research articles on protein synthesis

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Protein synthesis rates of muscle, tendon, ligament, cartilage, and bone tissue in vivo in humans

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

Affiliation Department of Human Biology, NUTRIM School of Nutrition and Translational Research in Metabolism, Maastricht University Medical Centre+, Maastricht, The Netherlands

Roles Conceptualization, Formal analysis, Investigation, Methodology, Resources, Supervision, Writing – review & editing

Roles Investigation, Resources, Writing – review & editing

Affiliation Department of Orthopedic Surgery, Maastricht University Medical Centre+, Maastricht, The Netherlands

Roles Formal analysis, Methodology, Software, Validation, Writing – review & editing

Roles Formal analysis, Methodology, Software, Writing – review & editing

Roles Conceptualization, Data curation, Investigation, Methodology, Supervision, Validation, Visualization, Writing – review & editing

* E-mail: [email protected]

• Joey S. J. Smeets, 
• Astrid M. H. Horstman, 
• Georges F. Vles, 
• Pieter J. Emans, 
• Joy P. B. Goessens, 
• Annemie P. Gijsen, 
• Janneau M. X. van Kranenburg, 
• Luc J. C. van Loon

• Published: November 7, 2019
• https://doi.org/10.1371/journal.pone.0224745
• Reader Comments

Skeletal muscle plasticity is reflected by a dynamic balance between protein synthesis and breakdown, with basal muscle tissue protein synthesis rates ranging between 0.02 and 0.09%/h. Though it is evident that other musculoskeletal tissues should also express some level of plasticity, data on protein synthesis rates of most of these tissues in vivo in humans is limited. Six otherwise healthy patients (62±3 y), scheduled to undergo unilateral total knee arthroplasty, were subjected to primed continuous intravenous infusions with L-[ring- 13 C 6 ]-Phenylalanine throughout the surgical procedure. Tissue samples obtained during surgery included muscle, tendon, cruciate ligaments, cartilage, bone, menisci, fat, and synovium. Tissue-specific fractional protein synthesis rates (%/h) were assessed by measuring the incorporation of L-[ring- 13 C 6 ]-Phenylalanine in tissue protein and were compared with muscle tissue protein synthesis rates using a paired t test. Tendon, bone, cartilage, Hoffa’s fat pad, anterior and posterior cruciate ligament, and menisci tissue protein synthesis rates averaged 0.06±0.01, 0.03±0.01, 0.04±0.01, 0.11±0.03, 0.07±0.02, 0.04±0.01, and 0.04±0.01%/h, respectively, and did not significantly differ from skeletal muscle protein synthesis rates (0.04±0.01%/h; P >0.05). Synovium derived protein (0.13±0.03%/h) and intercondylar notch bone tissue protein synthesis rates (0.03±0.01%/h) were respectively higher and lower compared to skeletal muscle protein synthesis rates ( P <0.05 and P <0.01, respectively). Basal protein synthesis rates in various musculoskeletal tissues are within the same range of skeletal muscle protein synthesis rates, with fractional muscle, tendon, bone, cartilage, ligament, menisci, fat, and synovium protein synthesis rates ranging between 0.02 and 0.13% per hour in vivo in humans.

Clinical trial registration : NTR5147

Citation: Smeets JSJ, Horstman AMH, Vles GF, Emans PJ, Goessens JPB, Gijsen AP, et al. (2019) Protein synthesis rates of muscle, tendon, ligament, cartilage, and bone tissue in vivo in humans. PLoS ONE 14(11): e0224745. https://doi.org/10.1371/journal.pone.0224745

Editor: Melissa M. Markofski, University of Houston, UNITED STATES

Received: April 30, 2019; Accepted: October 21, 2019; Published: November 7, 2019

Copyright: © 2019 Smeets et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are available from the Open Science Framework database (URL: https://osf.io/z7bgk/?view_only=9400d38f9c0749599a64cbf4e5682f91 ).

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Abbreviations: AA , amino acid; FSR , fractional synthesis rate; GC-IRMS , gas chromatography isotope ratio mass spectrometry; GC-MS , gas chromatography—mass spectrometry; LC-MS/MS , liquid chromatography—mass spectrometry; MPE , mole percent excess; PCA , perchloric acid; UPLC-MS , ultra-performance liquid chromatography mass spectrometry

Introduction

Skeletal muscle tissue plasticity is achieved by a dynamic equilibrium between muscle protein synthesis and breakdown rates. Temporary changes in either protein synthesis and/or protein breakdown result in net muscle protein accretion or loss. A routinely applied method to study skeletal muscle protein metabolism in vivo in humans is the continuous intravenous infusion of stable isotope labelled amino acids with frequent sampling of blood and skeletal muscle tissue using the percutaneous needle biopsy technique [ 1 , 2 ]. This contemporary stable isotope methodology has been applied for several decades to show that skeletal muscle tissue turns over at a rate of approximately 1–2% per day [ 3 ]. Though widely applied in skeletal muscle research, and to a lesser extent in tendon research [ 4 – 13 ], there are very few data on in vivo protein synthesis rates of other musculoskeletal tissues in humans.

It is evident that other musculoskeletal tissues such as tendon, ligaments, bone, and cartilage should also possess a certain degree of plasticity. Damage due to injury or surgery generally involves much more tissues than merely skeletal muscle. Obviously, recovery and rehabilitation requires plasticity of all tissues involved. Several studies have assessed tendon protein synthesis rates in vivo in humans using stable isotope methodology [ 4 – 12 , 14 , 15 ]. Furthermore, emerging research is now establishing the relevance of intramuscular as well as extramuscular collagen structures being required for the proper transduction of force generated by muscle contraction [ 16 , 17 ]. Clearly, connective tissue plasticity plays an important role in determining musculoskeletal strength and functional capacity [ 18 , 19 ].

The application of contemporary stable isotope methodology to assess tissue protein synthesis rates in other musculoskeletal tissues such as tendon, ligaments, bone, and cartilage is restricted due to the obvious logistical and medical ethical restraints of tissue sampling. To omit these restrictions we selected 6 otherwise healthy male ( n = 3) and female ( n = 3) adults, scheduled to undergo unilateral total knee arthroplasty, to participate in a study in which we applied contemporary stable isotope methodology to assess basal protein synthesis rates of a wide variety of musculoskeletal tissues including muscle, tendon, ligament, bone, cartilage, menisci, fat, and synovium. We hypothesized that basal protein synthesis rates of various musculoskeletal tissues are different compared to skeletal muscle tissue.

Materials and methods

Six otherwise healthy male ( n = 3) and female ( n = 3) adults (age: 62±3 y; body weight: 89.8±4.7 kg; body mass index: 28.6±1.1 kg/m 2 ), scheduled to undergo unilateral total knee arthroplasty, were recruited to participate in the present study. Subjects had no history of participating in any stable isotope infusion studies prior to this experiment. Exclusion criteria included secondary osteoarthritis of the knee, the use of intra-articular corticosteroid injections or bisphosphonates within 3 months prior to surgery, previous surgical intervention of the knee, rheumatoid arthritis or other systemic inflammatory diseases, and collagen disorders (e.g. Marfan and Ehlers-Danlos). All subjects were informed about the nature and possible risks of the experimental procedures, before their written informed consent was obtained. The study was approved by the Medical Ethical Committee of Zuyderland Medical Centre, Heerlen, The Netherlands, and conformed to the principles outlined in the declaration of Helsinki for use of human subjects and tissue.

Study design

The experimental protocol is outlined in Fig 1 . Each subject was diagnosed with knee osteoarthritis and, therefore, underwent unilateral total knee arthroplasty at the Department of Orthopedic Surgery at Maastricht University Medical Centre+, The Netherlands. Before and during surgery patients were subjected to primed continuous intravenous infusions with L-[ring- 13 C 6 ]-Phenylalanine. Four patients underwent general anesthesia and two patients underwent spinal anesthesia. Blood and tissue samples were collected throughout the surgical procedure to assess fractional muscle, tendon, ligament, bone, cartilage, menisci, fat, and synovium protein synthesis rates (FSR; %/h).

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

t = 0 min represents the start of the surgical procedure.

https://doi.org/10.1371/journal.pone.0224745.g001

Infusion protocol

All patients were fasted for at least 6 h prior to surgery. About 2.5 h before surgery a Teflon catheter was inserted into an antecubital vein for stable isotope infusion. A second Teflon catheter was inserted into a heated dorsal hand vein of the contralateral arm for perioperative blood sampling. After taking a baseline blood sample at t = -150 min, the serum phenylalanine pools were primed with a single dose of L-[ring- 13 C 6 ]-Phenylalanine (2 μmol/kg), after which continuous intravenous L-[ring- 13 C 6 ]-Phenylalanine (0.05 μmol/kg/min) infusion was initiated. Subsequently, blood samples were collected at t = -120, -90, -60, -30, 0 (start of the surgical procedure), 15, 30, 45, 60, 90, and 120 min ( Fig 1 ).

To determine basal tissue protein synthesis rates, tissue samples of the vastus lateralis muscle, patellar tendon, femur, tibia, cruciate ligaments, femoral cartilage, menisci, synovium, and Hoffa’s fat pad were obtained throughout the surgical procedure. All tissue samples were collected through surgical excision, except for the vastus lateralis muscle which was collected from the middle region of the vastus lateralis , approximately 15 cm above the patella and 3 cm below entry through the fascia, using the standard percutaneous needle biopsy technique [ 20 ]. Conventional muscle biopsy samples of the vastus lateralis were collected to assess skeletal muscle protein synthesis rates as a reference to the synthesis rates of the various musculoskeletal tissues obtained during surgery. All tissue samples were obtained directly after opening the joint and no tourniquet was used. For tissues that were visually affected by the disease process, such as cartilage and bone, we ensured that only parts of the tissues were sampled which appeared unaffected and healthy. Since vascularization and nutrient supply may differ substantially between and within tissues [ 21 – 26 ], all tissues were sampled in such a way that the obtained material was representative for the tissue. For bone tissues a mixture of cortical and trabecular bone was sampled (except for the predominantly trabecular notch bone tissue), and for the fibrous and intra-articular soft tissues complete cross sectional samples were obtained. Hereafter, samples were freed from any visible blood, immediately frozen in liquid nitrogen, and stored at -80°C until subsequent analysis. In addition, blood samples were collected at frequent intervals to determine L-[ring- 13 C 6 ]-Phenylalanine enrichment in serum protein. Blood samples were collected in serum tubes and centrifuged at 3500 g for 15 min at 20°C to obtain serum. Aliquots of serum were frozen in liquid nitrogen and stored at -80°C. For a schematic representation of the infusion protocol, please see Fig 1 .

Serum analyses

Serum amino acid concentrations and enrichments were determined by gas chromatography-mass spectrometry (GC-MS; Agilent 7890A GC/5975C; MSD, Little Falls, DE), as described in detail previously [ 27 ]. To measure concentrations, internal standards were added to the samples. The serum was deproteinized on ice with dry 5-sulfosalicylic acid. Free amino acids were purified using cation exchange AG 50W-X8 resin (mesh size: 100–200, ionic form: hydrogen (Bio-Rad Laboratories, Hercules, CA, USA)) columns. The free amino acids were converted to their tert-butyl dimethylsilyl (MTBSTFA) derivative before analysis by GC-MS. The amino acid concentrations were determined using electron impact ionization by monitoring ions at mass/charge (m/z) 336 and 346 for unlabelled phenylalanine and internal standards, respectively. The serum phenylalanine 13 C enrichments were determined using selective ion monitoring at m/z 336 and 342 for unlabelled and labelled phenylalanine, respectively. Standard regression curves were applied from a series of known standard enrichment values against the measured values to assess the linearity of the mass spectrometer and to account for any isotope fractionation that may have occurred during the analysis.

Tissue analyses

As described in detail previously [ 27 ], all tissues were freeze-dried, weighed and crushed. Subsequently, samples were homogenized in ice-cold 2% perchloric acid (PCA) using ultrasonic disintegration (Soniprep; MSE, London, UK). Samples were incubated on ice for 10 min. Following centrifugation, the supernatant was collected for determination of L-[ring- 13 C 6 ]-Phenylalanine enrichments in the tissue free amino acid pool using GC-MS analysis. Therefore, the supernatant was processed in the same manner as the serum samples. The tissue protein pellets were washed three times with 1.5 mL of ice-cold 2% PCA and hydrolysed in 3 mL of 6 M HCl overnight at 120°C. The free amino acids were then dissolved in 50% acetic acid solution and passed over cation exchange AG 50W-X8 resin (mesh size: 100–200, ionic form: hydrogen (Bio-Rad Laboratories, Hercules, CA, USA)) columns. The amino acids were eluted with 2 M NH 4 OH and dried under a continuous N 2 -stream for 48 h for measurement of L-[ring- 13 C 6 ]-Phenylalanine enrichment in tissue protein. To determine the L-[ring- 13 C 6 ]-Phenylalanine enrichment of tissue protein, the purified amino acids were derivatized into their N(O,S)-ethoxycarbonyl ethyl ester derivatives with ethyl chloroformate (ECF). The derivatives were then measured by GC-combustion-isotope ratio MS (GC-IRMS; MAT 253; Thermo-Scientific, Bremen, Germany) using an Agilent J&W DB-17MS (60 m) GC-column (Agilent Technologies, Santa Clara, CA, USA), and monitoring of ion masses 44, 45, and 46. Standard regression curves were applied to assess the linearity of the mass spectrometer and to control for the loss of tracer.

Amino acid concentrations

Quantification of amino acids in the different tissues was performed using ultra-performance liquid chromatograph mass spectrometry (UPLC-MS; ACQUITY UPLC H-Class with QDa; Waters, Saint-Quentin, France), as described in detail previously [ 27 ]. At least 5 mg of freeze-dried tissue was hydrolysed in 3 mL of 6 M HCl for 12 h at 120°C and dried under a continuous N 2 -stream. 5 mL of 0.1 M HCl was used to reconstitute the hydrolysates after which 50 μL of each protein hydrolysate was deproteinized using 100 μL of 10% SSA with 50 μM of MSK-A2 internal standard (Cambridge Isotope Laboratories, Massachusetts, USA). Subsequently, 50 μL of ultra-pure demineralized water was added and samples were centrifuged. After centrifugation, 10 μL of supernatant was added to 70 μL of Borate reaction buffer (Waters, Saint-Quentin, France). In addition, 20 μL of AccQ-Tag derivatizing reagent solution (Waters, Saint-Quentin, France) was added after which the solution was heated to 55°C for 10 min. Of this 100 μL derivative 1 μL was injected and measured using UPLC-MS.

Protein identification

As described in detail previously [ 27 ], tissue samples were homogenized in 50 mM ammonium bicarbonate and 5 M urea buffer, freeze-dried in three cycles, vortexed for 1 min and centrifuged at 20000 g for 30 min at 10°C. The supernatant was collected and stored at -80°C until further analysis. Protein concentrations were determined with the Protein Assay Kit (Bio-Rad, Veenendaal, the Netherlands). Subsequently, a total of 75 μg protein in 50 μL 50 mM ammonium bicarbonate with 5 M urea was used for further analysis. 5 μL of DTT solution (20 mM final) was added and incubated at room temperature for 45 min. Proteins were alkylated by adding 6 μL of IAA solution (40 mM final) and incubated at room temperature for 45 min in darkness. Alkylation was stopped by adding 10 μL DTT solution (to consume any unreacted IAA) and incubation at room temperature for 45 min. Subsequently, 3 μg trypsin/lysC was added to the protein and incubated at 37°C for 2 h. 200 μL of 50 mM ammonium bicarbonate was added to dilute the urea concentration and the solution was further incubated at 37°C for 18 h. The digestion mixture was centrifuged at 2500 g for 5 min and the supernatant was collected. The digestion mixture was fourfold diluted for the use of LC-MS/MS analysis. LC-MS/MS was performed using a nanoflow HPLC instrument (Dionex ultimate 3000) coupled on-line to a Q Exactive (Thermo Scientific) with a nano-electrospray Flex ion source (Proxeon). The digest/peptide mixture was loaded onto a C18-reversed phase column (Thermo Scientific Acclaim PepMap C18 column, 75-μm inner diameter x 15 cm, 2-μm particle size). Peptides were separated with a 90 min linear gradient of 4–45% buffer (80% acetonitrile and 0.08% formic acid) at a flow rate of 300 nL/min. Proteins were identified using Proteome Discoverer v2.1 Sequest HT search engine (Thermo Scientific). The false discovery rate (FDR) was set to 0.01 for proteins and peptides.

Calculations

All data are expressed as means±SEM. Paired t tests were used to compare intracellular free and tissue protein bound L-[ring- 13 C 6 ]-Phenylalanine enrichments between vastus lateralis muscle (as the reference tissue) and each of the different musculoskeletal tissues. Likewise, fractional synthesis rate of each of the different musculoskeletal tissues was compared with vastus lateralis muscle tissue fractional synthesis rates using paired t tests. For these comparisons, the 95% confidence interval (95% CI) of the difference, and the effect size (Cohen’s d ) were calculated. No statistical analyses were performed between the different musculoskeletal tissues. Missing data was accounted for using pairwise deletion. Due to the exploratory nature of the experiment, multiplicity adjustments were not performed. For all analyses, significance was set at P <0.05. All calculations were performed using SPSS (version 23.0, IBM Corp., Armonk, NY, USA).

Serum enrichments

As shown in Fig 2 , serum L-[ring- 13 C 6 ]-Phenylalanine enrichments did not change significantly throughout the infusion period, despite the surgical setting of the experiment. Throughout the surgical procedure, serum L-[ring- 13 C 6 ]-Phenylalanine enrichments averaged 6.53±0.20 MPE.

Serum L-[ring- 13 C 6 ]-Phenylalanine enrichments are expressed as mole percent excess (MPE). t = 0 min represents the start of the surgical procedure. Values represent means+SEM. Serum L-[ring- 13 C 6 ]-Phenylalanine enrichments did not change significantly throughout the experiments.

https://doi.org/10.1371/journal.pone.0224745.g002

Tissue free and protein bound enrichments

Tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments in skeletal muscle tissue averaged 4.74±0.39 MPE. The majority of musculoskeletal tissues did not significantly differ in their tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments when compared to skeletal muscle tissue, except for femoral bone, tibial bone, and Hoffa’s fat pad which all possessed lower tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments (4.14±0.39, 4.19±0.35, and 4.32±0.41 MPE, respectively; P < 0.05; Table 1 ).

https://doi.org/10.1371/journal.pone.0224745.t001

Protein bound L-[ring- 13 C 6 ]-Phenylalanine enrichments in skeletal muscle tissue averaged 0.010±0.002 MPE. The highest enrichment levels were found in synovium and Hoffa’s fat pad (0.029±0.008 and 0.023±0.007 MPE, respectively), whereas the lowest L-[ring- 13 C 6 ]-Phenylalanine enrichments were observed in patellar bone (0.005±0.002 MPE). The various musculoskeletal tissues did not significantly differ in their observed L-[ring- 13 C 6 ]-Phenylalanine enrichments when compared to skeletal muscle, except for intercondylar notch bone tissue (i.e. the deep groove or notch between the two femoral condyles) that did have a lower L-[ring- 13 C 6 ]-Phenylalanine enrichment level (0.010±0.002 vs 0.008±0.001, respectively; P <0.01; Table 1 ).

Tissue protein synthesis rates

Tissue-specific protein synthesis rates, using serum L-[ring- 13 C 6 ]-Phenylalanine enrichments as the precursor pool, are shown in Figs 3 and 4 . In line with previous data, basal protein synthesis rates averaged 0.04±0.01%/h in skeletal muscle tissue. Synovium protein synthesis rates were significantly higher when compared to skeletal muscle protein synthesis rates (0.13±0.03%/h; P <0.05), whereas intercondylar notch bone tissue protein synthesis rates were significantly lower when compared to muscle tissue protein synthesis rates (0.03±0.01%/h; P <0.01). Fractional synthesis rates of other musculoskeletal tissues varied between 0.02 and 0.11%/h and did not significantly differ from skeletal muscle tissue. Please see Table 2 for the mean±SD of the difference, 95% CI of the difference, effect size (Cohen’s d ), and P -value for the comparison of each musculoskeletal tissue with vastus lateralis muscle tissue protein synthesis rates. Similar tissue-specific protein synthesis rates were observed when using tissue free precursor enrichments, though significant differences were no longer present ( S1 Fig ).

Tissue protein synthesis rates (FSR) based on incorporation of L-[ring- 13 C 6 ]-Phenylalanine in human musculoskeletal tissue protein with serum L-[ring- 13 C 6 ]-Phenylalanine enrichments used as precursor pool. Values represent means±SEM.

https://doi.org/10.1371/journal.pone.0224745.g003

Fractional tissue protein synthesis rates (FSR) based on incorporation of L-[ring- 13 C 6 ]-Phenylalanine in human musculoskeletal tissue protein with serum L-[ring- 13 C 6 ]-Phenylalanine enrichments used as precursor pool. Values represent means+SEM. The number of pairs included in each comparison for both protein bound and tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments is n = 6, except for tibial bone, trochlea, notch, and patellar bone tissue (all n = 5). The right-hand section x-axis represents averaged individual fractional synthesis rates per tissue class. * Significantly different from vastus lateralis muscle, P <0.05.

https://doi.org/10.1371/journal.pone.0224745.g004

https://doi.org/10.1371/journal.pone.0224745.t002

Tissue protein content and amino acid composition

Tissue protein contents ranged between 16 and 98% of the raw (dry) material ( S1 Table ). Protein contents of skeletal muscle averaged 74% and were higher when compared to the various bone tissue samples, which ranged between 16 and 31% of dry tissue weight. Compared to skeletal muscle tissue, protein contents of tendon, ligaments, cartilage, and menisci were higher and ranged between 79 and 98% of dry tissue weight. Protein content of synovium and Hoffa’s fat pad were lower compared to skeletal muscle tissue protein content (23±7 and 29±19% of dry tissue weight, respectively).

Essential amino acid contents of all musculoskeletal tissues ranged between 15 and 25% of total amino acid content and were considerably lower when compared to skeletal muscle tissue (43% of total amino acids). Non-essential amino acid contents of the different musculoskeletal tissues ranged between 75 and 85% of total amino acids, were substantially higher compared to skeletal muscle (57% of total amino acids), and were mainly attributed to the high alanine, glycine, and proline contents ( S1 Table ). Amino acid profiles of tendon, cartilage, and bone differed substantially from skeletal muscle, with glycine contents as high as 42% of total amino acid content in patellar tendon, 39% in cartilage, and 40% in femoral bone tissue and as low as 9% in skeletal muscle tissue. In addition, proline contents appeared to be higher in all musculoskeletal tissues as well, with 14% of total amino acid content in patellar tendon and cartilage, 13.5% in femoral bone tissue, and 6% in skeletal muscle tissue. An overview of amino acid profile and amino acid composition, i.e. essential vs non-essential amino acid ratios, is provided in Fig 5 .

Amino acid content is presented in % of total AA content. Note: Tryptophan, Asparagine, and Glutamine were not measured. ƩEAA, sum of all essential amino acids; ƩNEAA, sum of all non-essential amino acids.

https://doi.org/10.1371/journal.pone.0224745.g005

Supplemental S2 Table provides a list of all identified proteins and their corresponding estimated abundances. In total 1374 different proteins have been identified in the different musculoskeletal tissues of one subject ( n = 1).

The current study provides insight into tissue protein metabolism of a wide variety of musculoskeletal tissues in vivo in humans. Using stable isotope methodology, we showed that average basal protein synthesis rates of various musculoskeletal tissues are within the same range of skeletal muscle protein synthesis rates, with fractional muscle, tendon, bone, cartilage, ligament, menisci, fat, and synovium protein synthesis rates ranging between 0.02 and 0.13% per hour in vivo in humans.

Skeletal muscle protein synthesis rates observed in the present study averaged 0.04±0.01%/h ( Fig 3 ). These rates are similar to muscle protein synthesis rates assessed previously in an overnight fasted state in a wide variety of subjects studied in our lab [ 1 , 29 – 34 ] as well as in many other laboratories [ 3 , 35 – 37 ]. With protein synthesis rates ranging between 1–2% per 24 h, skeletal muscle tissue shows extensive remodeling within a matter of weeks to months [ 3 ]. These protein synthesis rates allow skeletal muscle tissue to adapt to changes in habitual use, with muscle hypertrophy following increased levels of physical activity [ 38 – 40 ] or muscle atrophy developing during periods of reduced physical activity or disuse [ 41 – 43 ]. Though a certain level of plasticity has been well established for skeletal muscle tissue, there is less data on in vivo tissue protein synthesis rates of most other musculoskeletal tissues in humans.

Tendon and ligaments play an important role in the force-transmitting function of the musculoskeletal system [ 18 , 44 ]. Tendon protein synthesis rates in this study averaged 0.06±0.01%/h ( Fig 3 ), which is in line with previous literature describing patellar tendon protein synthesis rates ranging between 0.01 and 0.07%/h [ 4 – 13 , 15 ]. These data suggest that tendon tissue possesses similar protein synthesis characteristics as skeletal muscle tissue and may, therefore, also express some level of plasticity to external stimuli. In agreement, exercise has been reported to increase tendon protein synthesis rates [ 9 , 13 ]. In addition to tendon tissue, total knee arthroplasty provided us with the unique opportunity to sample both anterior and posterior cruciate ligaments. With fractional synthetic rates averaging 0.07±0.02 and 0.04±0.01%/h for the anterior and posterior cruciate ligaments, respectively, we observed that both cruciate ligaments turn over at similar rates compared to skeletal muscle tissue ( Fig 3 ). Interestingly, protein synthesis rates of patellar tendon tissue appeared to be similar to anterior cruciate ligament protein synthesis rates. Though not statistically tested, from a clinical perspective this may be relevant since patellar tendon tissue is often used in the surgical reconstruction of anterior cruciate ligament injuries [ 45 ]. In addition, protein synthesis rates of the anterior cruciate ligament tended to be higher compared to the posterior cruciate ligament ( Fig 4 ). Whether this difference in protein synthesis rates between cruciate ligaments is reflective of differences in incidence of injuries [ 46 ] or the tissue’s capacity to repair, remains to be established.

Bone tissue quality is determined by multiple mechanical properties such as elasticity, resistance to bending, and toughness on impact [ 47 , 48 ]. Though bone tissue has always been considered to possess limited remodeling capacity, few studies have actually investigated this. Previous studies have provided semi-quantitative estimates on adult bone remodeling of 3–25% per year [ 49 ], and on bone calcium turnover of ~8–15% per year [ 50 , 51 ]. Here, we assessed bone tissue protein synthesis rates directly by measuring the incorporation of infused L-[ring- 13 C 6 ]-Phenylalanine in the bone tissue protein pool of a variety of bone tissue samples. Bone tissue protein synthesis rates ranged between 0.02 and 0.03%/h ( Fig 3 ). These data indicate that bone tissue may possess a much greater remodeling capacity than previously assumed. Though in this study most of the bone tissue samples showed protein synthesis rates similar to skeletal muscle tissue, much higher bone collagen synthesis rates (0.06±0.01%/h) have been observed previously by Babraj et al . using stable isotope methodology [ 52 ]. This discrepancy in synthesis rates may be caused by the type of protein studied. Babraj et al . [ 52 ] have measured bone collagen synthesis rates, whereas we have assessed mixed bone tissue protein synthesis rates. Obviously, protein synthetic rates may differ substantially between different proteins and protein fractions.

To our knowledge, no previous study has applied stable isotope methodology to assess fractional protein synthesis rates of cartilage and menisci in vivo in humans. These structures have always been suggested to turn over slowly. Especially human cartilage has been considered an essentially permanent structure with little to no ability to remodel after a certain age [ 53 ]. Our data, however, show that cartilage tissue protein synthesis rates average 0.04±0.01%/h. In addition, we observed that both medial and lateral menisci have a protein synthesis rate of 0.04±0.01 and 0.04±0.01%/h, respectively as well ( Fig 3 ). Since cartilage tissue is avascular [ 25 ] and menisci tissue’s vascularization differs substantially within the tissue [ 21 ], tissue-specific protein synthesis rates were also calculated based on tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments. However, a similar pattern of tissue-specific protein synthesis rates was observed when using tissue free precursor enrichments.

Our assessment of musculoskeletal tissue protein synthesis rates represents the integrated synthesis rates of all available proteins in the musculoskeletal tissues that were sampled. However, different proteins or protein fractions within a tissue likely possess different turnover rates. Indeed, others have shown that protein turnover of the collagen matrix of human articular cartilage is negligible whereas turnover rates of the cartilage glycosaminoglycan matrix are substantially higher [ 53 ]. Hence, it seems that differences between proteins and protein fractions within different musculoskeletal tissues may be important in interpreting these data. To obtain more insight into the contribution of individual proteins and/or protein fractions to the mixed tissue protein synthesis rates we report here, we additionally applied liquid chromatography/mass spectrometry (LC-MS/MS) in all tissue samples of a single subject to identify the proteins present and their (semi-quantitative) abundance in the tissues. Supplementary S2 Table provides a list of 1374 identified proteins and their estimated abundances in each of the musculoskeletal tissues. Though such data provide more in-depth insight, they do not show which proteins or protein fractions are synthesized more or less rapidly. The combination of contemporary stable isotope methodology and proteomics analyses methods do not yet allow us to detect fractional protein synthesis rates at the level of individual proteins in tissues in vivo in humans. Limitations are present in the ability to identify all proteins present, their (relative) abundances in the tissues, as well as their degree of label enrichments.

To obtain some insight into the differences in amino acid composition between the different tissues, we applied ultra-performance liquid chromatography mass spectrometry (UPLC-MS) to assess tissue-specific amino acid composition of all musculoskeletal tissues ( S1 Table ). Fig 5 presents a selection of 4 different musculoskeletal tissues and their amino acid composition. Essential amino acid contents of patellar tendon, cartilage, and femoral bone are much lower when compared to skeletal muscle tissue, whereas the amino acids glycine and proline are substantially more abundant in patellar tendon, cartilage, and femoral bone when compared to skeletal muscle tissue. Proline and glycine are both amino acids known to be important in musculoskeletal collagen metabolism [ 54 – 56 ]. Future research should evaluate the impact of such differences in amino acid content on basal and, more importantly, post-prandial tissue protein synthesis rates under various conditions.

Apart from musculoskeletal tissues that allow for adequate movement and power transfer, the human knee joint also contains metabolically active intra-articular soft tissues such as Hoffa’s fat pad and synovium. Hoffa’s fat pad is known to share morphological similarities with subcutaneous fat [ 57 ], possess an abundant peripheral anastomotic blood supply [ 58 ], and has been suggested to play a modulatory role in the inflammatory pathways in osteoarthritis [ 59 ]. Synovium tissue lines the inner surface of the knee joint and primarily produces synovial fluid to lubricate the joint. However, it is also known to become inflamed in patients with different stages of knee osteoarthritis [ 60 ]. In the present study Hoffa’s fat pad and synovium tissue protein synthesis rates averaged 0.11±0.03 and 0.13±0.03%/h, respectively ( Fig 3 ), which are substantially higher when compared to skeletal muscle tissue protein synthesis rates ( Fig 4 ). These higher tissue protein synthesis rates of Hoffa’s fat pad and synovium tissue might be reflective of the metabolic active nature of these tissues or their suggested role in disease dependent pathways. For all tissues we ensured that only visually unaffected parts of the tissue were sampled, thereby ensuring that we were sampling only healthy tissue. However, the potential influence of disease dependent processes on our measurement of tissue protein synthesis rates in Hoffa’s fat pad and synovium should not be ignored. Nevertheless, these observations for the first time show tissue protein synthesis rates of intra-articular soft tissues when compared to other musculoskeletal tissues in vivo in humans.

From a research perspective, the present study provides interesting data on protein metabolism of a wide variety of musculoskeletal tissues of the human knee. Whereas protein synthesis rates have been assessed in some of the referred tissues, no study has directly assessed in vivo protein synthesis rates of all of these musculoskeletal tissues in a single study in humans. Knowledge of basal musculoskeletal tissue protein synthesis rates enables us to further explore the capacity of these tissues to regenerate. Skeletal muscle plasticity is well appreciated since muscle tissue has shown to be highly responsive to both anabolic and catabolic stimuli. To assess whether other musculoskeletal tissues are capable of displaying some degree of plasticity, more work is required to address the impact of various factors on tissue protein synthesis rates. It has previously been observed that gelatin supplementation can stimulate collagen synthesis following exercise [ 55 ], and collagen hydrolysate supplementation has been reported to increase collagen content in the knee of osteoarthritis patients [ 61 ], and may decrease knee pain in athletes with activity-related joint pain [ 62 ]. From a clinical perspective this is more than interesting, because identifying which specific proteins or protein fractions are responsive to external stimuli may enable us to develop more effective therapies in treating and/or preventing injuries.

In conclusion, basal fractional muscle, tendon, bone, cartilage, ligament and menisci protein synthesis rates range between 0.02 and 0.13% per hour in vivo in humans. Fractional tissue protein synthesis rates of tendon, bone, cartilage, ligament and menisci do not differ substantially from muscle tissue protein synthesis rates, suggesting that these musculoskeletal tissues may express a greater level of tissue plasticity than generally believed.

Supporting information

S1 fig. musculoskeletal tissue protein synthesis rates..

Fractional tissue protein synthesis rates (FSR) based on incorporation of L-[ring- 13 C 6 ]-Phenylalanine in human musculoskeletal tissue protein with tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments used as precursor pool. Values represent means+SEM. The number of pairs included in each comparison for both protein bound and tissue free L-[ring- 13 C 6 ]-Phenylalanine enrichments is n = 6, except for tibial bone, trochlea, notch, and patellar bone tissue (all n = 5). * Significantly different from vastus lateralis muscle, P <0.05.

https://doi.org/10.1371/journal.pone.0224745.s001

S1 Table. Protein and amino acid content of various musculoskeletal tissues.

Protein content is presented in % of raw material based on the determined nitrogen content multiplied by 6.25 as the standard conversion factor. Amino acid content is presented in % of total AA content. Note: Tryptophan, Asparagine, and Glutamine were not measured. ƩEAA, sum of all essential amino acids; ƩNEAA, sum of all non-essential amino acids. URL: https://osf.io/z7bgk/?view_only=9400d38f9c0749599a64cbf4e5682f91 .

https://doi.org/10.1371/journal.pone.0224745.s002

S2 Table. Protein identification.

Protein identification and semi-quantitative analyses of relative abundances were performed by LC-MS/MS. URL: https://osf.io/z7bgk/?view_only=9400d38f9c0749599a64cbf4e5682f91 .

https://doi.org/10.1371/journal.pone.0224745.s003

Acknowledgments

The authors greatly acknowledge the enthusiasm and dedication of the participants in this study.

• View Article
• PubMed/NCBI
• Google Scholar
• 3. Waterlow JC. Protein turnover. Oxfordshire: CABI; 2006.

Suggestions or feedback?

MIT News | Massachusetts Institute of Technology

• Machine learning
• Sustainability
• Black holes
• Classes and programs

Departments

• Aeronautics and Astronautics
• Brain and Cognitive Sciences
• Architecture
• Political Science
• Mechanical Engineering

Centers, Labs, & Programs

• Abdul Latif Jameel Poverty Action Lab (J-PAL)
• Picower Institute for Learning and Memory
• Lincoln Laboratory
• School of Architecture + Planning
• School of Engineering
• School of Humanities, Arts, and Social Sciences
• Sloan School of Management
• School of Science
• MIT Schwarzman College of Computing

New technology enables fast protein synthesis

Press contact :, media download.

*Terms of Use:

Images for download on the MIT News office website are made available to non-commercial entities, press and the general public under a Creative Commons Attribution Non-Commercial No Derivatives license . You may not alter the images provided, other than to crop them to size. A credit line must be used when reproducing images; if one is not provided below, credit the images to "MIT."

Previous image Next image

Many proteins are useful as drugs for disorders such as diabetes, cancer, and arthritis. Synthesizing artificial versions of these proteins is a time-consuming process that requires genetically engineering microbes or other cells to produce the desired protein.

MIT chemists have devised a protocol to dramatically reduce the amount of time required to generate synthetic proteins. Their tabletop automated flow synthesis machine can string together hundreds of amino acids, the building blocks of proteins, within hours. The researchers believe their new technology could speed up the manufacturing of on-demand therapies and the development of new drugs, and allow scientists to design artificial proteins by incorporating amino acids that don’t exist in cells.

“You could design new variants that have superior biological function, enabled by using non-natural amino acids or specialized modifications that aren’t possible when you use nature’s apparatus to make proteins,” says Brad Pentelute, an associate professor of chemistry at MIT and the senior author of the study.

In a paper appearing today in Science , the researchers showed that they could chemically produce several protein chains up to 164 amino acids in length, including enzymes and growth factors. For a handful of these synthetic proteins, they performed a detailed analysis showing their function is comparable to that of their naturally occurring counterparts.

The lead authors of the paper are former MIT postdoc Nina Hartrampf, who is now an assistant professor at the University of Zurich, MIT graduate student Azin Saebi, and former MIT technical associate Mackenzie Poskus.

Rapid production

The majority of proteins found in the human body are up to 400 amino acids long. Synthesizing large quantities of these proteins requires delivering genes for the desired proteins into cells that act as living factories. This process is used to program bacterial or yeast cells to produce insulin and other drugs such as growth hormones.

“This is a time-consuming process,” says Thomas Nielsen, head of research chemistry at Novo Nordisk, who is also an author of the study. “First you need the gene available, and you need to know something about the cellular biology of the organism so you can engineer the expression of your protein.”

An alternative approach for protein production, first proposed in the 1960s by Bruce Merrifield, who was later awarded the Nobel Prize in chemistry for his work on solid-phase peptide synthesis, is to chemically string amino acids together in a stepwise fashion. There are 20 amino acids that living cells use to build proteins, and using the techniques pioneered by Merrifield, it takes about an hour to perform the chemical reactions needed to add one amino acid to a peptide chain.

In recent years, Pentelute’s lab has invented a more rapid method to perform these reactions, based on a technology known as flow chemistry. In their machine, chemicals are mixed using mechanical pumps and valves, and at every step of the overall synthesis they cycle through a heated reactor containing a resin bed. In the optimized protocol, forming each peptide bond takes on average 2.5 minutes, and peptides up to 25 amino acids long can be assembled in less than an hour.

Following the development of this technology, Novo Nordisk, which makes several protein drugs, became interested in working with Pentelute’s lab to synthesize longer peptides and proteins. To achieve that, the researchers needed to improve the efficiency of the reactions that form peptide bonds between amino acids in the chain. For each reaction, their previous efficiency rate was between 95 and 98 percent, but for longer proteins, they needed it to be over 99 percent.

“The rationale was if we got really good at making peptides, we could expand the technology to make proteins,” Pentelute says. “The idea is to have a machine that a user could walk up to and put in a protein sequence, and it would string together these amino acids in such an efficient manner that at the end of the day, you can get the protein you want. It’s been very challenging because if the chemistry is not close to 100 percent for every single step, you will not get any of the desired material.”

To boost their success rate and find the optimal recipe for each reaction, the researchers performed amino-acid-specific coupling reactions under many different conditions. In this study, they assembled a universal protocol that achieved an average efficiency greater than 99 percent for each reaction, which makes a significant difference when so many amino acids are being linked to form large proteins, the researchers say.

“If you want to make proteins, this extra 1 percent really makes all the difference, because byproducts accumulate and you need a high success rate for every single amino acid incorporated,” Hartrampf says.

Using this approach, the researchers were able to synthesize a protein that contains 164 amino acids — Sortase A, a bacterial protein. They also produced proinsulin, an insulin precursor with 86 amino acids, and an enzyme called lysozyme, which has 129 amino acids, as well as a few other proteins. The desired protein has to be purified and then folded into the correct shape, which adds a few more hours to the overall synthesis process. All of the purified synthesized proteins were obtained in milligram quantities, making up between 1 and 5 percent of the overall yield.

Medicinal chemistry

The researchers also tested the biological functions of five of their synthetic proteins and found that they were comparable to those of the biologically expressed variants.

The ability to rapidly generate any desired protein sequence should enable faster drug development and testing, the researchers say. The new technology also allows amino acids other than the 20 encoded by the DNA of living cells to be incorporated into proteins, greatly expanding the structural and functional diversity of potential protein drugs that could be created.

“This is paving the way for a new field of protein medicinal chemistry,” Nielsen says. “This technology really complements what is available to the pharmaceutical industry, providing new opportunities for rapid discovery of peptide- and protein-based biopharmaceuticals.”

The researchers are now working on further improving the technology so that it can assemble protein chains up to 300 amino acids long. They are also working on automating the entire manufacturing process, so that once the protein is synthesized, the cleavage, purification, and folding steps also occur without any human intervention required.

Pentelute is a co-founder of a company called Amide Technologies that has licensed aspects of the peptide synthesis technology for possible commercial development. The research was funded by Novo Nordisk, a National Science Foundation Graduate Research Fellowship, and an MIT Dean of Science Fellowship.

Share this news article on:

Related links.

• Pentelute Lab
• Department of Chemistry

Related Topics

• Drug development
• National Science Foundation (NSF)

Related Articles

An experimental peptide could block Covid-19

Chemists synthesize millions of proteins not found in nature

New technology offers fast peptide synthesis

Previous item Next item

More MIT News

Helping robots zero in on the objects that matter

Read full story →

Where flood policy helps most — and where it could do more

MIT launches new Music Technology and Computation Graduate Program

New security protocol shields data from attackers during cloud-based computation

How social structure influences the way people share money

Mars’ missing atmosphere could be hiding in plain sight

• More news on MIT News homepage →

Massachusetts Institute of Technology 77 Massachusetts Avenue, Cambridge, MA, USA

• Map (opens in new window)
• Events (opens in new window)
• People (opens in new window)
• Careers (opens in new window)
• Accessibility
• Social Media Hub
• MIT on Facebook
• MIT on YouTube
• MIT on Instagram

Our systems are now restored following recent technical disruption, and we’re working hard to catch up on publishing. We apologise for the inconvenience caused. Find out more: https://www.cambridge.org/universitypress/about-us/news-and-blogs/cambridge-university-press-publishing-update-following-technical-disruption

We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings .

Login Alert

• > Journals
• > Proceedings of the Nutrition Society
• > Volume 80 Issue 2
• > Protein metabolism and related body function: mechanistic...

Article contents

Molecular mechanisms for the sensing of protein and amino acid sufficiency, a tight control of protein intake, protein intake, protein synthesis and body composition, protein metabolism and related body function: mechanistic approaches and health consequences.

Published online by Cambridge University Press:  14 October 2020

The development and maintenance of body composition and functions require an adequate protein intake with a continuous supply of amino acids (AA) to tissues. Body pool and AA cellular concentrations are tightly controlled and maintained through AA supply (dietary intake, recycled from proteolysis and de novo synthesis), AA disposal (protein synthesis and other AA-derived molecules) and AA losses (deamination and oxidation). Different molecular regulatory pathways are involved in the control of AA sufficiency including the mechanistic target of rapamycin complex 1, the general control non-derepressible 2/activating transcription factor 4 system or the fibroblast growth factor 21. There is a tight control of protein intake, and human subjects and animals appear capable of detecting and adapting food and protein intake and metabolism in face of foods or diets with different protein contents. A severely protein deficient diet induces lean body mass losses and ingestion of sufficient dietary energy and protein is a prerequisite for body protein synthesis and maintenance of muscle, bone and other lean tissues and functions. Maintaining adequate protein intake with age may help preserve muscle mass and strength but there is an ongoing debate as to the optimal protein intake in older adults. The protein synthesis response to protein intake can also be enhanced by prior completion of resistance exercise but this effect could be somewhat reduced in older compared to young individuals and gain in muscle mass and function due to exercise require regular training over an extended period.

A daily intake of an adequate quantity of protein from foods provides nitrogen and amino acid (AA) to support the synthesis of body proteins and as precursors of various nitrogenous and other important compounds in the body. A continuous supply of AA to tissues, and particularly essential AA (EAA) which are not de novo synthesised in the body, is required for survival, for the development and maintenance of body composition and to support AA-dependent metabolic processes and most if not all physiological functions. AA sufficiency in the body and in tissues and cells is tightly and continuously controlled through different sensing and signalling processes that modulate and adapt protein and energy metabolism and feeding behaviour to prevent or counteract protein deficiency and to reach and maintain a well-balanced protein status.

AA play a central role in the metabolism and their body pool and cellular concentrations are tightly controlled and maintained through AA supply (dietary intake, recycled from proteolysis and de novo synthesis), AA disposal (protein synthesis and other AA-derived molecules) and AA losses (deamination and oxidation). At the body and cellular levels, the control and maintenance of AA sufficiency proceeds through complex sensing and signalling pathways ( Reference Broer and Broer 1 ) . Different mediators, hormones, signalling regulatory molecules and their upstream and downstream pathways are involved in the control and maintenance of AA sufficiency including insulin, the insulin growth factor 1 (IGF1), the fibroblast growth factor 21 (FGF21), the mechanistic target of rapamycin (mTOR) complex 1 (mTORC1), the AMP-activated protein kinase and the general control non-derepressible 2 (GCN2)/activating transcription factor 4 (ATF4) system. Numerous AA transport processes and transporters adapt to intracellular AA level and modulate cellular AA exchange with the extracellular medium through AA uptake and excretion ( Reference Broer and Broer 1 – Reference Palacin, Estevez and Bertran 5 ) .

The cellular availability of AA is involved in the anabolic response to feeding through mRNA translation and protein synthesis ( Reference Broer and Broer 1 , Reference Anthony, Anthony and Kimball 6 – Reference Lynch, Halle and Fujii 10 ) . The mTOR pathway, and more precisely the mTORC1 complex constituted by the protein mTOR and the regulatory associated protein of mTOR is a central regulatory component in the sensing and signalling of cellular AA sufficiency ( Reference Li, Yin and Tan 9 , Reference Avruch, Long and Ortiz-Vega 11 – Reference Zheng, Zhang and Zhou 16 ) . It is a main regulator of cell growth with the capacity to phosphorylate target proteins involved in cellular anabolic pathways including protein synthesis, and in catabolic pathways including autophagy ( Reference Dunlop and Tee 17 – Reference Tan and Miyamoto 19 ) . Under low intracellular AA concentration, mTORC1 is inactivated, leading to reduced protein synthesis and increased proteolysis through protein autophagy ( Reference Tan and Miyamoto 19 ) . Under high intracellular AA concentration, mTORC1 is activated, promoting protein synthesis and inhibiting protein degradation and autophagy ( Reference Kimball 7 ) . The active mTORC1 complex initiates mRNA translation and protein synthesis by phosphorylation of downstream target effectors including the 70-kDa ribosomal protein S6 kinase and the eukaryotic initiation factor 4E binding proteins 1 and 2 ( Reference Dennis, Jefferson and Kimball 20 ) . The activation of mTORC1 is associated with its recruitment to the surface of the lysosome with a direct role in the control of autophagy and lysosomal biogenesis ( Reference Rabanal-Ruiz and Korolchuk 21 ) .

Signals that modulate mTORC1 activity and mTORC1-dependant metabolic pathways involved in the anabolic response to feeding protein, associate hormones, growth factors, increased AA concentration as precursors of protein synthesis and some specific AA also identified as signal molecules, including particularly leucine, other branched-chain AA, arginine, glutamine and lysine. Sestrin2 and CASTOR1 proteins are proposed as molecular sensors for leucine and arginine, inducing through the same cascade of cellular events mTORC1 activation ( Reference Chantranupong, Scaria and Saxton 22 – Reference Wolfson, Chantranupong and Saxton 25 ) . The ingestion of protein, free branched-chain AA or free leucine, is associated with higher cellular uptake of leucine through specific transport systems, its transfer to the lysosome, the colocalisation of mTORC1 with the lysosome and the activation of mTORC1 ( Reference Sancak, Peterson and Shaul 26 ) . Glutamine could increase the cellular uptake of leucine through solute carriers expressed at the plasma and lysosomal membrane and also participate to induce lysosomal mTORC1 colocalisation and activation ( Reference Duran, Oppliger and Robitaille 27 – Reference Nicklin, Bergman and Zhang 29 ) . The ingestion of a meal containing 20–30 g leucine-rich proteins induces the activation of mTORC1 and the stimulation of skeletal muscle synthesis within 2 h, and this effect is reinforced by regular physical training ( Reference Borack and Volpi 30 ) .

The GCN2/ATF4 system controls AA insufficiency and imbalance in mammalian cells and subsequently increases the cellular AA pool by reducing translation and AA oxidation and enhancing AA uptake and biosynthesis ( Reference Balasubramanian, Butterworth and Kilberg 31 – Reference Ye, Kumanova and Hart 33 ) . In these processes, AA and insulin exert a coordinated action on translation involving mTOR, AMP-activated protein kinase and GCN2 transduction pathways and inhibition of AMP-activated protein kinase and activation of mTOR transduction pathways are required for the downregulation of the protein ubiquitination proteolytic pathway in response to high AA and insulin concentrations ( Reference Chotechuang, Azzout-Marniche and Bos 34 , Reference Chotechuang, Azzout-Marniche and Bos 35 ) . A situation of AA deficiency induces an increase in uncharged transfer ribonucleic acid that binds to GCN2 with subsequent phosphorylation of eukaryotic initiation factor-2 and the glutamate receptor 1. Interestingly, FGF21 is under the control of GCN2 that senses AA deficiency through the ATF4 pathway ( Reference Hill, Berthoud and Munzberg 36 , Reference Hill, Laeger and Dehner 37 ) . FGF21 appears as a signal of protein insufficiency and has been involved in the downstream control of metabolic processes in different organs (liver, brown adipose tissue and skeletal muscle), such as lipid oxidation, ketogenesis and glucose uptake, and in the stimulation of adaptive diet-induced thermogenesis in response to a low protein diet or to EAA restriction as shown for leucine, methionine or threonine ( Reference Hill, Berthoud and Munzberg 36 , Reference Chalvon-Demersay, Even and Tome 38 – Reference Yap, Rusu and Chan 46 ) . Both energy expenditure and food intake are increased after intracerebroventricular infusion of FGF21 without affecting body composition ( Reference Laeger, Henagan and Albarado 47 , Reference Sarruf, Thaler and Morton 48 ) .

Subjects are able to detect and adapt food and protein intake and metabolism to maintain or restore an adequate protein status in face of different foods or diets with different protein contents classified as high (above 25–30 % energy as protein), normal (10–20 % energy), moderately restricted (5–8 % energy) or severely restricted (2–3 % energy) in protein ( Reference Tomé, Chaumontet and Even 49 ) . Protein and AA sufficiency is controlled in the body including a control of the availability of the nine EAA which are not synthesised in the body and must be provided by the diet.

With an adequate or high-protein content of the diet there is no signal of AA insufficiency, and the control of food intake is mainly driven by the need for energy although conditioning and learning processes contribute to maintain a motivation for consuming protein to prevent protein deficiency. High-protein feeding usually stimulates satiety pathways by increasing anorexigenic signals and reducing both orexigenic signals and the sensitivity to feeding of reward-driven mechanisms in the brain ( Reference Darcel, Fromentin and Raybould 50 – Reference Tome, Schwarz and Darcel 53 ) . This is associated with low ghrelin and high leptin plasma concentrations, low neuropeptide Y and high proopiomelanocortin levels in the hypothalamus ( Reference Tomé, Chaumontet and Even 49 , Reference Faipoux, Tome and Gougis 51 , Reference Chaumontet, Recio and Fromentin 54 – Reference Zeeni, Nadkarni and Bell 57 ) , increased neuronal activity in the nucleus of the tractus solitary ( Reference Faipoux, Tome and Gougis 51 , Reference Fromentin, Darcel and Chaumontet 58 , Reference Schwarz, Burguet and Rampin 59 ) , reduced neuronal activity in the amygdala ( Reference Min, Tuor and Koopmans 60 ) and lower sensitivity of dopamine-dependent reward pathways to feeding food and protein ( Reference Chaumontet, Recio and Fromentin 54 , Reference Drummen, Dorenbos and Vreugdenhil 61 – Reference Leidy, Ortinau and Douglas 63 ) . This is also associated with lower body weight gain, and fat mass without affecting lean body mass ( Reference Bensaid, Tome and L'Heureux-Bourdon 64 – Reference Jean, Rome and Mathe 66 ) . Such high-protein diets have been repeatedly discussed in the context of body weight management and prevention or treatment of overweight and obesity ( Reference Journel, Chaumontet and Darcel 52 , Reference Leidy, Clifton and Astrup 67 , Reference Westerterp-Plantenga, Nieuwenhuizen and Tome 68 ) . In human subjects, a protein threshold of at least 30 g protein is required to increase fullness ratings and to elicit satiety responses compared to low-protein preloads ( Reference Dhillon, Craig and Leidy 69 , Reference Paddon-Jones and Leidy 70 ) . Interestingly, in rats submitted to food selection with protein-rich food, there is a trend to choose a high level of protein intake that is often significantly above the protein intake required to meet protein needs derived from nitrogen balance ( Reference Azzout-Marniche, Chalvon-Demersay and Pimentel 71 – Reference Makarios-Lahham, Roseau and Fromentin 73 ) .

With a diet marginally low in protein, the metabolic needs for protein are probably the main determinant of food intake and feeding behaviour, with different strategies such as a preference for protein-rich foods when a choice is offered or, when no food choice is allowed, hyperphagia and an increase in food intake ( Reference Chaumontet, Recio and Fromentin 54 , Reference Azzout-Marniche, Chalvon-Demersay and Pimentel 71 ) . Many studies indicate that, to preserve the protein intake with diets marginally reduced in protein content, subjects usually tend to eat more than a control group fed an adequate protein diet if they are not offered a choice in which protein-rich foods are included ( Reference Chalvon-Demersay, Even and Tome 38 , Reference Blais, Chaumontet and Azzout-Marniche 74 – Reference Sorensen, Mayntz and Raubenheimer 77 ) . Feeding moderately deficient low-protein diet or marginally deficient in some AA more often increases orexigenic pathways and appetite and motivation for food and induces an appetite and a preference for protein-rich foods ( Reference Gibson and Booth 78 – Reference Griffioen-Roose, Smeets and van den Heuvel 80 ) , to increase protein intake and compensate for protein and AA deficiency ( Reference Du, Higginbotham and White 75 , Reference Sorensen, Mayntz and Raubenheimer 77 , Reference Beaton, Feleki and Stevenson 81 – Reference White, He and Dean 84 ) . This correlates with low concentrations of leptin, insulin and IGF1, and high concentrations of ghrelin and FGF21 in the plasma with high levels of neuropeptide Y and corticotropin releasing hormone and low level of proopiomelanocortin ( Reference Chalvon-Demersay, Even and Tome 38 , Reference Chaumontet, Recio and Fromentin 54 , Reference White, He and Dean 84 – Reference Vinales, Begaye and Bogardus 90 ) in the hypothalamus, and with increased sensitivity of central regions involved in reward and increased response of reward-driven mechanisms to foods, protein-rich foods ( Reference Chaumontet, Recio and Fromentin 54 ) and savoury food cues ( Reference Griffioen-Roose, Smeets and van den Heuvel 80 ) . In both animals and human subjects, a moderate protein deficiency produced by foods with a low protein content or by protein deprivation induces a specific appetite for protein ( Reference Booth 91 , Reference Gibson, Wainwright and Booth 92 ) .

In rats a moderately low-protein diets induces hyperphagia, but the overconsumption of food remains limited and does not allow to meet an adequate intake of protein, and therefore, body protein is decreased, whereas growth and different metabolic pathways are altered ( Reference Blais, Chaumontet and Azzout-Marniche 74 , Reference Du, Higginbotham and White 75 , Reference Sorensen, Mayntz and Raubenheimer 77 ) . The higher energy intake also leads to a risk of excess fat deposition and body fats are increased, but this is also associated with an increase in energy expenditure that compensates in part for the overconsumption of energy and the resulting fat accumulation and adiposity ( Reference Chalvon-Demersay, Even and Tome 38 , Reference Blais, Chaumontet and Azzout-Marniche 74 , Reference Du, Higginbotham and White 75 , Reference White, He and Dean 84 , Reference Donald, Pitts and Pohl 93 , Reference Pezeshki, Zapata and Singh 94 ) . In human subjects, energy intake is also increased with a low-protein diet with either a low-protein high-fat diet or a low-protein high-carbohydrate diet ( Reference Gosby, Conigrave and Raubenheimer 95 ) , and this is also associated with an increase in energy expenditure ( Reference Vinales, Begaye and Bogardus 90 ) that could prevent in part for the excess fat accumulation. The origin of the increase in energy expenditure in low-protein-fed subjects remains unclear and has been related to an increase in both diet-induced thermogenesis in adipose tissue, the cost of muscle activity and spontaneous activity ( Reference Blais, Chaumontet and Azzout-Marniche 74 , Reference Pezeshki, Zapata and Singh 94 , Reference Aparecida de Franca, Dos Santos and Garofalo 96 – Reference de Franca, dos Santos and Przygodda 99 ) . It also remains unclear whether the energy expenditure in low-protein-fed subjects is secondary to the increased energy intake or if, inversely, it represents the primary specific response that is responsible for the increase in energy intake, because according to some studies, the higher energy expenditure induced by low-protein diets can occur without hyperphagia ( Reference Hill, Berthoud and Munzberg 36 , Reference Hill, Laeger and Albarado 100 – Reference Rothwell and Stock 102 ) .

Severely deficient protein diets (2–3 % energy as protein) or devoid in one EAA usually depress food intake and induce lower weight, fat mass and lean tissues, and lower plasma protein concentrations that are characteristic of the status of protein deficiency ( Reference Chaumontet, Recio and Fromentin 54 , Reference Du, Higginbotham and White 75 ) . When diets are severely deficient or devoid of protein or at least one EAA, both protein or AA intake cannot be efficiently increased, and this leads to metabolic imbalance that induces an aversive response to the diet by a learning process, allowing for detection and rejection of these diets ( Reference Rogers and Leung 103 ) . A diet severely deficient in at least one EAA is rapidly rejected by rats, and the animals can recognise the missing EAA when offered a choice between different foods ( Reference Gietzen, Hao and Anthony 104 ) . The deficiency is very rapidly detected within the hour following the ingestion in relation to the decrease in the corresponding EAA in the blood, leading to a rapid decrease in food intake. An incomplete protein diet is also recognised in human subjects and results in a decrease in food intake, through a signal of hunger suppression rather than satiation or satiety ( Reference Nieuwenhuizen, Hochstenbach-Waelen and Veldhorst 105 , Reference Veldhorst, Nieuwenhuizen and Hochstenbach-Waelen 106 ) . An extremely low-protein diet or diets severely deficient in at least one EAA induce imbalanced plasma and brain AA patterns, producing a conditioned taste aversion to the diet ( Reference Even, Rolland and Feurte 107 – Reference Feurte, Tome and Gietzen 109 ) . In this process, the decrease of the deficient EAA in the plasma, cerebrospinal fluid and brain is sensed by GCN2 that subsequently triggers a glutamatergic signalling ( Reference Hao, Sharp and Ross-Inta 110 ) . The associated loss of γ-aminobutyric acid (GABA)ergic/inhibitory control contributes to activating glutamatergic excitatory circuits, which project to different brain regions, leading to the modification of feeding behaviour ( Reference Anthony and Gietzen 111 ) .

Protein supply is required for the development and maintenance of body composition. A balanced diet provides an adequate quantity of protein (average requirement for adults is 0⋅66 g/kg/d according to WHO/FAO) containing an adequate quantity of each of the nine EAA to support protein synthesis and achieve nitrogen balance.

A severely protein deficient diet induces lean body mass losses, and ingestion of sufficient dietary energy and protein is a prerequisite for body protein synthesis and maintenance of muscle, bone and other lean tissues and their functions. Proteins, the main compartment of AA in the body, are in constant turnover with free-AA through protein synthesis and degradation, and a small fraction of free-AA that is lost by oxidation in the mitochondria is replaced by AA uptake and non-EAA de novo synthesis. Protein synthesis, protein degradation and AA oxidation are regulated by AA availability in the cell ( Reference Hinnebusch, Ivanov and Sonenberg 112 – Reference Kimball and Jefferson 116 ) . In the fed state with increased intracellular AA concentrations, protein synthesis is increased, protein degradation is reduced and AA oxidation is increased, whereas the inverse processes are induced in the fasting state. Body protein content is related to protein intake up to a plateau considered to indicate a well-balanced protein status, and above this plateau there is no more protein deposition and additional dietary AA are oxidised and lost. The exact amount of protein intake to meet the requirements for body remodelling is currently defined as the minimum amount resulting in a whole-body nitrogen and protein net balance but this remains debated due to difficulties in the identification of the more relevant markers to be used (e.g. nitrogen balance, whole-body protein mass, muscle mass, bone mass, physical function, immune function and metabolic function).

There are several mechanisms by which dietary protein improves muscle and bone mass and strength ( Reference Gaffney-Stomberg, Insogna and Rodriguez 117 ) . Increasing dietary protein increases circulating levels of IGF1, a key regulatory factor of bone growth but also of the skeletal response to anabolic signals, and conversely, a low-protein diet decreases IGF1 ( Reference Bonjour, Schurch and Chevalley 118 ) . Increasing the availability of AA induces an anabolic state that stimulates muscle protein synthesis and a mild suppression of protein breakdown with protein synthesis exceeding breakdown and leading to a positive protein balance ( Reference Biolo, Tipton and Klein 119 – Reference Wolfe 121 ) . There is a direct relationship between intracellular appearance of AA and muscle protein synthesis up to a plateau occurring with ingestion of 20–35 g high-quality protein ( Reference Deutz and Wolfe 122 – Reference Witard, Jackman and Breen 124 ) . However, with increasing protein ingestion, if protein synthesis rapidly increases but reaches a plateau, in contrast protein breakdown could decrease more slightly but more continuously even above the amount of protein intake at which synthesis reach a plateau. This could lead to an extra net protein balance due to decreased breakdown although this was not measured at the muscle level but mainly at the whole body level ( Reference Dideriksen, Reitelseder and Holm 125 , Reference Kim, Schutzler and Schrader 126 ) . EAA stimulates protein synthesis while non-EAA does not have any additional effect, and the intracellular availability of EAA is the primary determinant of the rate of muscle protein synthesis ( Reference Volpi, Kobayashi and Sheffield-Moore 127 ) . EAA (especially leucine) and insulin are anabolic stimuli for muscle and share a common pathway of action via activation of mTOR ( Reference Wullschleger, Loewith and Hall 128 ) . The branched chain AA leucine is a key regulator of anabolic signalling in skeletal muscles and bind with Sestrin2 to induce protein synthesis ( Reference Anthony, Anthony and Kimball 6 , Reference Borack and Volpi 30 , Reference Volpi, Kobayashi and Sheffield-Moore 127 , Reference Anthony, Yoshizawa and Anthony 129 – Reference Glynn, Fry and Drummond 131 ) . However, although leucine initiates the translational processes, the other EAA are probably required to efficiently induce a protein synthesis response following protein intake in human subjects ( Reference Wolfe 132 ) .

Dietary protein is important for muscle and bone development and maintenance. Short-term bed rest or disuse accelerates the loss of muscle mass, function and glucose intolerance ( Reference Galvan, Arentson-Lantz and Lamon 133 ) . It is known that muscle inactivity leads to loss of muscle mass, loss of muscle strength and reduced muscle oxidative capacity in human subjects ( Reference Dideriksen, Reitelseder and Holm 125 ) . Muscle protein synthesis response to protein intake is reduced in immobilised muscle and an elevated protein intake could be required to maintain whole-body postabsorptive protein turnover during inactivity ( Reference Biolo, Ciocchi and Lebenstedt 134 , Reference Stuart, Shangraw and Peters 135 ) . The onset of age-related bone and muscle losses associated with osteoporosis and sarcopoenia may be increased or decreased according to lifestyle practices in early-middle age ( Reference Paddon-Jones and Leidy 70 ) . The loss of muscle and bone tends to occur at approximately the same time, and changes in muscle and bone mass are correlated ( Reference Cohn, Vaswani and Zanzi 136 , Reference Frost 137 ) . Reduction in muscle mass and functional capacity related to ageing account for 3–8 % reduction in muscle mass per decade, starting in the fourth or fifth decade of life. There is evidence that mTORC1- and ATF4-mediated AA-sensing pathways, triggered by impaired AA delivery to skeletal muscle, reduced the anabolic responses to feeding, even if the anabolic sensitivity of tissues is not directly impaired ( Reference Kramer, Verdijk and Hamer 138 , Reference Moro, Ebert and Adams 139 ) . Maintaining adequate protein intake with age may help to preserve muscle mass and strength ( Reference Sahni, Mangano and Hannan 140 ) . There is an ongoing discussion on the optimal protein intake in older adults that could be higher than usually proposed and on the optimal feeding profile between meals ( Reference Devries, McGlory and Bolster 141 – Reference Phillips, Chevalier and Leidy 145 ) . Increasing dietary protein also resulted in lower markers of bone resorption and higher circulating levels of IGF1 in healthy older men and women ( Reference Dawson-Hughes, Harris and Rasmussen 146 ) . Additionally, IGF1 and skeletal muscle fibre decrease in older women fed a low-protein diet, suggesting that increasing IGF1 by increasing dietary protein intake may promote muscle protein synthesis ( Reference Castaneda, Gordon and Fielding 147 ) .

Muscle inactivity decreases and muscle activity increases muscle protein synthesis and the net balance of AA incorporated into muscle protein, but gain in muscle mass and function due to exercise require regular training over an extended period of time ( Reference Dideriksen, Reitelseder and Holm 125 , Reference Cermak, Res and de Groot 148 ) . Exercise sensitises the muscle to AA and potentiates the rise in protein synthesis that, when repeated over time, results in gradual radial growth of skeletal muscle and improved muscle strength and function ( Reference Loenneke, Loprinzi and Murphy 149 , Reference Stokes, Hector and Morton 150 ) . Protein supplementation during resistance exercise training increases muscle mass gain with a protein dose-dependent effect on translational regulation through mTORC1 signalling that could be enhanced by an adequate leucine intake ( Reference Witard, Jackman and Breen 124 , Reference Phillips 144 , Reference Phillips, Chevalier and Leidy 145 , Reference Davies, Carson and Jakeman 151 – Reference Schoenfeld and Aragon 153 ) . Regular physical activity may preserve and even enhance the responsiveness of skeletal muscle to protein intake in older subjects ( Reference Franzke, Neubauer and Cameron-Smith 142 ) . The protein synthesis response to protein intake can also be enhanced by prior completion of resistance exercise in older subjects but this effect could be somewhat lower in older compared to young individuals ( Reference Dideriksen, Reitelseder and Holm 125 , Reference Cermak, Res and de Groot 148 , Reference Moro, Brightwell and Deer 154 – Reference Shad, Thompson and Breen 157 ) .

An adequate protein status is required for survival to balance nitrogen losses, for the development and maintenance of body composition, and to support most if not all physiological functions. Protein quality is based on EAA content to meet metabolic needs. The control of protein sufficiency proceeds through complex sensing and signalling pathways that control protein and food consumption and the metabolic response to feeding. Ingestion of protein promotes body protein synthesis and maintenance of lean and muscle mass and function and this effect is potentiated by exercise. These processes are particularly important to prevent lean body mass losses and decreased muscle strength and bone health with ageing. There is a direct relationship between intracellular appearance of AA and the rate of muscle protein synthesis. With increasing protein ingestion, body and muscle protein synthesis increases fast but reaches a plateau above which additionally provided AA are catabolised. Whether ageing is associated with diminished accretion of muscle proteins after the ingestion of protein or EAA remains under discussion. The strategies to improve protein synthesis, lean mass and muscle performance in older subjects, include per meal dose and frequency of protein consumption. The protein synthesis response to protein intake can also be enhanced by prior completion of resistance exercise but this effect could be somewhat reduced in older compared to young individuals and gain in muscle mass and function due to exercise requires regular training over an extended period.

Financial Support

This research activity was supported by INRAE.

Conflict of Interest

The authors had joint responsibility for all aspects of preparation of this paper.

This article has been cited by the following publications. This list is generated based on data provided by Crossref .

• Google Scholar

View all Google Scholar citations for this article.

Save article to Kindle

To save this article to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle .

Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

• Volume 80, Issue 2
• Daniel Tomé (a1) , Simon Benoit (a1) and Dalila Azzout-Marniche (a1)
• DOI: https://doi.org/10.1017/S0029665120007880

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox .

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive .

Reply to: Submit a response

- No HTML tags allowed - Web page URLs will display as text only - Lines and paragraphs break automatically - Attachments, images or tables are not permitted

Your details

Your email address will be used in order to notify you when your comment has been reviewed by the moderator and in case the author(s) of the article or the moderator need to contact you directly.

You have entered the maximum number of contributors

Conflicting interests.

Please list any fees and grants from, employment by, consultancy for, shared ownership in or any close relationship with, at any time over the preceding 36 months, any organisation whose interests may be affected by the publication of the response. Please also list any non-financial associations or interests (personal, professional, political, institutional, religious or other) that a reasonable reader would want to know about in relation to the submitted work. This pertains to all the authors of the piece, their spouses or partners.

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Journal Proposal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Journal Menu

• Nutrients Home
• Aims & Scope
• Editorial Board
• Reviewer Board
• Topical Advisory Panel
• Instructions for Authors
• Special Issues
• Sections & Collections
• Article Processing Charge
• Indexing & Archiving
• Editor’s Choice Articles
• Most Cited & Viewed
• Journal Statistics
• Journal History
• Journal Awards
• Society Collaborations
• Conferences
• Editorial Office

Journal Browser

• arrow_forward_ios Forthcoming issue arrow_forward_ios Current issue
• Vol. 16 (2024)
• Vol. 15 (2023)
• Vol. 14 (2022)
• Vol. 13 (2021)
• Vol. 12 (2020)
• Vol. 11 (2019)
• Vol. 10 (2018)
• Vol. 9 (2017)
• Vol. 8 (2016)
• Vol. 7 (2015)
• Vol. 6 (2014)
• Vol. 5 (2013)
• Vol. 4 (2012)
• Vol. 3 (2011)
• Vol. 2 (2010)
• Vol. 1 (2009)

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

Nutrition and Regulation of Muscle Protein Synthesis

• Print Special Issue Flyer
• Special Issue Editors

Special Issue Information

Benefits of publishing in a special issue.

• Published Papers

A special issue of Nutrients (ISSN 2072-6643). This special issue belongs to the section " Sports Nutrition ".

Deadline for manuscript submissions: closed (20 May 2023) | Viewed by 41582

Share This Special Issue

Special issue editor.

Dear Colleagues,

Skeletal muscle is a crucial tissue for human health since it controls several metabolic activities, including thermal regulation, nutritional balance, glucose uptake, and the endocrine system. Alterations in body glucose homeostasis, falls, fractures, disability, and chronic diseases are associated with loss of skeletal muscle mass. Protein synthesis and degradation are two important regulated processes that act in concert to maintain muscle mass. In atrophic conditions, protein synthesis impairment is associated with several conditions, including physical inactivity, sarcopenia, aging- and muscle-wasting-related diseases, as well as malnutrition, corticosteroid therapies, or inflammation. The availability of nutrients is one of the factors which can influence protein turnover. In fact, when muscle protein breakdown exceeds the rate of muscle protein synthesis, loss of protein occurs involving the ubiquitin-proteasomal system, autophagy, and the calpain signaling. The knowledge of these pathways helps to understand their role in muscle remodeling and in response to diet, knowing that a balanced nutrition intake represents a potential clinical intervention to reactivate protein synthesis during atrophy. The aim of this Special Issue is to collect original articles or reviews which discuss the therapeutic strategies, based on current nutrition interventions, able to modulate muscle protein turnover in muscle wasting conditions. 

Dr. Sara Salucci Guest Editor

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website . Once you are registered, click here to go to the submission form . Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Nutrients is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2900 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• skeletal muscle disorders
• protein synthesis
• protein degradation
• micronutrients and macronutrients
• Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
• Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
• Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
• External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
• e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here .

Published Papers (10 papers)

Jump to: Research , Review

Jump to: Editorial , Review

Graphical abstract

Jump to: Editorial , Research

Further Information

Mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed, biochemistry, protein synthesis, affiliations.

• 1 Kaiser Permanente Oakland Medical Center
• 2 University of Louisville
• PMID: 31424745
• Bookshelf ID: NBK545161

Our understanding of each of the biological sciences becomes heightened by the study of biochemistry and molecular biology. In the last few decades, advances in laboratory techniques for the study of these microscopic sciences have led us to a greater understanding of the central dogma of molecular biology – that DNA transcribes RNA which then gets translated into protein. Understanding protein synthesis is paramount in studying various medical fields, from the molecular basis of genetic diseases through antibiotic development to expressing recombinant proteins as drugs or clinical laboratory reagents. As one of the foundational concepts in biology, protein synthesis is sufficiently complex that many believe it evolved once, giving the protein synthetic machinery in all organisms on the planet a common ancestry. Despite having certain underlying similarities in their mechanism, protein synthesis in the three major lines of descent (bacteria, archaea, and eukaryotes) has diverged to the point that substantive mechanistic differences have arisen. These differences have been exploited in nature as organisms produce compounds targeting the protein synthetic machinery of competitors as they vie for limited resources. Science has modified many of these compounds that target the machinery for protein synthesis in pathogenic microbes for use in the clinic as antibiotics. As our understanding of the mechanisms of protein synthesis continues to grow, there will likely be countless additional applications for this knowledge in medicine, research, and industry.

Copyright © 2024, StatPearls Publishing LLC.

PubMed Disclaimer

Conflict of interest statement

Disclosure: Jacob Hoerter declares no relevant financial relationships with ineligible companies.

Disclosure: Steven Ellis declares no relevant financial relationships with ineligible companies.

• Introduction
• Fundamentals
• Issues of Concern
• Cellular Level
• Pathophysiology
• Clinical Significance
• Review Questions

Similar articles

• An Experimental Approach to Genome Annotation: This report is based on a colloquium sponsored by the American Academy of Microbiology held July 19-20, 2004, in Washington, DC. [No authors listed] [No authors listed] Washington (DC): American Society for Microbiology; 2004. Washington (DC): American Society for Microbiology; 2004. PMID: 33001599 Free Books & Documents. Review.
• The Global Genome Question: Microbes as the Key to Understanding Evolution and Ecology: This report is based on a colloquium, “The Global Genome Question: Microbes as the Key to Understanding Evolution and Ecology,” sponsored by the American Academy of Microbiology and held October 11-13, 2002, in Longboat Key, Florida. [No authors listed] [No authors listed] Washington (DC): American Society for Microbiology; 2004. Washington (DC): American Society for Microbiology; 2004. PMID: 33119236 Free Books & Documents. Review.
• Japan-China Joint Medical Workshop on Drug Discoveries and Therapeutics 2008: The need of Asian pharmaceutical researchers' cooperation. Nakata M, Tang W. Nakata M, et al. Drug Discov Ther. 2008 Oct;2(5):262-3. Drug Discov Ther. 2008. PMID: 22504718
• Translational Metabolomics of Head Injury: Exploring Dysfunctional Cerebral Metabolism with Ex Vivo NMR Spectroscopy-Based Metabolite Quantification. Wolahan SM, Hirt D, Glenn TC. Wolahan SM, et al. In: Kobeissy FH, editor. Brain Neurotrauma: Molecular, Neuropsychological, and Rehabilitation Aspects. Boca Raton (FL): CRC Press/Taylor & Francis; 2015. Chapter 25. In: Kobeissy FH, editor. Brain Neurotrauma: Molecular, Neuropsychological, and Rehabilitation Aspects. Boca Raton (FL): CRC Press/Taylor & Francis; 2015. Chapter 25. PMID: 26269925 Free Books & Documents. Review.
• Chronoastrobiology: proposal, nine conferences, heliogeomagnetics, transyears, near-weeks, near-decades, phylogenetic and ontogenetic memories. Halberg F, Cornélissen G, Regal P, Otsuka K, Wang Z, Katinas GS, Siegelova J, Homolka P, Prikryl P, Chibisov SM, Holley DC, Wendt HW, Bingham C, Palm SL, Sonkowsky RP, Sothern RB, Pales E, Mikulecky M, Tarquini R, Perfetto F, Salti R, Maggioni C, Jozsa R, Konradov AA, Kharlitskaya EV, Revillam M, Wan C, Herold M, Syutkina EV, Masalov AV, Faraone P, Singh RB, Singh RK, Kumar A, Singhs R, Sundaram S, Sarabandi T, Pantaleoni G, Watanabe Y, Kumagai Y, Gubin D, Uezono K, Olah A, Borer K, Kanabrockia EA, Bathina S, Haus E, Hillman D, Schwartzkopff O, Bakken EE, Zeman M. Halberg F, et al. Biomed Pharmacother. 2004 Oct;58 Suppl 1:S150-87. doi: 10.1016/s0753-3322(04)80025-8. Biomed Pharmacother. 2004. PMID: 15754855
• Schmeing TM, Ramakrishnan V. What recent ribosome structures have revealed about the mechanism of translation. Nature. 2009 Oct 29;461(7268):1234-42. - PubMed
• Chee NT, Lohse I, Brothers SP. mRNA-to-protein translation in hypoxia. Mol Cancer. 2019 Mar 30;18(1):49. - PMC - PubMed
• Harish A, Caetano-Anollés G. Ribosomal history reveals origins of modern protein synthesis. PLoS One. 2012;7(3):e32776. - PMC - PubMed
• Tolsma TO, Hansen JC. Post-translational modifications and chromatin dynamics. Essays Biochem. 2019 Apr 23;63(1):89-96. - PubMed
• Simonović M, Steitz TA. A structural view on the mechanism of the ribosome-catalyzed peptide bond formation. Biochim Biophys Acta. 2009 Sep-Oct;1789(9-10):612-23. - PMC - PubMed

Publication types

• Search in PubMed
• Search in MeSH
• Add to Search

Related information

Linkout - more resources, full text sources.

• NCBI Bookshelf

Research Materials

• NCI CPTC Antibody Characterization Program

• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 27 September 2024

Tyrp1 is the mendelian determinant of the Axolotl ( Ambystoma mexicanum ) copper mutant

• Raissa F. Cecil 1 ,
• Lloyd Strohl 2 ,
• Maddie K. Thomas 1 ,
• James L. Schwartz 1 ,
• Nataliya Timoshevskaya 3 ,
• Jeramiah J. Smith 3 &
• S. Randal Voss 1  

Scientific Reports volume  14 , Article number:  22399 ( 2024 ) Cite this article

20 Accesses

Metrics details

• Disease model
• Herpetology

Several dozen Mendelian mutants have been discovered in axolotl ( Ambystoma mexicanum ) populations, including several that affect pigmentation. Four recessive mutants have been described in the scientific literature and genes for three of these have been identified. Here we describe and genetically dissect copper , a mutant with an albino-like phenotype known only from the pet trade. We performed a cross segregating copper and wildtype color phenotypes and used bulked segregant RNA-Seq to identify a region on chromosome 6 that was enriched for single-nucleotide polymorphisms (SNPs) between the color phenotypes. This region included Tyrosinase-like Protein 1 ( Tyrp1 ), a melanin synthesis protein that when mutated, is associated with lighter than black melanin coloration in animal models and oculocutaneous albinism in humans. Inspection of RNA-Seq reads identified a single nucleotide deletion that is predicted to change the coding frame, introduce a premature stop codon in exon 6 and yield a truncated Tyrp1 protein in copper individuals. Using CRISPR-Cas9 editing, we show that wildtype Tyrp1 crispants exhibit copper pigmentation, thus confirming Tyrp1 as the copper locus. Our results suggest that commercial and hobbyist axolotl populations may harbor useful mutants for biological research.

Similar content being viewed by others

Identification of mutant gene for Black crystal coat and non-allelic gene interactions in Neogale vison

Integrative functional genomic analyses identify genetic variants influencing skin pigmentation in Africans

Dopachrome tautomerase variants in patients with oculocutaneous albinism

Introduction.

The axolotl ( Ambystoma mexicanum ) has a long and storied history in biological research. Axolotls provide models for studies of embryonic and post-embryonic development, including most famously the study of whole organ regeneration 1 . The primary stock center for axolotl research is a captive bred population that has been maintained for almost 100 years 2 , 3 . Many mutants have been maintained in this population over time, including four different color mutants that are determined by single genes: white , melanoid , albino , and axanthic . Each of these mutants has an interesting history. The white mutant was presumably collected from Mexico in 1862 along with 33 wildtype individuals 4 . These axolotls were transported to Paris to establish the first laboratory axolotl population. The white mutant is caused by a Edn3 splicing defect 3 . The melanoid mutant was originally identified from laboratory crosses made using wild-caught axolotls from Mexico, and like white , melanoid is presumably also a natural color variant 5 . Subsequent analyses revealed that melanoid is genetically associated with Ltk 6 . Similarly, albino was originally identified in a wild captured tiger salamander and crossed into axolotl stocks 7 . In contrast, axanthic appears to have arisen spontaneously within a laboratory axolotl strain 8 . Although the gene for axanthic has not been identified, Woodcock et al. 3 showed that albino is caused by a deletion in the Tyr coding sequence.

For as long as axolotls have been studied in research, they have also been highly prized as aquarium pets. All axolotl color mutants known to biological research, and even transgenic axolotls, are available in the pet trade. Additionally, pet breeders have identified color variants that have not received biological study, including a recessive Mendelian mutant called copper . Pet breeders describe copper axolotls as having copper-colored bodies with yellow xanthophores, brownish melanophores, and iridescent iridophores. The brownish and not black color of melanophores suggests copper is a form of albinism, perhaps involving mutation of a melanin synthesis protein. Given the importance of animal models in the study of albinism diseases, we used bulked segregant RNA-Seq (BSR-Seq) 9 and CRISPR-Cas9 10 to identify Tyrp1 as the copper locus. Our results suggest that commercial and private axolotl collections may harbor genetic variants that may prove useful as animal models in biological research.

BSR-Seq 9 was used to identify SNPs linked to the copper locus. Embryos from a cross that segregated copper and wildtype individuals were used to create two copper and two wildtype RNA pools ( N  = 36 and 27 for copper and wildtype respectively, per pool) that were each subjected to Illumina short-read RNA sequencing to identify single nucleotide polymorphisms (SNPs). Analysis of SNPs across the axolotl genome revealed a region on chromosome 6p with dissimilar frequencies between the pools (Fisher’s exact test F = 686, p  = 2.93E −123 at peak association) (Fig.  1 a; Supplementary File 1). Examination of genes within this region identified Tyrosinase-like protein 1 ( Tyrp1 ) as a candidate gene. Tyrp1 encodes an enzyme that functions in melanin synthesis. Tyrp1 mutations generate lighter than black melanin shades of color in animal models 11 , 12 , 13 and Oculocutaneous Albinism 3 (OCA3) in humans 14 . Additionally, significantly more Tyrp1 transcripts were identified from the wild-type vs. copper RNA-Seq pools (Dseq2 Wald Test = 20.03, adjusted P-value = 1.06E-84) (Fig.  1 b). The axolotl Tyrp1 locus (AMEX60DD301043361.1) is distributed across 8 exons and the 1663 bp coding sequence encodes a 524 amino acid protein. A single nucleotide deletion (chr6p:1,225,541,490) was identified in RNA-Seq reads generated for the copper RNA pool (Fig.  1 c; Supplementary File 2). This deletion, in exon 6, was confirmed by sequencing genomic DNA isolated from two copper and two wildtype individuals that were used to construct the RNA BSR-Seq pools (Fig.  1 c). The deletion in the copper Tyrp1 sequence is predicted to change the reading frame and introduce a premature stop codon at amino acid position 416 (Supplementary File 3).

Identification of Tyrp1 as a candidate gene for copper . ( A ) Plot showing genome wide SNP (allele) frequency differences between copper and wildtype BSA-RNA-Seq pools. ( B ) Transcript counts identified for Tyrp1 between copper and wildtype BSA-RNA-Seq pools. The difference between pools is statistically significant (W = 20.03, adjusted P-value = 1.06E-84). ( C ) Genomic map of the axolotl Tyrp1 locus and deletion detected in exon 6 between copper and wildtype alleles.

CRISPR-CAS9 10 was performed to determine if disruption of the Tyrp1 coding sequence would yield individuals that presented copper pigmentation. Two guide RNAs targeting exons 2 and 6 of the Tyrp1 coding sequence were injected into 100, one-cell stage wildtype embryos and visually assessed for color at developmental stage 42 15 . Crispant individuals were characterized by having fewer melanophores than non-injected controls and presented a yellowish overall color, as is typical of copper larvae in the pet trade. PCR and DNA sequencing of Tyrp1 regions targeted by CRISPR-Cas9 confirmed that crispant embryos ( N  = 12) were edited at both gRNA-target sites (Fig.  2 ). Five Tyrp1 crispants (that were edited at both gRNA1 and gRNA2) were reared to approximately 17 cm total body length (1.2 years old) and photographed to document pigmentation. Relative to the yellowish color observed during the larval stage, all five Tyrp1 crispants presented a darker copper skin color that was very similar to the color of adult copper axolotls (Fig.  3 ). The pigmentation phenotypes resulting from CRISPR-Cas9 genome editing strongly implicate Tyrp1 as the gene for copper.

CRISPR-CAS9 knockdown of Tyrp1 . Melanin pigmentation was greatly reduced in F0 Tyrp1 crispant larvae relative to F0 non-injected control siblings. F0 Tyrp1 crispant electropherograms showed overlapping sequence traces at the gRNA target site for forward and reverse DNA sequencing reactions, consistent with genome editing. F0 non-injected control DNA sequence does not show evidence of genome editing. The gRNA target sequence (underlined) overlaps the deleted nucleotide in the copper Tyrp1 allele.

Images of F0 Tyrp1 crispants relative to a wildtype axolotl and copper mutant. Salamanders are 15–18 cm total body length and 1-1.2 years old. A USA copper penny (1.9 cm diameter) is provided as a color and size reference.

Several different axolotl pigmentation variants are present in laboratories and households around the world. Several of these variants, including white , albino and melanoid , have received considerable attention in biological research. Previously, we identified genes for these variants to increase their value as research models 3 , 5 . In this study we identified a new gene in an axolotl pigmentation variant known only from the pet trade. Specifically, we show that copper , an axolotl mutant with copper coloration, is likely determined by a single nucleotide deletion in the Tyrp1 coding sequence. Tyrp1 encodes an enzyme that functions in melanophores to produce a black pigment called melanin. Mutations in Tyrp1 are associated with decreased production of melanin and structural alterations that result in lighter than expected coloration, for example brown coat color in mice 11 and blonde hair in Melanesian humans 16 . copper axolotls in the larval stage have fewer dark pigmented melanophores than those observed in wildtype animals, and in absence of melanophores pigmentation is primarily determined by yellow xanthophores. As copper axolotls age, a brownish pigment emerges and individuals developed a uniform copper skin color, as is typical of animal models with Tyrp1 mutations. To functionally validate Tyrp1 for copper , we generated copper- like pigmentation in wildtype individuals using CRISPR-Cas9 genome editing of Tyrp1 . The combination of SNP genome association data with CRISPR-Cas9 functional genomics results strongly implicates Tyrp1 as the copper locus.

The copper deletion is predicted to alter the Tyrp1 coding sequence and introduce a premature stop codon in exon 6, yielding a truncated protein with an altered function. We note that the exon 6 deletion in copper Tyrp1 parallels an exon 6 deletion identified in the first human case report of OCA3 14 , with both deletions leading to a premature stop codon. No Tyrp1 mRNA or protein was detected in the human case while Tyrp1 transcripts were recovered by BSR-Seq in copper axolotls, albeit at lower transcript abundances. We speculate that Tyrp1 transcript abundances are lower in copper axolotls because the premature stop codon causes nonsense-mediated mRNA decay. Although it remains to be determined if a copper Tyrp1 protein is generated in axolotl, a nonsense mutation in the catalytic, tyrosinase-like domain of Tyrp1 is expected to affect protein function and melanogenesis 17 .

Domestic plant and animal populations have long served as reservoirs for phenotypes of commercial, biological, and biomedical relevance 18 , 19 , 20 . While the pet trade has gained access to axolotls from research labs, including for example GFP transgenics, axolotl researchers have not assessed commercial and domestic populations for new research models. Now that Tyrp1 has been identified as the causative gene for copper , this new axolotl model can be used to study molecular functions underlying OCA3 phenotypes. As is observed in humans with OCA3, copper is characterized by a reduction in melanin. A reduction in melanin could trace to many different cellular mechanisms as Tyrp1 functions in multiple pathways that directly or indirectly regulate melanin biosynthesis, as well as affecting melanophore cell division and death 21 . In addition to copper , other axolotl pigment variants in the pet trade include hypomelanistic, which is characterized by a reduction in melanin and UV-light excitable face/cranium fluorescence, and “starburst” which presents increased numbers of iridophores in albino axolotls. These and other phenotypic variants in the axolotl pet trade may provide useful models for biological research.

Materials and methods

Approval for animal experiments.

Animal care procedures were approved under University of Kentucky IACUC protocol 2017–2580 and performed in accordance with ARRIVE 22 and AAALAC 23 guidelines and standards.

Animal procedures

Axolotls used in this study were obtained from a commercial population (Strohls Herptiles: wildtype and copper sibling larvae) and the Ambystoma Genetic Stock Center (AGSC RRID: SCR_006372; wildtype axolotl embryos RRID: AGSC_100E). All experiments were performed using either 50% (pre-hatching) or 100% (post-hatching) axolotl rearing water (ARW: 1.75 g NaCl, 100 mg MgSO4, 50 mg CaCl2, and 25 mg KCl per liter, buffered with NaHCO3 to pH 7.3–7.5) in a room maintained at 17–18 °C. Larvae were housed in 1 L glass or 2 L polypropylene bowls at either low density (8–10 per bowl) or one per container. After larvae reached 5 cm they were transferred to 9 L boxes on an Aquatic Enterprises recirculating system. Larvae were initially fed newly hatched brine shrimp until 3 cm total body length and then transitioned to California Black worms (J&R Enterprises) until large enough to be fed fish pellets (Rangen). Animals were anesthetized using a 0.02% benzocaine solution. Benzocaine was first dissolved in 4 ml 100% EtOH and then the chemical and solvent were diluted in 1 L of ARW.

Bulked segregant RNA-Seq

A total of 36 copper and 27 wildtype siblings were sampled from a cross between a homozygous copper male and a heterozygous female carrier and reared to approximately 3 cm. Tail tissue was collected from each while under benzocaine anesthesia. The resulting tail tips were pooled into separate 1.7 ml plastic tubes and flash frozen with dry ice to create copper and wildtype bulk segregant pools for RNA isolation. Tail tissue samples were dissociated with 23- and 26-gauge needles, and RNA was isolated with TRIzol and then further purified using a QIAGEN RNeasy Mini Kit with DNase treatment. The resulting RNA pools were used to generate outsourced Poly(A) RNA-seq libraries that were sequenced on an Illumina NovaSeq X Plus (150 bp paired end reads) by Novogene. Reads from each pool of copper or wildtype individuals were mapped to the axolotl genome assembly 24 using HiSat2 25 v. 2.2.0 ( http://daehwankimlab.github.io/hisat2 ). SNPs were identified using BCFtools 26 v. 1.12-57-g0c2765b ( https://samtools.github.io/bcftools ) and SNP frequencies at polymorphic sites were compared between copper and wildtype pools using Popoolation2 27 v. 1.201 ( https://sourceforge.net/projects/popoolation2 ). Additionally, DEseq2 28 v. 1.44.0 ( https://www.bioconductor.org/packages/release/bioc/html/DESeq2.html ) was used to compare Tyrp1 transcript abundances between cooper and wildtype pools. RNA sequence data will be published in the Short Read Archive at the National Center for Biotechnology Information.

Two guide RNAs (gRNAs) (CTGGCCACTGCGGAGAGCCT, TTTGTCCTCCAGTTCCATTC) were designed to target the 2nd and 6th exons, respectively, of axolotl Tyrp1 (TYRP1|AMEX60DD301043361.1). To generate targeted mutations, gRNAs were first duplexed with Alt-R tracrRNA, aliquoted and stored at -80⁰ C. All RNA products were synthesized by Integrated DNA Technologies. Guide RNAs and Alt-R tracrRNA were mixed with Cas9 to generate ribonucleoproteins that were injected into 100, 1-cell stage wildtype AGSC embryos as described previously 29 . A total of 12 injected embryos were reared to approximately 3 cm and tail tissue was collected from each while under benzocaine anesthesia. During animal rearing, larvae were fed brine shrimp. The resulting tail tips were placed into separate 1.7 ml plastic tubes and placed on ice for DNA isolation. DNA isolations were performed using a New England Biolabs (NEB) Monarch genomic DNA isolation kit. DNA concentrations were determined using a Nanodrop (Thermo Scientific) and all samples were diluted to 30 ng / ul for PCR. PCR primers were designed to amplify DNA amplicons spanning gRNA target sites. The flanking PCR primers for gRNA1: Tyrp1_gRNA1_5. 1 TTGGTTTTATAGTTCCAGGTCCCG and Tyrp1_gRNA1_3.1 AGACAGAAGCCATTCATCGACTG. The flanking PCR primers for gRNA2: Tyrp1_gRNA2_5.1 AAGGTGGTTGAATCTTGTCTCCTT and Tyrp1_gRNA2_3.1 TTTTAAGACAGGTTACCCCCGAG. PCR amplicons were treated with NEB Exo-CIP and shipped to Eurofins for Sanger sequencing. The resulting sequences were compared using Geneious Prime software to evaluate CRISPR editing.

Homozygous copper and wildtype (axolotls not carrying copper alleles) larvae were sampled from two separate spawns. PCR was performed using Tyrp1_gRNA2_5.1 and Tyrp1_gRNA2_3.1 primers to amplify Tyrp1 exon 6 genomic sequence from 4 individuals (2 copper , 2, wildtype ) and amplicons were sequenced as described above. The resulting sequences were aligned using Geneious Prime software and searched for polymorphisms.

Data availability

Raw DNA sequence reads and transcript abundance estimates may be found under GEO GSE269079.

Voss, S. R., Epperlein, H. H. & Tanaka, E. M. Ambystoma mexicanum, the axolotl: a versatile amphibian model for regeneration, development, and evolution studies. Cold Spring Harbor Protoc. 8 , artpdbemo128 (2009).

Article   Google Scholar  

Voss, S. R., Woodcock, M. R. & Zambrano, L. A tale of two axolotls. Bioscience . 65 , 1134–1140 (2015).

Article   PubMed   PubMed Central   Google Scholar  

Woodcock, M. R. et al. Identification of mutant genes and introgressed tiger salamander DNA in the laboratory axolotl. Ambystoma mexicanum Sci. Rep. 7 , 6 (2017).

Article   PubMed   Google Scholar  

Smith, H. M. Discovery of the axolotl and its early history in biological research. In: Developmental Biology of the Axolotl (eds (eds Armstrong, J. B. & Malacinski, G. M.) 3–12 (Oxford University Press (1989).

Humphrey, R. R. & Bagnara, J. T. A color variant in the Mexican axolotl. J. Hered . 58 , 251–256 (1967).

Kabangu, M. et al. Leukocyte tyrosine kiltke ( Ltk ) is the mendelian determinant of the axolotl melanoid color variant. Genes (Basil) . 14 , 904 (2023).

Article   CAS   Google Scholar  

Humphrey, R. R. Albino axolotls from an albino tiger salamander through hybridization. J. Hered . 58 , 95–101 (1967).

Article   CAS   PubMed   Google Scholar  

Lyerla, T. A. & Dalton, H. C. Genetic and developmental characteristics of a new color variant, axanthic , in the Mexican axolotl, Ambystoma mexicanum Shaw. Dev. Biol. 24 , 1–18 (1971).

Liu, S., Yeh, C-T., Tang, H. M., Nettleton, D. & Schnable, P. S. Gene mapping via bulked segregant RNA-Seq (BSR-Seq). PLoS ONE . 7 , e36406 (2012).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Doudna, J. A. & Charpentier, E. Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science . 346 , 1258096 (2014).

J Jackson, I. A cDNA encoding tyrosinase-related protein maps to the brown locus in mouse. Proc. Nat. Acad. Sci. U S A . 85 , 4392–4396 (1988).

Article   ADS   CAS   Google Scholar  

Schmidt-Küntzel, A., Eizirik, E., O’Brien, S. J. & Menotti-Raymond, M. Tyrosinase and tyrosinase related protein 1 alleles specify domestic cat coat color phenotypes of the albino and brown loci. J. Hered . 96 , 289–301 (2005).

Peterson, S. M. et al. Genetic variants in melanogenesis proteins TYRP1 and TYR are associated with the golden rhesus macaque phenotype. G3 13, jkad168 (2023).

Boissy, R. E. et al. Mutation in and lack of expression of tyrosinase-related protein-1 (TRP-1) in melanocytes from an individual with brown oculocutaneous albinism: a new subtype of albinism classified as OCA3. Am. J. Hum. Genet. 58 , 1145–1156 (1996).

CAS   PubMed   PubMed Central   Google Scholar  

Bordzilovskaya, N. P., Dettlaff, T. A., Duhon, S. T. & Malacinski, G. M. Developmental stage series of axolotl embryos. In: Developmental Biology of the Axolotl (eds (eds Armstrong, J. B. & Malacinski, G. M.) 201–209 (Oxford University Press (1989).

Kenny, E. E. et al. Melanesian blond hair is caused by an amino acid change in TYRP1. Science . 336 , 554 (2012).

Lai, X., Wichers, H. J., Soler-Lopez, M. & Dijkstra, B. W. Structure of human tyrosinase related protein 1 reveals a binuclear zinc active site important for Melanogenesis. Angew Chem. Int. Ed. Engl. 56 , 9812–9815 (2017).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Bennett, D. C. & Lamoreux, M. L. The color loci of mice–a genetic century. Pigment Cell. Res. 16 , 333–344 (2003).

Andersson, L. Domestic animals as models for biomedical research. Upsala J. Med. Sci. 121 , 1–11 (2016).

Kantar, M. B., Nashoba, A. R., Anderson, J. E., Blackman, B. K. & Rieseberg, L. H. The genetics and genomics of plant domestication. BioScience . 67 , 971–982 (2017).

Kamaraj, B. & Purohit, R. Mutational analysis of oculocutaneous albinism: a compact review. Biomed. Res. Int. 905472 (2014). (2014).

Percie du Sert. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLoS Biol. 18 , e3000410 (2020).

National Research Council. Guide for the Care and Use of Laboratory Animals: Eighth Edition (National Academies, 2011). https://doi.org/10.17226/12910

Schloissnig, S. et al. The giant axolotl genome uncovers the evolution, scaling and transcriptional control of complex gene loci. Proc. Nat. Acad. Sci. U.S.A. 118, e2017176118 (2021).

Kim, D. et al. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat. Biotechnol. 37 , 907–915 (2019).

Li, H. A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics . 27 , 2987–2993 (2011).

Kofler, R., Pandey, R. V. & Schlötterer, C. PoPoolation2: identifying differentiation between populations using sequencing of pooled DNA samples (Pool-Seq). Bioinformatics . 27 , 3435–3436 (2011).

Love, M. I., Huber, W. & Anders, S. Moderated estimation of Fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15 , 550 (2014).

Trofka, A. et al. Genetic basis for an evolutionary shift from ancestral preaxial to postaxial limb polarity in non-urodele vertebrates. Curr. Biol. 31 , 4923–4934 (2021).

Download references

Acknowledgements

This work was funded by the Office of Infrastructure Programs at the National Institutes of Health (R24OD010435, P40OD019794).

Author information

Authors and affiliations.

Department of Neuroscience, Spinal Cord and Brain Injury Research Center, and Ambystoma Genetic Stock Center, University of Kentucky, Lexington, KY, 40536, USA

Raissa F. Cecil, Maddie K. Thomas, James L. Schwartz & S. Randal Voss

10496 Stevens Rd, Vevay, IN, 47043, USA

Lloyd Strohl

Department of Biology, University of Kentucky, Lexington, KY, 40506, USA

Nataliya Timoshevskaya & Jeramiah J. Smith

You can also search for this author in PubMed   Google Scholar

Contributions

R.F.C. performed animal care procedures, collected tissues for RNA and DNA isolation, isolated RNA and DNA, designed PCR primers and gRNAs for CRISPR-CAS9, performed PCR, prepared samples for RNA and DNA sequencing, analyzed and summarized results from PCR and DNA sequencing, and contributed to writing of the manuscript. L.S. contributed to study design, supplied copper and wildtype larvae for RNA-Seq, provided information about axolotl stocks in the pet trade, assessed candidate genes for association to copper, and contributed to writing of the manuscript. M.T. and J.S. performed animal care procedures and took pictures of salamanders. N.T. developed and applied methods and pipelines for bioinformatic analyses and contributed to writing of the manuscript. J.J.S. contributed to study design, developed and applied methods and pipelines for bioinformatic analyses, summarized results from bioinformatic analyses, and contributed to writing of the manuscript. S.R.V. contributed to study design, performed animal care procedures, collected tissues for RNA and DNA isolation, performed embryo microinjections, photographed salamanders, analyzed, and summarized results from the study, and drafted the original manuscript.

Corresponding author

Correspondence to S. Randal Voss .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Supplementary file 1

Manhattan plots showing the degree of association of segregating genotypes with copper phenotype vs. wildtype in BSR-Seq pools. (A) Values shown are -log10(p-values) from Fisher’s exact tests. (B) Values are Z-scores from statistical tests of allele frequency difference.

Supplementary file 2

Nucleotide sequence for wildtype Tyrp1 showing where a nucleotide deletion (nonsense mutation at 1156 bp) occurs in the copper Tyrp1 allele.

Supplementary file 3

Predicted amino acid sequence for wildtype Tyrp1 showing where the copper amino acid sequence is altered by a nonsense mutation that results in a stop codon at amino acid residue 415.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary material 2, supplementary material 3, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Cecil, R.F., Strohl, L., Thomas, M.K. et al. Tyrp1 is the mendelian determinant of the Axolotl ( Ambystoma mexicanum ) copper mutant. Sci Rep 14 , 22399 (2024). https://doi.org/10.1038/s41598-024-73283-1

Download citation

Received : 05 June 2024

Accepted : 16 September 2024

Published : 27 September 2024

DOI : https://doi.org/10.1038/s41598-024-73283-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• Open access
• Published: 28 September 2024

Protective effect of afamin protein against oxidative stress related injury in human ovarian granulosa cells

• Qiang Zhang 1 ,
• Xiaoyu Zheng 2 ,
• Xueying Zhang 1 &
• Lianwen Zheng 1  

Journal of Ovarian Research volume  17 , Article number:  189 ( 2024 ) Cite this article

Metrics details

Ovarian granulosa cells (GCs) play crucial roles in follicular growth and development. Their normal function is influenced by various factors, including oxidative stress, which is a significant factor. Afamin protein is a vitamin E-specific binding protein that acts as a vitamin E carrier in follicular fluid. Although the mechanism of the protective effect of afamin on human ovarian GCs is still unclear, there is evidence it has an antioxidant effect in neuronal cells.

In this study, we investigated the protective effects of afamin proteins on testosterone propionate (TP)-induced ovarian GCs using a human ovarian tumor granulosa cell line (KGN).

The results showed that afamin reduced TP-induced oxidative stress in KGN cells by decreasing the levels of oxidative damage markers, enhancing the activity of antioxidant enzymes, and exerting a protective effect on GCs. Supplementation with afamin repaired mitochondrial dysfunction by improving mitochondrial membrane potential damage and ATP levels. It counteracted TP-induced apoptosis by decreasing the activity of Caspase-3 and upregulating the expression of the anti-apoptotic gene (BCL-2) while downregulating the expression of the pro-apoptotic gene BCL-2-associated X protein (BAX). In addition, afamin regulated the expression of genes related to ovarian steroid hormone synthesis, ameliorating the endocrine dysfunction observed in TP-induced KGN cells.

Therefore, Afamin proteins may serve as important complementary factors that protect GCs from other types of damage, such as oxidative stress, and may help improve ovarian follicle quality and female reproductive function.

Introduction

Ovarian granulosa cells (GCs) regulate oocyte development and are critical for follicle growth [ 1 ]. These cells primarily secrete estrogen and progesterone, with cytochrome P450 aromatase (CYP19A1) playing a crucial role in the synthesis of estrogen. Additionally, 3β-hydroxysteroid dehydrogenase (HSD3B1) is involved not only in the production of progesterone but also in the conversion of androgens [ 2 ]. GCs and oocytes exchange substances and signals through gap junctions to provide the most nutrients required for oocyte growth and development. They also interact with each other to regulate the processes of follicular maturation, ovulation, and fertilization [ 3 , 4 ]. Damage to GCs results in pathophysiological changes in ovaries. In a hyperandrogenic microenvironment, excess androgens inhibit the proliferation of GCs and promote apoptosis by acting through the androgen receptor (AR). This affects follicular development and leads to follicular atresia, resulting in polycystic changes of the ovary and causing the development of Polycystic Ovary Syndrome (PCOS) [ 5 , 6 ]. Damage to ovarian GCs can negatively affect female reproductive health.

Oxidative stress (OS), which is an imbalance between the production of reactive oxygen species (ROS) and the body’s antioxidant defense system [ 7 ], can also harm ovarian function. It is important to control ROS levels to maintain normal ovarian function as GCs are sensitive to ROS. Elevated ROS levels can disrupt redox signaling and redox-regulated cellular processes, leading to an oxidative stress (OS) response in GCs. This can cause DNA damage in GCs and affect their function [ 8 , 9 , 10 ]. Antioxidants have been shown to inhibit GC progression, restore disturbances in cellular redox balance, and reduce ROS levels in GCs, improving ovarian function [ 11 ]. Thus, inhibition of OS is considered an effective method to alleviate reproductive dysfunction caused by oxidative damage to GCs.

Afamin, the fourth member of the albumin gene family, has been identified as a vitamin E-specific binding protein [ 12 ]. Under physiological conditions, afamin acts as a vitamin E carrier in body fluids such as human follicular fluid [ 13 , 14 ]. Afamin proteins have multiple binding sites with specific affinities for α-tocopherol and γ-tocopherol. These tocopherols can substitute for vitamin E transport in body fluids when the lipoprotein system is inadequate [ 13 ]. Vitamin E is an essential antioxidant. Afamin protein, a specific binding protein for vitamin E, has been shown to protect cortical neurons from apoptosis and act as an antioxidant against neuronal emergence of oxidative stress in neuronal cellular assays in vitro. This effect is observed either alone or in synergy with vitamin E [ 15 , 16 ]. However, the role of afamin in ovarian GCs remains unclear. Therefore, the purpose of this study is to treat KGN cells using testosterone propionate (TP) to simulate high androgen levels and induce cellular damage. This approach will help to elucidate the protective effects of afamin at the cellular level in ovarian GCs, thereby providing new insights and potential therapeutic strategies for treating ovarian diseases related to OS and other forms of cellular damage.

Materials and methods

The KGN human ovarian granulosa cell line used in this study was purchased from Wuhan Punosai Life Science and Technology Co., Ltd. The afamin protein was purchased from BIO-techne (USA), and fetal bovine serum was purchased from Clark (USA). Dulbecco’s Modified Eagle’s Medium/Nutritional Mixture F-12Ham, trypsin, dimethyl sulfoxide, and TriPure Isolation Reagent were purchased from Sigma (USA). Trypsin, dimethyl sulfoxide (DMSO), and TriPure Isolation Reagent were purchased from Sigma. Takara (Japan) supplied the RNA reverse transcription kit.

Cell culture and processing

The experiment used cells derived from GCs of patients with invasive ovarian cancer, specifically the human ovarian granulosa tumor cell line (KGN). KGN cells maintain the physiological properties of normal ovarian GCs and are widely used in various functional GC studies. KGN cells that had been preserved in liquid nitrogen were thawed by rapid shaking in warm water, transferred to sterile centrifuge tubes containing a serum-free medium, and then centrifuged at 1000 rpm for 7 min at room temperature. The culture flask was pre-loaded with DMEM containing 10% fetal bovine serum. After centrifugation, the supernatant was discarded, and the cells were resuspended by pipetting. Transfer 2 × 10 6 cells into the pre-loaded T75 culture flask. The flask was gently shaken horizontally and vertically to ensure proper mixing and then placed in a cell culture incubator at 37 °C with 5% CO2 for growth. The morphology and density of the cells were observed under a microscope. Once the cell density reached approximately 80% and the morphology was satisfactory, trypsin containing 0.05% EDTA was used to digest the cells, which were passaged. Then, seed the KGN cells into six-well plates at a density of 2.5 × 10 5 cells per well for subsequent experiments.

The experiments were randomly divided into three groups: control, TP, and Afamin + TP. Based on our previous experimental study and relevant literature [ 17 ], we selected 50 μmol/L TP and 50 ng/mL afamin protein for the experiment. Once the KGN cell density reached 80% or higher, the medium was replaced with DMEM containing 2% fetal bovine serum. Afamin protein at a concentration of 50 ng/mL was added to the Afamin + TP group, whereas the remaining two groups were incubated with DMEM containing 2% fetal bovine serum. After 4 h, KGN cells in the TP and Afamin + TP groups were treated separately with TP at a concentration of 50 μmol/L for 24 h. Subsequent experiments were then conducted.

ROS level detection

KGN cells were treated as previously described. Intracellular ROS levels were detected using a dichlorodihydrofluorescein diacetate (DCFH-DA) fluorescent probe from Beyotime Biotechnology (Shanghai, China). DCFH-DA was diluted with a serum-free medium to a final concentration of 10 μmol/L. The old medium in the six-well plate was discarded, and the cells were washed with PBS three times before aspirating the liquid. DCFH-DA (1 mL) was added to each well. After incubation at 37 °C in the dark, the working solution was discarded. The cells were washed three times with PBS and collected by trypsin digestion. After centrifugation, the supernatant was discarded, and the fluorescence intensity of the cells was measured by flow cytometry.

Superoxide anion level detection

Superoxide anion levels were measured using flow cytometry and dihydroethidium (DHE) fluorescent probe (Beyotime Biotechnology). The superoxide assay working solution was prepared, and the DHE probe was diluted to a final concentration of 10 μmol/L using serum-free medium at a ratio of 1:1000. The cells were washed three times with PBS, and the old medium was aspirated. Add 1 mL of DHE working solution to each well. The working solution was discarded at the end of the incubation period at 37 °C and was protected from light. The cells were washed with PBS buffer three times. The cells were collected by trypsin digestion and centrifuged, and the supernatant was discarded. The fluorescence intensity of each group of cells was determined by flow cytometry.

Malondialdehyde (MDA) level detection

MDA levels were determined using an MDA Assay Kit (Beyotime Biotechnology) to evaluate the presence of oxidative damage in the cells. The TBA storage solution and MDA assay working solution were prepared, the standards were diluted, and the reaction system and conditions for each step of the assay were established. The samples in each group were reacted, and the absorbance at 532 nm was measured using an enzyme marker. The MDA content of the samples was calculated based on protein concentrations.

Antioxidant system detection

This study measured the activities of four antioxidant enzymes: glutathione peroxidase (GSH-Px), glutathione reductase (GR), superoxide dismutase (SOD), and catalase (CAT). The corresponding kits from Beyotime Biotechnology were used to assay enzymes. After the appropriate treatment of the cells, the proteins in the samples were extracted. The protein concentration of each sample was determined using a BCA Protein Concentration Assay Kit (Beyotime Biotechnology). After treating the cells appropriately, the proteins were extracted, and their concentrations were determined using a BCA Protein Concentration Assay Kit (Beyotime Biotechnology). The instructions for each antioxidant enzyme kit were followed to measure the amount of each enzyme using an enzyme labeler.

Mitochondrial membrane potential (MMP) level detection

The cells were stained with JC-1 using a Mitochondrial Membrane Potential Assay Kit (Beyotime Biotechnology). JC-1 staining the working solution was prepared according to the manufacturer’s instructions. Then, the prepared staining working solution was added. The old cell culture solution in the six-well plate was discarded, and the cells were washed with PBS. Finally, the working solution was discarded at the end of the incubation at 37 °C and protected from light. The cells were washed with a pre-prepared JC-1 staining buffer, collected via trypsin digestion, and centrifuged. The supernatant was discarded, and the MMP levels of each group of cell samples were measured using a flow cytometer.

ATP level detection

ATP levels were measured using an ATP Assay Kit (Beyotime Biotechnology). The cell culture solution in the six-well plate was discarded, and 200 μL of the cell lysate was added to each well. After sufficient lysis, supernatants were collected for subsequent assays. The standard was diluted and an ATP assay working solution was prepared. To each well of a 96-well plate, 100 μL of ATP Assay Work Solution was added and allowed to sit at room temperature for 3 to 5 min to consume any background ATP and reduce background noise. Next, the standard and sample were added to each well, mixed quickly and thoroughly, and RLU values were determined using a chemiluminescence meter. Finally, ATP concentrations of the samples were calculated based on their respective protein concentrations.

Mitochondrial superoxide level detection

Mitochondrial superoxide levels were measured using MitoSOX Red mitochondrial superoxide indicator (Thermo Fisher, USA). To prepare a 5 mM MitoSOX storage solution, DMSO was added, and MitoSOX was diluted with serum-free culture solution at a ratio of 1:1000 to obtain a final concentration of 5 μM. The old culture was aspirated, and the cells were washed with PBS before adding 1 ml of diluted MitoSOX working solution to each well. After incubation at 37 °C, the working solution was discarded, and the cells were washed with PBS. Next, the cells were collected by trypsin digestion and the supernatant was discarded after centrifugation. Finally, the fluorescence intensity of each group of cells was determined by flow cytometry.

Apoptosis flow cytometry detection

Apoptosis was detected using the Annexin V-FITC Apoptosis Detection Kit (Beyotime Biotechnology). The previous cell culture was collected in a 15 ml centrifuge tube, washed with PBS buffer, trypsin-digested, and the cell suspension was mixed with the previous culture in the centrifuge tube. The cells from the old culture were collected in a 15 mL centrifuge tube and washed with PBS buffer. The collected cells were trypsin-digested and the resulting cell suspension was mixed with the old culture in a centrifuge tube. After centrifugation, the supernatant was discarded, and the cells were washed again with PBS. Add 195 μL of Annexin V-FITC conjugate and 5 μL of Annexin V-FITC dye. Mix well and add 10 μL of propidium iodide staining solution. The assay was completed within 1 h after incubation, protected from light.

Caspase-3 activity detection

Caspase-3 activity was measured using the Caspase-3 Activity Assay Kit (Beyotime Biotechnology). Standard dilutions were prepared, the pNA standards were diluted, and A405 was measured using an enzyme labeler. The results are plotted on a standard curve. The cell culture was collected, the adherent cells were digested with trypsin, transferred to the same centrifuge tube as the collected culture, and the supernatant was discarded after centrifugation. After washing, lysis solution was added, and the cells were lysed in an ice bath for 15 min. The resulting mixture was centrifuged to collect the supernatant in a pre-cooled EP tube. A portion of the supernatant was collected to determine protein concentration. Next, each well of a 96-well plate was sequentially added to the assay buffer, the samples to be tested, and Ac-DEVD-pNA (2 mM). The mixture was thoroughly mixed and incubated at 37 °C. The absorbance at 405 nm was measured using an enzyme labeler after a significant change in color was observed. Caspase-3 activity was calculated based on the protein concentration in each sample.

Steroids hormone secretion level detection

The levels of E 2 and P 4 secretion in the cells were determined using human estrogen and progesterone ELISA kits (Cusabio Technology LLC, Houston, TX, USA). Once the cell density reached 70–80%, the medium was changed to serum-free medium, and the cells were treated according to the above grouping. The culture supernatant of each group of cells was collected, centrifuged, and stored at -80 °C. The levels of E 2 and P 4 secreted in the supernatant were measured using ELISA, following the manufacturer’s instructions.

Real-time polymerase chain reaction (PCR)

RNA was extracted from KNG cells using an RNA extraction kit and reverse-transcribed to cDNA using an RNA reverse transcription kit. Gene expression levels were analyzed by RT-qPCR using FS Universal SYBR Green Real Master (Roche) and the 2 −ΔΔCt method, based on the sample cycling reaction threshold (Ct). The primer sequences used are listed in Table  1 .

Statistical analysis

Data were analyzed and plotted using SPSS software (version 25.0) and GraphPad Prism 9.4.0 software. Significant differences were analyzed using one-way analysis of variance (ANOVA), and the data were tested for normality using chi-square tests. The mean ± standard deviation (Mean ± SD) was used to express the data, and differences were considered statistically significant when P  < 0.05.

Afamin protein ameliorates TP-induced oxidative stress levels in KGN cells

The DCFH-DA fluorescent probe could freely pass through the cell membrane. Once inside, it is oxidized by intracellular ROS to fluorescent DCF. The average fluorescence intensity of DCF was used to analyze ROS levels. As shown in (Fig.  1 A-B), the levels of ROS were significantly higher in the TP group than in the control group, suggesting that TP-induced OS in the KGN cells. Compared with the TP group, afamin protein significantly reduced the expression of ROS in KGN cells, protecting them from OS-induced damage.

DHE, one of the most commonly used fluorescent probes, was employed to detect superoxide anions. Upon dehydrogenation, the red fluorescence of the probe indicates the presence of superoxide anions. Therefore, the levels of superoxide anions in the KGN cells were examined using a DHE fluorescent probe. The data presented in (Fig.  1 C-D) indicate that the level of superoxide anions was significantly increased in the TP-treated group compared to the control group, suggesting the occurrence of OS. However, the addition of afamin resulted in a significant reduction in superoxide anion levels in KGN cells compared to that in the TP group, indicating that afamin mitigated the oxidative damage caused by TP in KGN cells.

ROS induces lipid peroxidation in vivo and in vitro, leading to an increase in MDA levels. MDA is a product of lipid peroxidation and an indicator of cellular oxidative damage. MDA levels were significantly higher in the TP-treated group than in the control group (Fig.  1 E). However, the addition of afamin protein to KGN cells resulted in a significant decrease in MDA levels compared to those in the TP group. It has been suggested that afamin proteins may have damaging antioxidant effects.

Afamin protein ameliorates TP-induced oxidative stress in KGN cells. ( A - B ) Afamin reduced TP-induced ROS levels in KGN cells. ( C - D ) Afamin reduced superoxide anion levels in TP-induced KGN cells. ( E ) Afamin reduces TP-induced MDA levels in KGN cells. The experiment was repeated for three times, results are expressed as mean ± standard deviation, * P  < 0.05

OS occurs due to the limited ability of antioxidants to scavenge excess ROS, resulting in diminished antioxidant capacity. Four antioxidant enzymes, GSH-Px, GR, SOD, and CAT, were assayed using relevant antioxidant assay kits. Our data revealed that the activities of four antioxidant enzymes (GSH-Px, GR, SOD, and CAT) decreased in the KGN cells of the TP group compared to those in the control group, indicating that TP-induced OS in KGN cells. The addition of afamin significantly elevated the activities of the four antioxidant enzymes (SOD, CAT, GSH-Px, and GR) in KGN cells compared to those in the TP group(Fig.  2 A-D). These findings indicated that afamin proteins may mitigate OS damage in KGN cells by modulating the expression of antioxidant enzymes.

Afamin protects KGN cells from oxidative stress damage by increasing the activity of antioxidant enzymes. ( A ) GSH-Px enzymatic activity in KGN cells. ( B ) GR enzyme activity in KGN cells. ( C ) SOD activity in KGN cells. ( D ) CAT activity in KGN cells. The experiment was repeated for three times, results are expressed as mean ± standard deviation, * P  < 0.05

Afamin protein ameliorates TP-induced mitochondrial damage in KGN cells

MMP levels were measured using flow cytometry. The results indicate that the MMP level was significantly decreased in the TP group compared to that in the control group, suggesting that the mitochondrial function of KGN cells in the TP group may be impaired. The inclusion of afamin significantly increased MMP expression in KGN cells compared to that in the TP group (Fig.  3 A-B), indicating that afamin improved mitochondrial membrane potential in KGN cells.

Mitochondria are crucial intracellular organelles involved in ATP synthesis. ATP levels in KGN cells were measured using an ATP assay kit. The results indicate that ATP levels decreased in the TP-treated group, whereas ATP levels increased significantly with the addition of afamin protein (Fig.  3 C). Previous studies have suggested that TP leads to mitochondrial dysfunction in KGN cells and that afamin ameliorates this impairment.

Mitochondrial superoxide levels were measured using MitoSOX Red mitochondrial superoxide indicator and analyzed by flow cytometry. The data presented in (Fig.  3 D-E) indicate that intracellular mitochondrial superoxide levels increased after TP treatment, suggesting impaired mitochondrial function. Treatment with Afamin protein reduced the elevated mitochondrial superoxide levels compared to those in the TP group, indicating the effectiveness of afamin protein in restoring mitochondrial function in KGN cells.

Afamin protein attenuates TP-induced mitochondrial damage in KGN cells. ( A - B ) Afamin reduced TP-induced MMP damage in KGN cells. ( C ) Afamin increased ATP levels in TP-treated KGN cells. ( D - E ) Afamin reduces TP-induced mitochondrial superoxide levels in KGN cells. The experiment was repeated for three times, results are expressed as mean ± standard deviation, * P  < 0.05

Afamin protein ameliorates TP-induced apoptosis in KGN cells

Mitochondrial dysfunction induces apoptosis. Flow cytometry was used to detect apoptosis in KGN cells. The results showed a significant increase in the apoptosis of KGN cells after exposure to TP. However, afamin protein treatment ameliorated TP-induced apoptosis(Fig.  4 A).

The activity of Caspase-3, an important factor in apoptosis, was analyzed to confirm these results. The results indicate that Caspase-3 activity increased in the TP group and decreased in the afamin group(Fig.  4 B), indicating that afamin protein ameliorated TP-promoted apoptosis.

To investigate the mechanisms of apoptosis in KGN cells, we analyzed the expression of the pro-apoptotic gene BAX and anti-apoptotic gene BCL-2 using fluorescence quantitative PCR. The results in (Fig.  4 C-D) indicate that exposure to TP led to a significant increase in the expression of the pro-apoptotic BAX gene and a significant decrease in the expression of the anti-apoptotic BCL-2 gene in KGN cells. Conversely, pretreatment with afamin resulted in a decrease in BCL-2-associated X protein (BAX) expression and increased BCL-2 expression. These findings are objective and are based solely on the data presented. Afamin appears to ameliorate TP-induced apoptosis in KGN cells by regulating BAX/BCL-2 and acting as an inhibitor of apoptosis.

Afamin protein inhibits TP-induced apoptosis in KGN cells. ( A ) Flow cytometry analysis suggests that afamin can reduce TP-induced apoptosis in KGN cells. ( B ) Afamin decreased caspase-3 activity in TP-treated KGN cells. ( C - D ) Afamin protein reduced TP-induced apoptosis in KGN cells by increasing BCL-2 mRNA expression and decreasing BAX mRNA expression. The experiment was repeated for three times, results are expressed as mean ± standard deviation, * P  < 0.05

Afamin protein ameliorates TP-induced abnormal steroid hormone secretion in KGN cells

Excessive ovarian stimulation leads to impaired function of GCs, which are the main sources of steroidal estrogens (E 2 ) and progesterone (P 4 ). Therefore, we determined the levels of E 2 and P 4 secreted by KGN cells into the supernatant using ELISA. The results showed abnormal levels of E 2 and P 4 hormone secretion in the TP group compared to those in the control group. However, the addition of afamin restored abnormal levels of E 2 and P 4 secretion in KGN cells (Fig.  5 A-B). To investigate the mechanism of the effect of afamin protein on steroid hormone synthesis in TP-treated KGN cells, we detected the mRNA expression of the hormone synthesis-related genes CYP19A1, CYP11A1, steroidogenic acute regulatory (STAR), and HSD3B1 using fluorescence quantitative PCR.

The results in (Fig.  5 C-F) indicate a significant decrease in the mRNA expression of STAR, HSD3B1, and CYP11A1, and an increase in CYP19A1 mRNA expression in the TP group. In contrast, KGN cells treated with Afamin protein exhibited elevated STAR, HSD3B1, and CYP11A1 mRNA expression levels, along with reduced CYP19A1 mRNA expression. It has been suggested that afamin may regulate the expression of genes related to steroid hormone synthesis, ameliorating the abnormal secretory function observed in TP-treated KGN cells.

Afamin protects TP-induced secretory function of KGN cells. ( A - B ) Afamin ameliorates the abnormal secretion of E 2 and P 4 in TP-treated KGN cells. ( C - F ) Afamin repairs the secretory function of KGN cells by regulating the mRNA expression of CYP19A1, CYP11A1, STAR, and HSD3B1. The experiment was repeated for three times, results are expressed as mean ± standard deviation, * P  < 0.05

GCs play crucial roles in follicle growth and development. They provide nutrients and maturation factors to maintain oocyte maturation and protect oocytes from oxidative damage [ 18 , 19 ]. In recent years, damage caused by OS, which is characterized by the excessive production of ROS, has garnered significant attention from researchers. OS is considered a potential inducing factor in the pathogenesis of PCOS. For patients with PCOS, elevated androgen levels and insulin resistance could exacerbate oxidative stress, adversely affecting the function of granulosa cells (GCs) and the normal growth and development of follicles [ 20 , 21 ]. ROS, an important marker of OS, can lead to DNA damage in GCs, affecting female oocyte maturation and follicle growth and development [ 22 , 23 ]. Excessive ROS production can induce lipid peroxidation both in vivo and in vitro. This process commonly results in the formation of malondialdehyde (MDA) [ 24 , 25 ], which is a harmful stimulus that can cause protein misfolding and exacerbate the OS response [ 26 ]. Studies have shown that patients with PCOS typically exhibit elevated levels of MDA, indicating significant lipid peroxidation and oxidative damage [ 27 ]. Furthermore, the body relies on four antioxidant enzymes (SOD, CAT, GSH-Px, and GR) to defend against OS by neutralizing excess ROS and preventing damage to cellular structures [ 28 , 29 ]. SOD serves as the first line of defense in the antioxidant defense system by converting superoxide anions into hydrogen peroxide, thereby reducing levels of ROS. In PCOS, SOD activity is commonly reduced, indicating a weakened antioxidant defense capability. CAT further decomposes hydrogen peroxide into water and oxygen, thus reducing its toxicity to cells. Reduced CAT activity leads to the accumulation of hydrogen peroxide, resulting in cellular damage and dysfunction. GSH-Px reduces lipid peroxides and hydrogen peroxide, protecting cells from oxidative damage. In PCOS, decreased GSH-Px levels indicate compromised antioxidant capacity, exacerbating cellular oxidative damage. Additionally, GR maintains intracellular glutathione levels by reducing oxidized glutathione to its reduced form, thereby enhancing cellular antioxidant capacity. Reduced GR activity weakens cellular antioxidant defenses, increasing the risk of cellular damage [ 8 , 30 , 31 ]. However, it is well known that vitamin E is an important antioxidant. Afamin, as a specific binding protein for vitamin E, can take over the role of transporting vitamin E in body fluids when the lipoprotein system is inadequate [ 13 ]. In in vitro experiments, researchers have discovered that afamin protects cortical neurons from apoptosis and acts as an antioxidant against neuronal OS [ 16 ]. In this study, exposure to TP led to elevated levels of the OS biomarkers ROS, superoxide anions, and MDA in KGN cells. Furthermore, the levels of four antioxidant enzymes (SOD, CAT, GSH-Px, and GR) decreased. However, afamin protein could downregulate the levels of ROS, superoxide anions, and MDA while increasing the activities of the antioxidant enzymes GSH-Px, GR, SOD, and CAT. Therefore, afamin protects KGN cells from oxidative damage by reducing OS and increasing their antioxidant capacity.

Mitochondria are highly susceptible to oxidative damage due to their proximity to ROS-generating sources, lack of histone and DNA repair mechanisms, and presence of a circular genome known as mitochondrial DNA (mtDNA) between the inner and outer mitochondrial membranes [ 32 ]. When ROS are produced in large quantities, they can damage mitochondria through free radical reactions, resulting in impaired mitochondrial function. This impairment is mainly characterized by alterations in MMP. MMP deficiency can cause defects in the mitochondrial electron transport chain, leading to an increase in ATP consumption and a decrease in energy metabolism [ 33 , 34 ]. This study demonstrates that mitochondrial function in KGN cells may be impaired after TP treatment. Afamin supplementation resulted in a significant increase in MMP and ATP levels as well as a significant reduction in mitochondrial superoxide levels, indicating that afamin may improve mitochondrial dysfunction in KGN cells. Furthermore, oxidative damage to mitochondria can induce mitochondrial outer membrane permeabilization (MOMP), resulting in a reduction in the mitochondrial transmembrane potential. This opens the mitochondrial membrane permeability transition pore (mPTP), leading to the release of cytochrome c from the mitochondria into the cytoplasm. Cytochrome c binds to apoptotic protease activating factor 1 (APAf-1) to form the “apoptosome” complex, which sequentially activates a series of cysteine caspases, leading to apoptosis [ 34 , 35 ]. BAX is a key regulator of mitochondrial apoptosis. Although it physiologically maintains tissue homeostasis, dysregulation of BAX can cause aberrant cell death [ 36 ]. However, BCL-2 prevents MOMP by maintaining mitochondrial integrity, which prevents cytochrome c release. BCL-2 overexpression inhibits cell apoptosis [ 37 ]. The study results indicated that exposure to TP significantly increased apoptosis of KGN cells, elevated BAX expression, decreased BCL-2 expression, and significantly increased caspase-3 activity. The addition of afamin reversed these effects, suggesting that afamin can ameliorate apoptosis in KGN cells.

GCs play a crucial role in steroid hormone synthesis. Oxidative damage to GCs can lead to disorders of steroid hormone secretion in vivo [ 38 , 39 ]. Steroid hormone synthesis involves several enzymes and proteins. Changes in these key enzymes can affect the production of steroid hormones [ 40 ]. The STAR facilitates cholesterol transfer from the outer mitochondrial membrane to the inner mitochondrial membrane. This process is the rate-limiting step in steroid hormone formation. The P450 cholesterol side-chain cleavage enzyme (CYP11A1) is situated on the stromal side of the inner mitochondrial membrane. It can convert cholesterol to pregnenolone. After leaving the mitochondrion, pregnenolone is converted to progesterone by the 3β-hydroxysteroid dehydrogenase (HSD3B1) in the mitochondrial compartment [ 41 , 42 ]. Furthermore, the conversion of androgens to estrogens is primarily conducted by cytochrome P450 aromatase (CYP19A1), which is widely expressed in the ovarian GCs. This enzyme is a crucial component of estrogen synthesis [ 43 ]. In this study, KGN cells exposed to TP and damaged by OS exhibited abnormal secretion of E 2 and P 4 , and abnormal expression of genes related to steroid hormone synthesis. Oxidative damage to KGN cells may affect the expression of genes involved in steroid hormone synthesis, leading to abnormal synthesis. However, Afamin protein reduced TP-induced OS damage in KGN cells, reversed the expression levels of steroid hormone synthase genes, and improved the secretory function of KGN cells. These results indicate that afamin protein inhibits intracellular ROS levels, increases antioxidant enzyme activities in KGN cells, and protects against OS and related injuries in human ovarian GCs.

Conclusions

In summary, the findings of this study demonstrate that the protective effect of Afamin protein against TP-induced damage in KGN cells is achieved by enhancing the antioxidant capacity of these cells. This enhancement helps to prevent oxidative stress (OS), reduce mitochondrial damage and apoptosis, and improve hormone secretion functions in KGN cells(Fig.  6 ). This study presents a theoretical foundation for further investigation into the function and regulatory mechanism of Afamin protein, as well as a potential molecular basis for the early treatment of ovarian reproductive endocrine disorders, such as PCOS, in women.

Afamin protects against TP-induced oxidative stress-related damage in KGN cells

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Abbreviations

Adenosine triphosphate

BCL2-Associated X

B-cell lymphoma-2

Cytochrome P450 Family 11 Subfamily A Member 1

Cytochrome P450 Family 19 Subfamily A Member 1

Glutathione reductase

Glutathione peroxidase

Hydroxy-delta-5-steroid dehydrogenase, 3beta- and steroid delta-isomerase 1

Malondialdehyde

Reactive oxygen species

Superoxide dismutase

Steroidogenic acute regulatory protein

Jozkowiak M, Piotrowska-Kempisty H, Kobylarek D, Gorska N, Mozdziak P, Kempisty B, Rachon D, Spaczynski RZ. Endocrine disrupting chemicals in polycystic ovary syndrome: the relevant role of the theca and granulosa cells in the pathogenesis of the ovarian dysfunction. Cells. 2022;12(1). https://doi.org/10.3390/cells12010174

Havelock JC, Rainey WE, Carr BR. Ovarian granulosa cell lines. Mol Cell Endocrinol. 2004;228(1–2):67–78. https://doi.org/10.1016/j.mce.2004.04.018

Article   CAS   PubMed   Google Scholar  

Herlands RL, Schultz RM. Regulation of mouse oocyte growth: probable nutritional role for intercellular communication between follicle cells and oocytes in oocyte growth. J Exp Zool. 1984;229(2):317–25. https://doi.org/10.1002/jez.1402290217

Acuña-Hernández DG, Arreola-Mendoza L, Santacruz-Márquez R, García-Zepeda SP, Parra-Forero LY, Olivares-Reyes JA, Hernández-Ochoa I. Bisphenol A alters oocyte maturation by prematurely closing gap junctions in the cumulus cell-oocyte complex. Toxicol Appl Pharmacol. 2018;344:13–22. https://doi.org/10.1016/j.taap.2018.02.011

Rodriguez Paris V, Bertoldo MJ. The mechanism of androgen actions in PCOS etiology. Med Sci (Basel). 2019;7(9). https://doi.org/10.3390/medsci7090089

McFee RM, Romereim SM, Snider AP, Summers AF, Pohlmeier WE, Kurz SG, Cushman RA, Davis JS, Wood JR, Cupp AS. A high-androgen microenvironment inhibits granulosa cell proliferation and alters cell identity. Mol Cell Endocrinol. 2021;531:111288. https://doi.org/10.1016/j.mce.2021.111288

Pizzino G, Irrera N, Cucinotta M, Pallio G, Mannino F, Arcoraci V, Squadrito F, Altavilla D, Bitto A. Oxidative stress: harms and benefits for human health. Oxid Med Cell Longev. 2017;2017:8416763 https://doi.org/10.1155/2017/8416763

Article   CAS   Google Scholar  

Zeber-Lubecka N, Ciebiera M, Hennig EE. Polycystic ovary syndrome and oxidative stress-from bench to bedside. Int J Mol Sci. 2023;24(18). https://doi.org/10.3390/ijms241814126

Jones DP. Redefining oxidative stress. Antioxid Redox Signal. 2006;8(9–10):1865–79. https://doi.org/10.1089/ars.2006.8.1865

Luderer U. Ovarian toxicity from reactive oxygen species. Vitam Horm. 2014;94:99–127. https://doi.org/10.1016/B978-0-12-800095-3.00004-3

Liu S, Jia Y, Meng S, Luo Y, Yang Q, Pan Z. Mechanisms of and potential medications for oxidative stress in ovarian granulosa cells: a review. Int J Mol Sci. 2023;24(11). https://doi.org/10.3390/ijms24119205

Lichenstein HS, Lyons DE, Wurfel MM, Johnson DA, McGinley MD, Leidli JC, Trollinger DB, Mayer JP, Wright SD, Zukowski MM. Afamin is a new member of the albumin, alpha-fetoprotein, and vitamin D-binding protein gene family. J Biol Chem. 1994;269(27):18149–54.

Voegele AF, Jerković L, Wellenzohn B, Eller P, Kronenberg F, Liedl KR, Dieplinger H. Characterization of the vitamin E-binding properties of human plasma afamin. Biochemistry. 2002;41(49):14532–8. https://doi.org/10.1021/bi026513v

Dieplinger H, Dieplinger B. Afamin–A pleiotropic glycoprotein involved in various disease states. Clin Chim Acta. 2015;446:105–10. https://doi.org/10.1016/j.cca.2015.04.010

Azzi A. Oxidative stress: what is it? Can it be measured? Where is it located? Can it be good or bad? Can it be prevented? Can it be cured? Antioxid (Basel). 2022;11(8). https://doi.org/10.3390/antiox11081431

Heiser M, Hutter-Paier B, Jerkovic L, Pfragner R, Windisch M, Becker-André M, Dieplinger H. Vitamin E binding protein afamin protects neuronal cells in vitro. J Neural Transm Suppl. 2002(62):337–45. https://doi.org/10.1007/978-3-7091-6139-5_32

Jin J, Ma Y, Tong X, Yang W, Dai Y, Pan Y, Ren P, Liu L, Fan HY, Zhang Y, et al. Metformin inhibits testosterone-induced endoplasmic reticulum stress in ovarian granulosa cells via inactivation of p38 MAPK. Hum Reprod. 2020;35(5):1145–58. https://doi.org/10.1093/humrep/deaa077

Article   CAS   PubMed   PubMed Central   Google Scholar  

Kranc W, Budna J, Kahan R, Chachuła A, Bryja A, Ciesiółka S, Borys S, Antosik MP, Bukowska D, Brussow KP, et al. Molecular basis of growth, proliferation, and differentiation of mammalian follicular granulosa cells. J Biol Regul Homeost Agents. 2017;31(1):1–8.

CAS   PubMed   Google Scholar  

Rimon-Dahari N, Yerushalmi-Heinemann L, Alyagor L, Dekel N. Ovarian folliculogenesis. Results Probl Cell Differ. 2016;58:167–90. https://doi.org/10.1007/978-3-319-31973-5_7

Zeng X, Xie YJ, Liu YT, Long SL, Mo ZC. Polycystic ovarian syndrome: correlation between hyperandrogenism, insulin resistance and obesity. Clin Chim Acta. 2020;502:214–21. https://doi.org/10.1016/j.cca.2019.11.003

Kaufman MJ, Kanayama G, Hudson JI, Pope HG Jr. Supraphysiologic-dose anabolic-androgenic steroid use: a risk factor for dementia? Neurosci Biobehav Rev. 2019;100:180–207. https://doi.org/10.1016/j.neubiorev

Mello Filho AC, Hoffmann ME, Meneghini R. Cell killing and DNA damage by hydrogen peroxide are mediated by intracellular iron. Biochem J. 1984;218(1):273–5. https://doi.org/10.1042/bj2180273

Agarwal A, Gupta S, Sharma RK. Role of oxidative stress in female reproduction. Reprod Biol Endocrinol. 2005;3:28. https://doi.org/10.1186/1477-7827-3-28

Liu S, Navarro G, Mauvais-Jarvis F. Androgen excess produces systemic oxidative stress and predisposes to beta-cell failure in female mice. PLoS ONE. 2010;5(6):e11302. https://doi.org/10.1371/journal

Article   PubMed   PubMed Central   Google Scholar  

Sousa BC, Pitt AR, Spickett CM. Chemistry and analysis of HNE and other prominent carbonyl-containing lipid oxidation compounds. Free Radic Biol Med. 2017;111:294–308. https://doi.org/10.1016/j.freeradbiomed.2017.02.003

Wang L, Zhang M, Fang Z, Bhandari B. Gelation properties of myofibrillar protein under malondialdehyde-induced oxidative stress. J Sci Food Agric. 2017;97(1):50–7. https://doi.org/10.1002/jsfa

Tan W, Dai F, Yang D, Deng Z, Gu R, Zhao X, Cheng Y. MiR-93-5p promotes granulosa cell apoptosis and ferroptosis by the NF-kB signaling pathway in polycystic ovary syndrome. Front Immunol. 2022;13:967151. https://doi.org/10.3389/fimmu

Lv H, Liu Q, Wen Z, Feng H, Deng X, Ci X. Xanthohumol ameliorates lipopolysaccharide (LPS)-induced acute lung injury via induction of AMPK/GSK3β-Nrf2 signal axis. Redox Biol. 2017;12:311–24. https://doi.org/10.1016/j.redox.2017.03.001

Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M. Free radicals, metals and antioxidants in oxidative stress-induced cancer. Chem Biol Interact. 2006;160(1):1–40. https://doi.org/10.1016/j.cbi.2005.12.009

Ma Y, Zheng L, Wang Y, Gao Y, Xu Y. Arachidonic acid in follicular fluid of PCOS induces oxidative stress in a human ovarian granulosa tumor cell line (KGN) and upregulates GDF15 expression as a response. Front Endocrinol (Lausanne). 2022;13:865748. https://doi.org/10.3389/fendo

Article   PubMed   Google Scholar  

Abudawood M, Tabassum H, Alanazi AH, Almusallam F, Aljaser F, Ali MN, Alenzi ND, Alanazi ST, Alghamdi MA, Altoum GH, et al. Antioxidant status in relation to heavy metals induced oxidative stress in patients with polycystic ovarian syndrome (PCOS). Sci Rep. 2021;11(1):22935 https://doi.org/10.1038/s41598-021-02120-6

Mustafa MF, Fakurazi S, Abdullah MA, Maniam S. Pathogenic mitochondria DNA mutations: current detection tools and interventions. Genes (Basel). 2020;11(2). https://doi.org/10.3390/genes11020192

Zeng X, Huang Q, Long SL, Zhong Q, Mo Z. Mitochondrial dysfunction in polycystic ovary syndrome. DNA Cell Biol. 2020;39(8):1401–9. https://doi.org/10.1089/dna.2019.5172

Kim SH, Kim H. Inhibitory effect of astaxanthin on oxidative stress-induced mitochondrial dysfunction-a mini-review. Nutrients. 2018;10(9). https://doi.org/10.3390/nu10091137

Caroppi P, Sinibaldi F, Fiorucci L, Santucci R. Apoptosis and human diseases: mitochondrion damage and lethal role of released cytochrome C as proapoptotic protein. Curr Med Chem. 2009;16(31):4058–65. https://doi.org/10.2174/092986709789378206

Spitz AZ, Gavathiotis E. Physiological and pharmacological modulation of BAX. Trends Pharmacol Sci. 2022;43(3):206–20. https://doi.org/10.1016/j.tips.2021.11.001

Green DR. At the gates of death. Cancer Cell. 2006;9(5):328–30. https://doi.org/10.1016/j.ccr.2006.05.004

Jozkowiak M, Hutchings G, Jankowski M, Kulcenty K, Mozdziak P, Kempisty B, Spaczynski RZ, Piotrowska-Kempisty H. The stemness of human ovarian granulosa cells and the role of resveratrol in the differentiation of MSCs-a review based on cellular and molecular knowledge. Cells. 2020;9(6). https://doi.org/10.3390/cells9061418

Karabulut S, Korkmaz O, Kutlu P, Gozel HE, Keskin I. Effects o follicular fluid oxidative status on human mural granulosa cells, oocyte competency and ICSI parameters. Eur J Obstet Gynecol Reprod Biol. 2020;252:127–36. https://doi.org/10.1016/j.ejogrb.2020.06.021

Payne AH, Hales DB. Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocr Rev. 2004;25(6):947–70. https://doi.org/10.1210/er.2003-0030

Stocco DM. StAR protein and the regulation of steroid hormone biosynthesis. Annu Rev Physiol. 2001;63:193–213. https://doi.org/10.1146/annurev.physiol.63.1.193

Miller WL, Auchus RJ. The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders. Endocr Rev. 2011;32(1):81–151. https://doi.org/10.1210/er.2010-0013

Zhang Y, Chen X, Zhou Z, Tian X, Yang P, Fu K. CYP19A1 may influence lambing traits in goats by regulating the biological function of granulosa cells. Anim (Basel). 2022;12(15). https://doi.org/10.3390/ani12151911

Download references

Acknowledgements

Not applicable.

This study was supported by the Natural Science Foundation of Jilin Province (No. YDZJ202301ZYTS434).

Author information

Authors and affiliations.

Reproductive Medical Center, The Second Hospital of Jilin University, 4026 Yatai Street, Changchun, 130022, China

Qiang Zhang, Xueying Zhang & Lianwen Zheng

Department of Gynecology, Dongguan Songshan Lake Tungwah Hospital, NO.1 Songshan Lake Science Development Seven Road, DongGuan, 523822, China

Xiaoyu Zheng

You can also search for this author in PubMed   Google Scholar

Contributions

Qiang Zhang reviewed the literature and statistical analyses of the data and wrote the manuscript. Xiaoyu Zheng and Xueying Zhang contributed to the conception and design of the study, and Professor Lianwen Zheng supervised and revised the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Lianwen Zheng .

Ethics declarations

Ethics approval and consent to participate, consent for publication.

The manuscript has been approved for publication by all authors.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Zhang, Q., Zheng, X., Zhang, X. et al. Protective effect of afamin protein against oxidative stress related injury in human ovarian granulosa cells. J Ovarian Res 17 , 189 (2024). https://doi.org/10.1186/s13048-024-01511-3

Download citation

Received : 26 June 2024

Accepted : 05 September 2024

Published : 28 September 2024

DOI : https://doi.org/10.1186/s13048-024-01511-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Human ovarian tumor granulosa cell line(KGN)
• Oxidative stress(OS)
• Mitochondria

Journal of Ovarian Research

ISSN: 1757-2215

• General enquiries: [email protected]

• Open access
• Published: 30 September 2024

HIF-1α-HPRT1 axis promotes tumorigenesis and gefitinib resistance by enhancing purine metabolism in EGFR-mutant lung adenocarcinoma

• Pengyu Geng 1 , 2 ,
• Peng Dou 1 , 2 ,
• Chunxiu Hu 1 , 2 ,
• Jiarui He 4 ,
• Jinhui Zhao 1 , 2 ,
• Qi Li 1 , 2 ,
• Miao Bao 5 ,
• Xiangnan Li 3 ,
• Xinyu Liu 1 , 2 &
• Guowang Xu 1 , 2  

Journal of Experimental & Clinical Cancer Research volume  43 , Article number:  269 ( 2024 ) Cite this article

Metrics details

The mutations of oncogenic epidermal growth factor receptor (EGFR) is an important cause of lung adenocarcinoma (LUAD) malignance. It has been knowm that metabolic reprogramming is an important hallmark of malignant tumors, and purine metabolism is a key metabolic pathway for tumor progression and drug resistance, but its relationship with the EGFR-mutant LUAD is unclear.

Metabolic reprogramming was studied through capillary electrophoresis-time of flight mass spectrometry (CE-TOF/MS)-based metabolic profiling analysis. Cell proliferation in vitro was evaluated by EdU staining and cell cycle assay. Tumorigenicity in vivo was tested by subcutaneous tumor formation experiment in nude mice. The binding of hypoxia-inducible factor-1 alpha (HIF-1α) and hypoxanthine phosphoribosyltransferase 1 (HPRT1) was detected by DNA pull‑down assay and Chromatin immunoprecipitation (ChIP) assays. HIF-1α, HPRT1, DNA damage and cell apoptosis related genes were examined by western blot. In addition, RNA sequencing, mass spectrometry and bioinformatics analysis were performed.

We found that mutated EGFR (muEGFR) upregulates HPRT1 to promote purine metabolism and tumorigenesis of EGFR-mutant LUAD. Mechanistically, muEGFR increases HIF-1α expression through protein stability. Meanwhile, up-regulated HIF-1α bound to the promoter of HPRT1 and transcriptionally activates HPRT1 expression, enhancing purine metabolism to maintain rapid tumor cell proliferation in EGFR-mutant LUAD. Further, gefitinib inhibited the synthesis of purine nucleotides, and HPRT1 inhibition increased the sensitivity of gefitinib to EGFR-mutant LUAD.

Conclusions

Our study reveals that muEGFR-HIF-1α-HPRT1 axis plays a key role in EGFR-mutant LUAD and provides a new strategy-inhibiting purine metabolism for treating EGFR-mutant LUAD.

Mutations in the epidermal growth factor receptor (EGFR) are a pivotal determinant in the etiology of lung adenocarcinoma (LUAD) [ 1 ]. Currently, EGFR-tyrosine kinase inhibitors (TKIs), such as gefitinib and osimertinib, are the mainstay of first-line therapy for advanced lung cancer with EGFR mutations [ 2 ]. Despite their initial high response rates, the development of resistance to these EGFR-TKIs is an unfortunate eventuality that can significantly impair patient outcomes over time [ 3 , 4 ].

EGFR-mutant LUAD experiences significant alterations in a variety of metabolic pathways, including key metabolic processes such as glycolysis, pentose phosphate pathway, glutathione metabolism and lipid metabolism [ 5 , 6 , 7 , 8 ]. Metabolic reprogramming, as a defined hallmark of malignant tumor, plays a crucial role in tumor biology [ 9 ].Notably, abnormal purine metabolism is intimately linked to tumor progression [ 10 ]. Purine nucleotides not only play essential roles in DNA and RNA biosynthesis, but also furnish the necessary energy and cofactor molecules that are critical for sustaining cell survival and driving cell proliferation [ 11 ]. Multiple enzymes in purine metabolism are disregulated, which is related to the enhanced proliferation and development of drug resistance in tumor cells [ 12 ]. Hypoxanthine phosphoribosyltransferase 1 (HPRT1) is a key enzyme in the salvage pathway of purine nucleotides, facilitating the conversion of hypoxanthine and guanine to their respective mononucleotides [ 13 ]. This enzyme is highly expressed in various cancers, and is implicated in the dynamics of cancer progression and the acquisition of drug resistance [ 14 , 15 , 16 ]. In small cell lung cancer (SCLC), HPRT1 promotes cell proliferation by enhancing salvage purine synthesis metabolism in glutamine starvation condition [ 17 ]. However, the specific mechanism of mutated EGFR (muEGFR) regulated the HPRT1 to reprogram purine metabolism in EGFR-mutant LUAD remains unexplored and has not yet been reported.

In this study, a critical role for HPRT1-mediated purine metabolism in promoting cell proliferation and tumorigenesis in EGFR-mutant LUAD were uncovered. Moreover, the regulatory mechanism and function of hypoxia-inducible factor-1 alpha (HIF-1α) on HPRT1 were clarified. Finally, the effect of targeting purine metabolism on the therapeutic efficacy of EGFR-TKIs was explored.

Materials and methods

Cell culture.

EGFR wild type lung adenocarcinoma cell (H1299) and EGFR mutant type lung adenocarcinoma cells (PC9, H3255 and H1975) were purchased from the cell bank of the Committee on Type Culture Collection of the Chinese Academy of Sciences (CTCC). H1299, PC9, H3255 and H1975 cell lines were cultured in RPMI 1640 medium (Gibco) supplemented with 10% FBS (Meilun) and 1% penicillin/streptomycin (Meilun) at 37℃ with 5% CO2 in a humidified incubator (Thermo).

Plasmids construction and viral infection

The HPRT1 overexpression plasmid was constructed in the pCDH-puro vector, and the HPRT1 and HIF-1α short hairpin RNA plasmids were constructed in the pLKO.1-puro backbone. The viruses were produced from 293T cells with lentivirus packaging vectors (PRRE, VSVG and REV) and then infected into target cell lines following manufacturer’s instructions. The sequences of shRNAs are given below:

shHPRT1#1: CCAGGTTATGACCTTGATTTA

shHPRT1#2: GCACTGAATAGAAATAGTGAT

shHIF-1α#1: GTGATGAAAGAATTACCGAAT

shHIF-1α#2: GCCGCTGGAGACACAATCATA

RNA extraction and real-time PCR

First, total RNA was extracted with RNAiso Plus reagent (Takara, #9108) following the manufacturer’s instructions. Second, the cDNA was performed using PrimeScript RT reagent kit with gDNA Eraser (TaKaRa, #RR047). Finally, real-time PCR was performed using TB Green Premix Ex Taq II (TaKaRa, #RR036) on PCR machine (LightCycler 96, #Roche). The sequences of primers are as follows:

GAPDH forward: TCCAAAATCAAGTGGGGCGA

GAPDH reverse: TGATGACCCTTTTGGCTCCC

HPRT1 forward: ACAGGACTGAACGTCTTGCT

HPRT1 reverse: GTCCCCTGTTGACTGGTCATT

IMPDH1 forward: CGTGCCCTACCTCATAGCAG

IMPDH1 reverse: GCCGCTTTTCGTAAGAGTGC

IMPDH2 forward: AGCTCTTCAACTGCGGAGAC

IMPDH2 reverse: GGATGAAGCCAATACCGCCT

HIF-1α forward: GGCGCGAACGACAAGAAAAA

HIF-1α reverse: GTGGCAACTGATGAGCAAGC

Western blot

Cells were lysed in RIPA lysis buffer containing protease inhibitors. Supernatants were subjected to SDS-PAGE precast glue, and transferred to PVDF membrane (Merck Millipore, #IPVH00005). The membranes were blocked with 5% skim milk powder at 37 ℃ for 1 h, and then incubated with the primary antibodies and second antibodies sequentially. The proteins were visualized with enhanced chemiluminescence (ECL) reagent on Chemiluminescence imaging system (Tanon, #2500). The primary antibodies are as follows: GAPDH (ZEN BioScience, #R24404), HPRT1 (abcom, #ab109021), HIF-1α (Proteintech Technology, #20960-1-AP), EGFR (Cell Signaling Technology, #4267), p-EGFR (abcom, #ab5644), Cleaved Caspase 3 (Cell Signaling Technology, #9661,), p-Chk1 (Proteintech Technology, #28803-1-AP), Chk1 (Proteintech Technology, #25887-1-AP) and γ-H2A.X (abcom, #ab81299).

Cell proliferation assay

For cell proliferation assay, 1000 cells were seeded in 96-well plates and incubated overnight. Next day, the experiment was carried out in accordance with BeyoClick™ EdU Cell Proliferation Kit with TMB (Beyotime, #C0088).

Clinical samples

Ten pairs of EGFR wild lung adenocarcinoma cancer tissue and paracancerous tissue samples, as well as nine pairs of EGFR mutant lung adenocarcinoma cancer tissue and paracancerous tissue samples were obtained from the First Affiliated Hospital of Zhengzhou University. All patients provided written informed consent. The study was approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University (2019-ky-30).

Metabolite extraction

For tissue samples, 20 mg of tissue sample was plated into the EP tube, and then the grinding ball and 500 µL of pre-cooled methanol containing the internal standard solution 1 (Human Metabolome Technologies, #H3304-1002) were added. After ground for 2 min, 500 µL of chloroform was added, vortexing for 30 s. Subsequently, 200 µL of ultrapure water was added, after 30 s of vortex the sample was stood for 10 min, further centrifuged at 4 °C and 13,000 rpm for 15 min, the supernatant was filtered and freeze-dried. Dried metabolite samples were stored at -80 °C.

For cell samples, cells were washed three times with 5% mannitol and then quenched by immersing them in liquid nitrogen. One mL of 100% methanol containing the internal standard solution 1 was added and cells were scraped into the 5 mL EP tube with a scraper. The following process was the same as above.

For isotope-labeled cell samples, cells were treated with 4 mM amide- 15 N-glutamine (Sigma Aldrich, #490024) or 60 µM 15 N 4 -hypoxanthine (Cambridge Isotope Laboratories, #NLM-8500-PK) for 24 h before collection.

CE − TOF/MS analysis

The experiment was performed based on a capillary electrophoresis system (CE, #G7100A, Agilent) equipped with a 1260 ISO pump (Agilent, #G1310B) coupled with a time-of-flight mass spectrometry system (TOF/MS, Agilent, #G6224A) with an electrophoresis-electrospray ionization-MS spray kit (Agilent, #G1607A). Agilent’s coaxial sheath fluid interface was used to connect capillary electrophoresis and mass spectrometer. CE and MS are controlled through ChemStation software (Agilent, B.04.03) and Mass Hunter Workstation software (Agilent, B.04.00). Metabolomics data were acquired in both cation-positive (CP) and anion-negative (AN) modes. The original CE-TOF/MS analysis data were processed using the software Qualitative analysis (Agilent, B.04.00), Quantitative Analysis (Agilent, B.04.00), and MethodMarker (Human Metabolome Technologies). Metabolite identification is based on a database constructed from 500 standards. Before statistical analysis, the exported data were sequentially subjected to internal standard normalization.

RNA sequencing

Total RNA was firstly extracted using a TRIzol total RNA extraction kit (TIANGEN, #DP424), then the library was constructed according to the manufacturer’s instructions and sequenced using the sequencing platform (Illumina, NovaSeq 6000). Gene set enrichment analysis (GSEA) was performed by the function in package clusterProfiler.

Luciferase reporter assay

293T cells (5 × 10 4 ) were seeded in 24-well plates at 5 × 10 4 /well, and then co-transfected with pLenti-EGFR-WT/pLenti-EGFR-Mut (E746-A750del; del19), pGL3.0-HPRT1 promoter and Renilla plasmids with Lipofectamine 2000 (Invitrogen, #11668500) in the next day. pLenti-EGFR-WT (Cat No.: PPL00063-4c) and pLenti-EGFR-Mut (E746-A750del; del19) (Cat No.: PPL00063-4d) plamsids were purchased from Geneppl technology, co, Ltd. pcDNA3.1-HIF-1α, pGL3.0-HPRT1 promoter/pGL3.0-HPRT1 promoter Mut and Renilla plasmids were co-transfected in 293T cells. After 24 h, dual-lucy assay kit (Solarbio, #D0010) was used to examined luciferase activity.

DNA pull‑down assay

The promoter region of HPRT1 was amplified by PCR using pGL3-Basic HPRT1 promoter as template and 5’-biotin-labeled forward primer. The HPRT1 HRE2 WT promoter sequence and the HPRT1 HRE2 Mut promoter sequence were synthesized by Sangon Biotech (Shanghai) Co., Ltd. The DNA probe was incubated with streptavidin agarose beads (Biovision, #6565-2). Then, the nuclear proteins were extracted from PC9 cells and added to the DNA-beads system and kept on a shaker 4 ℃ overnight. After multiple washes, the bound proteins were collected. Finally, the proteins were used for Coomassie Brilliant Blue staining, MS analysis and western blot. The primers sequences were listed below:

HPRT1 forward: biotin-GCTGACTGTACTGTCCTAAGTGCAT

HPRT1 reverse: AGGGCTCGTCGCAGCC

HPRT1 HRE2 WT: Biotin-AGCCACAGGTAGTGCAAGGTCTTGGGAATGGG ACGT CTGGTCCAAGGATTCACGCGATGACTGGAACCCGAA

HPRT1 HRE2 Mut: Biotin-AGCCACAGGTAGTGCAAGGTCTTGGGAATGGG AAAA CTGGTCCAAGGATTCACGCGATGACTGGAACCCGAA

Chromatin immunoprecipitation (ChIP) assays

Chip was performed using Pierce Agarose Chip Kit (Thermo Scientific, #26156) according to the manufacturer’s protocol. Cells were firstly fixed and crosslinked, then were lysed and digested with Micrococcal Nuclease. Chromatin was immnuoprecipitated with HIF-1α antibody (Cell Signaling Technology, #14179) or IgG antibody. Finally, protein/DNA complexes were eluted, and DNA was released. The later was amplified with the Chip primers. The primers sequences used were listed below:

HPRT1 HRE1 promoter forward: 5′-AGCCACAGGTAGTGCAAGG-3′

HPRT1 HRE1 promoter reverse: 5′-TTCGGGTTCCAGTCATCG-3′

HPRT1 HRE2 promoter forward: 5′-CTTAGAGGCTAGAAGAAA-3′

HPRT1 HRE2 promoter reverse: 5′-ATAGAGCTTGGCTCAATA-3′

Positive control (VEGF) forward: AGCAGGAACAAGGGCCTCTGTCT

Positive control (VEGF) reverse: GGAGGGAAGAGGACCTGTTGGAG

Negative control forward: TCAGGCTGTGAACCTTGGTGGGG

Negative control reverse: GCTCTGCGGACGCTCAGTGAAGC

Animal studies

All animal studies were approved by the Ethics Committee of Dalian Institute of Chemical Physics, Chinese Academy of Sciences (DICPEC2308). 4–6 weeks male nude mice were purchased from Liaoning Changsheng Biotechnology co., Ltd. The cells (5 × 10 6 ) were diluted with 100 µL PBS and mixed with 100 µL Matrigel. The suspension was injected subcutaneously into the dorsal flanks of 4–6 weeks male nude mice. Tumor size was measured every three days with a caliper, and the volume was calculated according to the formula: (major axe × minor axe 2 )/2. After the whole experiment, the mice were sacrificed and the tumors were removed and evaluated, followed by immunohistochemistry and metabolomics analysis. To explore whether the inhibition of purine metabolism could enhance gefitinib sensitivity, mice were injected with PC9 cells (5 × 10 6 ). When the tumor volume reached 100 mm 3 , the mice were randomly divided into four groups, mice in groups 1, 2, 3 and 4 were treated every day with normal saline, gefitinib (50 mg/kg) by gavage, 6-MP (100 mg/kg) by daily intraperitoneal injection, and the combination of gefitinib and 6-MP, respectively.

Immunohistochemistry (IHC)

Paraffin sections were dehydrated, and then incubated in 3% H 2 O 2 to eliminate endogenous peroxidase activity. Sections were blocked with 10% normal goat serum, followed by primary antibodies and biotin-labeled secondary antibodies. The experiment was performed following manufacturer’s instructions (ZSGB-BIO, #PV-9000). HPRT1 (1:150), Ki67 (1:50, Proteintech Technology, #27309-1-AP), HIF-1α (1:100) and γ-H2A.X (1:100) were used as the primary antibodies. H-score was used to assess the results of IHC and calculated according to following equation: H-score = Σpi(i + 1), where pi represents the percentage of positive cells to all cells in the slice, and i represents the staining intensity (0, negative; 1, weak; 2, moderate; and 3, intense).

Statistical analysis

Data were derived from three or more independent, replicate experiments, and presented as mean ± SD. Student’s t-test (unpaired, two-tailed) with p  < 0.05 was considered statistically significant between groups. Multiple data analysis was carried out using GraphPad Prism 6 Software (La Jolla, CA, USA), Gene Expression Profiling Interactive Analysis ( http://gepia2.cancer-pku.cn/#index ), Multi Experiment Viewer ( http://www.tm4.org ), SIMCA-P Software (Umetrics, Sweden), MetaboAnalyst ( http://www.metaboanalyst.ca ) and FlowJo 10.6.2 (Becton, Dickinson and Company).

Purine metabolism is enhanced in EGFR-mutant LUAD

Rapid proliferating tumors caused by EGFR mutations have abnormally activated purine metabolism [ 18 , 19 ]. To explore metabolic changes caused by muEGFR in LUAD, 10 pairs of adjacent normal and cancerous tissues from EGFR wild-type (WT) LUAD patients and 9 pairs of tissues from EGFR-mutant (Mut) LUAD patients were subjected to CE-TOF/MS-based metabolomics analysis. Compared with EGFR-WT LUAD tissues, quite a few metabolites altered in EGFR-Mut LUAD tissues (Fig.  1 A), significantly changed purine metabolism was found in the pathway enrichment analysis (Fig.  1 B). Further, we evaluated the metabolic differences between EGFR-WT and EGFR-Mut LUAD cells, heatmap showed that metabolites in purine metabolism had significant higher levels in EGFR-Mut LUAD cells, than in EGFR-WT LUAD cells (Fig.  1 C). Orthognonal partial least squares discriminant analysis (OPLS-DA) showed most of the metabolites had VIP (variable important for the projection) values greater than 1, revealing that metabolites involved in purine metabolism were discriminatory between EGFR-WT and EGFR-Mut LUAD cells (Fig.  1 D). To strengthen the verification of the effect of activated EGFR on purine metabolism, H1299, PC9 and H3255 cells were treated with gefitinib (an inhibitor of EGFR-TKIs), the results demonstrated that EGFR inhibition significantly reduced the metabolite contents of the purine metabolism in EGFR-Mut LUAD cells, while no effect was found in EGFR-WT LUAD cells (Fig.  1 E). These results suggested that EGFR mutations caused significant purine metabolism reprogramming in EGFR-Mut LUAD.

EGFR-mutant LUAD has enhanced purine metabolism. A Heatmap analysis of significantly differential metabolites. Metabolomics of tumor tissues and adjacent normal tissues derived from ten pairs of EGFR-WT and nine EGFR-Mut LUAD patients, then the ratio of metabolites in cancerous/adjacent normal tissue of the same patient was used to remove individual differences. p  < 0.05, unpaired two-tailed Student’s t-test. B Pathway enrichment analysis based on significantly differential metabolites. C Heatmap analysis of changed metabolites. Metabolites were extracted from EGFR-WT LUAD cells (H1299) and EGFR-Mut LUAD cells (PC9, H3255, H1975). n  = 4 independent cell cultures for each cell lines. p  < 0.05, unpaired two-tailed Student’s t-test. D Metabolites differentiating EGFR-WT from EGFR-Mut LUAD cell lines had variable importance (VIP scores > 1 were shown). The VIP value reflects the contribution of the metabolite to the classification of the model. Generally, VIP value greater than 1 is considered as a key contributor to the classification of the model. E Relative content of metabolites in purine metabolism pathway in H1299, PC9 and H3255 cells without or with gefitinib treatment for 48 h. n  = 5 independent cell cultures for each group sample. ns, not significant, * p  < 0.05, *** p  < 0.001

Mutated EGFR upregulates HPRT1 expression in EGFR-mutant LUAD

To elucidate the metabolic enzymes regulated by muEGFR, PC9 cells were treated with 100 nM gefitinib for 48 h and then subjected to RNA sequencing. Gene set enrichment analysis (GSEA) showed that genes in the purine metabolism pathway were significantly down-regulated (Fig.  2 A), the evidently changed genes were presented in Fig.  2 B. Real-time PCR and western blot were further carried out, and confirmed that EGFR inhibition reduced the expression of HPRT1 (a key synthetase in the purine salvage synthesis pathway) in EGFR-Mut LUAD cells, while no differential HPRT1 expression was found in EGFR-WT LUAD cells (Fig.  2 C, D). Gene expression analysis based on The Cancer Genome Atlas (TCGA) showed that HPRT1 was highly expressed in EGFR-Mut LUAD tissues compared with EGFR-WT LUAD tissues (Fig.  2 E).

HPRT1 is highly expressed and regulated by EGFR in EGFR-mutant LUAD. A Gene set enrichment analysis (GSEA) of purine metabolism gene sets. The data came from RNA sequencing performed in PC9 cells that treated with DMSO or 100 nM gefitinib for 48 h. B The heatmap of significantly changed genes of purine metabolism pathway. p  < 0.05. C-D RNA and protein expression of HPRT1 by real-time PCR and western blot. ns, not significant, * p  < 0.05, *** p  < 0.001. E Expression level of HPRT1 in EGFR-WT LUAD tumors ( n  = 103) and EGFR-Mut LUAD tumors ( n  = 29) based on TCGA dataset. *** p  < 0.001. F Fold change of HPRT1 protein expression in EGFR-WT and EGFR-Mut LUAD patients tumor tissues. * p  < 0.05. HPRT1 protein expression was shown in Fig. S1 A. G HPRT1 expression in EGFR-WT LUAD cells ( n  = 61) and EGFR-Mut LUAD cells ( n  = 16) based on CCLE dataset. H Protein expression of HPRT1 in EGFR-Mut LUAD cells compared to EGFR-WT LUAD cells by western blot. I pLenti-EGFR-WT/pLenti-EGFR-Mut (E746-A750del; del19), pGL3.0-HPRT1 promoter and pRL-TK plasmids were transfected into H1299 cells through lipofectamine 2000 transfection reagent for 48 h, and HPRT1 promoter activity was examined by luciferase reporter assay. ns: no significant; *** p  < 0.001

Subsequently, the expression of HPRT1 in cancerous tissues and the matched adjacent normal tissues from EGFR-WT and EGFR-Mut LUAD patients was examined. The results suggested that HPRT1 expression was higher in EGFR-Mut LUAD tissues (Fig. S1 A and Fig.  2 F). This discovery was confirmed at the cell line level via transcriptional data analysis obtained from the Cancer Cell Line Encyclopedia (CCLE) (Fig.  2 G). Moreover, detection of HPRT1 RNA and protein levels in EGFR-WT and EGFR-Mut cell lines also yielded consistent results (Fig. S1 B and Fig.  2 H). Further, the effect of muEGFR on HPRT1 promoter activity was assessed by luciferase reporter assay. The results revealed that mutated EGFR, but not wild-type EGFR, enhanced the activity of HPRT1 promoter (Fig.  2 I). Taken together, we found that LUAD with EGFR mutations had higher expression of HPRT1 than EGFR-WT LUAD.

HPRT1 promotes cell proliferation and tumorigenesis by enhancing purine metabolism in EGFR-mutant LUAD

To evaluate the biological function of HPRT1 in EGFR-mutant LUAD, HPRT1 was overexpressed in H1299, PC9 and H3255 cells. EdU cell proliferation assay demonstrated that HPRT1 overexpression promoted cell proliferation of EGFR-Mut LUAD cells, but not EGFR-WT LUAD cells (Fig.  3 A). Furthermore, stable transfected cell lines with HPRT1 knockdown in H1299, PC9 and H3255 cells were constructed. EdU incorporation assay confirmed that cell proliferation was inhibited in HPRT1 knockdown EGFR-Mut LUAD cells, whereas unremarkable cell proliferation change was observed in HPRT1 knockdown EGFR-WT LUAD cells (Fig.  3 B). Moreover, cell cycle assay showed that knockdown of HPRT1 prevented the G2/M progression, resulting in reduced cell proliferation ability in PC9 and H3255, however, HPRT1 inhibition had limited effect on H1299 cells (Fig.  3 C). HPRT1 overexpression further verified the effect on cell cycle (Fig. S2 A). At the metabolic level, we found that the contents of purine nucleotides were dramatically decreased in HPRT1-silenced PC9 and H3255 cells compared with HPRT1-silenced H1299 cells (Fig.  3 D). The conclusions drawn from the results of HPRT1 overexpressed cell lines were also consistent (Fig. S2 B). Together, these results suggested that HPRT1 was crucial for cell proliferation in EGFR-mutant LUAD cells.

HPRT1 promotes tumorigenesis by increasing the synthesis of purine nucleotides in EGFR-mutant LUAD. A The expression of HPRT1 (left), and cell proliferation (right) in HPRT1 overexpressing H1299, PC9 and H3255 cells. ns: not significant, *** p  < 0.001. B Western blot analysis of HPRT1 expression (left) and EdU incorporation assay (right) in HPRT1-silenced (shHPRT1) H1299, PC9 and H3255 cells,. ns: not significant, * p  < 0.05, *** p  < 0.001. C H1299/PC9/H3255-shHPRT1 and H1299/PC9/H3255-shCtrl cells were serum starved for 24 h, and then returned to normal culture for 24 h for cell cycle assay. n  = 3 independent replicates. * p  < 0.05, ** p  < 0.01, *** p  < 0.001. D Relative content of metabolites in purine metabolism pathway in H1299/PC9/H3255-shHPRT1 cells, H1299/PC9/H3255-shCtrl cells were used as control. ns: not significant, * p  < 0.05, ** p  < 0.01, *** p  < 0.001. E Tumor growth monitoration every three days after PC9-shCtrl and PC9-shHPRT1 cells injected subcutaneously into the dorsal flanks of nude mice. * p  < 0.05. F Real photographed tumor images. n  = 5. G Representative images of IHC staining of HPRT1 and Ki67 in xenograft tumor tissues (left), scale bars: 50 μm. H-score was used to assess the results of IHC (right). *** p  < 0.001. H Purine nucleotides measured by CE-TOF/MS in PC9-shCtrl and PC9-shHPRT1 cells-derived tumors. I Cell viability assay for various treatment, vehicle, 6-mercaptopurine (6-MP) and IMP alone, 6-MP and IMP combination. * p  < 0.05, ** p  < 0.01 and *** p  < 0.001. J Cell growth examination via crystal violet staining. Cells were plated in the 6-well plates, and were treated with vehicle, 6-MP, IMP, 6-MP and IMP combination for 96 h. Representative pictures were presented. K Cells from J were added with 33% acetic acid, and then the optical density was measured at 570 nm using a microplate reader. ** p  < 0.01, *** p  < 0.001

To further explore the effect of HPRT1 on the tumorigenic ability in EGFR-mutant LUAD in vivo, mouse xenograft models were constructed. The results indicated that knockdown of HPRT1 suppressed tumor growth and decreased tumor weight (Fig.  3 E, F). IHC assay showed that the expressions of HPRT1 and Ki67 were reduced in tumors formed by HPRT1-silenced cells (Fig.  3 G). Meanwhile, CE-TOF/MS-based metabolomics analysis revealed that purine nucleotides obviously decreased in the tumors derived from PC9-shHPRT1 cells compared to PC9-shCtrl cells-derived tumors (Fig.  3 H).

Next, to test whether HPRT1-mediated purine metabolism affected cell proliferation, 6-mercaptopurine (6-MP) was used to inhibit the activity of HPRT1 in H1299, PC9 and H3255 cells. Cell viability assay determined that 6-MP treatment decreased cell viability in EGFR-Mut LUAD cells, and IMP addition complemented the ability of cell proliferation, whereas this treatment had a limit effect on EGFR-WT LUAD cells (Fig.  3 I). Moreover, crystal violet staining experiment confirmed that the inhibition of cell proliferation caused by 6-MP could be complemented by IMP in EGFR-Mut LUAD cells but not in EGFR-WT LUAD cells (Fig.  3 J, K). Taken together, HPRT1 promoted cell proliferation and tumorigenesis through reprogramming purine metabolism in EGFR-mutant LUAD.

HIF-1α transcriptionally regulates the expression of HPRT1 in EGFR-mutant LUAD

Studies have shown that muEGFR could stabilize the protein expression of HIF-1α in EGFR-mutant NSCLC [ 20 ]. In our study, we found that the mRNA and protein expression of HIF-1α was higher in EGFR-mutant LUAD cells than in EGFR-wild LUAD cells (Fig. S3 A, B). Gefitinib was used to inhibit the activation of EGFR, real-time PCR and western blot experiments demonstrated that gefitinib impaired the expression of HIF-1α at RNA and protein level in PC9 and H3255 cells (Fig. S3 C, D). Further results showed that gefitinib caused a drastic shortened half-life of HIF-1α in PC9 cells with cycloheximide (CHX) treatment, and the effect of gefitinib on HIF-1α was blocked by MG132-proteasome inhibitor (Fig. S3 E, F). The results suggested that muEGFR increased the expression of HIF-1α through enhancing the protein stability.

To explore the upstream key factors that regulated HPRT1 expression in EGFR-mutant LUAD, biotinylated primer was designed and synthesized based on the HPRT1 promoter sequence, and biotinylated DNA probe was obtained by PCR amplification. Proteins bound to the HPRT1 promoter was captured using DNA pull down assay. Coomassie brilliant blue staining and mass spectrometry identified that the transcription factor HIF-1α might bind to the promoter of HPRT1 (Fig.  4 A, B). Furthermore, we verified the binding of HIF-1α to the HPRT1 promoter by western blot experiment (Fig.  4 C). Subsequently, The Eukaryotic Promoter Database (EPD) was used to search for potential HIF-1α binding sites on the HPRT1 promoter. We found that the HPRT1 promoter contained two potential HREs (HRE1, which was located at -1947–1940 bp, and HRE2, which was located at -329–322 bp). ChIP assay with HIF-1α antibody confirmed that HIF-1α could bind to the region containing HRE2 (Fig. S4). To further verify this result, biotin-labeled HRE2 WT and Mut probes were used to perform DNA pull down experiment, and results demonstrated that HRE2 was the binding site of HIF-1α (Fig.  4 D). To confirm whether HIF-1α transcriptionally regulated HPRT1, luciferase reporter assay was performed in 293T cells. The results showed that HIF-1α enhanced wild-type HPRT1 promoter activity, whereas it had no effect on the mutated HPRT1 promoter activity (Fig.  4 E). Furthermore, HIF-1α was knocked down by two small-hairpin (shRNA) in PC9 cells and H3255 cells. We found that HIF-1α knockdown led to obvious reduction in the binding of HIF-1α to HPRT1 promoter by ChIP assay (Fig.  4 F). Next, analysis based on the TCGA database showed a positive correlation between HIF-1α and HPRT1 in EGFR-mutant LUAD tissues (Fig.  4 G). At the cellular level, HPRT1 was significantly decreased in HIF-1α-silenced PC9 and H3255 cells (Fig.  4 H, I). Our results confirmed that the expression of HPRT1 was regulated by HIF-1α at the transcriptional level.

HIF-1α is HPRT1 promoter-binding protein, and regulates the expression of HPRT1. A Potential HPRT1 promoter-binding proteins after SDS-PAGE and Coomassie brilliant blue staining. B The binding proteins identified by mass spectrometry analysis. C The combination of HIF-1α and HPRT1 promoter verified by western blot. D DNA pulldown experiment performed with biotinylated HPRT1 HRE2 WT and Mut probes in PC9 cells. E Luciferase reporter assay performed in 293T cells co-transtected with HPRT1 promoter WT, HPRT1 promoter mut, HIF-1α plasmids (up). ns, not significant, ** p  < 0.01. The expression of HIF-1α examined by western blot (down). F The level of HIF-1α presence at the promoter of HPRT1 via qChIP assay. VEGF was used as a positive control, and irrelevant sequence was used as a negative control (up). ** p  < 0.01. The expression of HIF-1α examined by western blot (down). G The correlation analysis between HIF-1α and HPRT1 in EGFR-mutant LUAD patients based on TCGA database. H-I The expression of HPRT1 in PC9-shHIF-1α and H3255-shHIF-1α cells by real-time PCR and western blot experiments

HIF-1α promotes HPRT1-mediated purine metabolism and tumorigenesis

To elucidate the impact of HIF-1α on HPRT1 function in EGFR-mutant LUAD, we first examined the changes in HIF-1α-mediated metabolic reprogramming. Metabolites levels of PC9-shCtrl and PC9-shHIF-1α cells were analyzed using CE-TOF/MS. Pathway enrichment analysis showed the altered purine metabolism (Fig.  5 A). The contents of metabolites in the purine metabolism were significantly decreased in HIF-1α-silenced PC9 cells (Fig.  5 B). Further, a stable isotope-labeled metabolic flux technique was used to examine the effect of HIF-1α knockdown on purine nucleotide synthesis. Isotope labeling experiment using amide- 15 N-glutamine showed that HIF-1α knockdown didn’t alter de novo purine metabolism (Fig. S5A). However, using 15 N 4 -hypoxanthine for isotope labeling experiment, we found that HIF-1α knockdown reduced the synthesis rate of purine metabolism from the salvage synthesis pathway (Fig.  5 C). Taken together, the results indicated that HIF-1α regulated HPRT1-mediated purine salvage synthesis pathway.

HIF-1α upregulates HPRT1 to promote purine metabolism and tumorigenesis in EGFR-mutant LUAD. A-B Differential metabolites in PC9-shCtrl and PC9-shHIF-1α cells detected by CE-TOF/MS based metabolomics. Enrichment analysis based on different metabolites ( A ), relative abundance of purine metabolism intermediates in PC9-shCtrl and PC9-shHIF-1α cells ( B ). C Schematic illustrating 15 N 4 labeling of nucleotides from 15 N 4 -hypoxanthine (left), and 15 N 4 -hypoxanthine-labeled purine metabolism intermediates in PC9-shCtrl and PC9-shHIF-1α cells measured by CE-TOF/MS based metabolomics (right). Data were assessed by two-tailed Student’s t-test. * p  < 0.05, *** p  < 0.001. D The expression of HIF-1α, HPRT1 analyzed by western blot in PC9-shCtrl, PC9-shHIF-1α and PC9-shHIF-1α HPRT1 cells. E-F EdU incorporation assay (E) and cell cycle assay (F) performed in PC9-shHIF-1α and PC9-shHIF-1α HPRT1 cells, PC9-shCtrl cells was used as control. G Relative content of IMP, GMP, GDP, GTP, AMP, ADP and ATP evaluated based on CE-TOF/MS metabolomic analysis in PC9-shCtrl, PC9-shHIF-1α and PC9-shHIF-1α HPRT1 cells. H Tumor images of Balb/c nude mice after 23 days’ injection with PC9-shCtrl, PC9-shHIF-1α and PC9-shHIF-1α HPRT1 cells. I IHC staining of HIF-1α, HPRT1 and Ki67 in tumor tissues. The representative images were shown. Scale bars: 50 μm (left). The H-score of IHC results (right). *** p  < 0.001. J The analysis of purine nucleotides content derived from tumor tissues. ns, not significant, * p  < 0.05

Next, HPRT1 was overexpressed in HIF-1α-silenced PC9 cells (Fig. S5B and Fig.  5 D). The EdU incorporation assay demonstrated that the ability of cell proliferation was dramatically reduced by HIF-1α knockdown, and was restored when HPRT1 was overexpressed in PC9-shHIF-1α cells (Fig.  5 E). Cell cycle assay showed that the proportion of cells in the G2/M phase was severely reduced in PC9-shHIF-1α cells, and the effect was reversed by HPRT1 overexpression (Fig.  5 F). To verify whether HIF-1α-HPRT1 axis promoted purine metabolism, CE-TOF/MS-based metabolomics analysis was performed. The results revealed that the contents of purine nucleotides, such as AMP, GMP and IMP, were restored in HPRT1 overexpressed PC9-shHIF-1α cells (Fig.  5 G).

Finally, to examine the effect of the HIF-1α-HPRT1 axis on tumorigenesis in vivo, PC9-shCtrl, PC9-shHIF-1α and PC9-shHIF-1α HPRT1 cells were subcutaneously injected into the dorsal flank of mice to construct mouse xenograft models. The results showed that HIF-1α knockdown inhibited tumor growth, while HPRT1 overexpression restored it (Fig.  5 H). The expression of HIF-1α, HPRT1 and Ki67 in mouse tumor tissues was detected by IHC (Fig.  5 I). Moreover, the contents of purine nucleotides were decreased in PC9-shHIF-1α cells-derived tumor tissues compared to control group, while the overexpression of HPRT1 restored the contents of purine nucleotides (Fig.  5 J).

Inhibition of HPRT1 sensitizes EGFR-mutant LUAD to EGFR-TKI

Gefitinib was one of the first-line routine drugs for the treatment of advanced EGFR-mutant LUAD, whether HPRT1 promoted resistance to EGFR-targeted therapy was unclear. In our study, PC9-shCtrl and PC9-shHPRT1 cells were treated without or with gefitinib, cell viability assay indicated that the inhibitory rate of cell proliferation was more significant in PC9-shHPRT1 cells than in PC9-shCtrl cells (Fig.  6 A). Meantime, PC9 cells co-treated with 6-MP and gefitinib showed lower cell viability than other groups (Fig.  6 B). Cell apoptosis assay by flow cytometry demonstrated that PC9 cells treated with the combination of 6-MP and gefitinib appeared to have the strongest apoptosis (Fig.  6 C). Furthermore, the western blot experiment revealed that the combination of 6-MP and gefitinib induced cell apoptosis and DNA damage (Fig.  6 D). These results demonstrated that the inhibition of HPRT1 increased the sensitivity of gefitinib to EGFR-mutant LUAD cells.

HPRT1 inhibition enhances the sensitivity of EGFR-mutant LUAD cells to EGFR-TKI. A Cell viability assay performed in PC9-shCtrl and PC9-shHPRT1 cells with 25 nM gefitinib treatment for 48 h, and the inhibition rate of proliferation was calculated. ** p  < 0.01, *** p  < 0.001. B Cell viability of PC9 cells treated with 6-MP and gefitinib alone or combination for 48 h. ** p  < 0.01, *** p  < 0.001. C Cell apoptosis analyzed by PI/Annexin V staining of PC9 cells, which subjected to the same treatment with that in Fig B,. The positive proportion of annexin V was calculated. ** p  < 0.01, *** p  < 0.001. n  = 3 independent replicates. D Protein expression of Cleaved Caspase 3 (Cl. Casp-3), γ-H2A.X and p-Chk1 examined by western blot in PC9 cells, which subjected to the same treatment with that in Fig B. E Real photographed tumor images. After the formation of PC9 cells derived xenograft tumors, mice were treated with 6-MP and gefitinib alone or combination every day. After 18 days of administration, the tumors were removed. F Tumor weighs were calculated. * p  < 0.05, ** p  < 0. 01. G Representative images of IHC staining of Ki67 and γ-H2A.X in xenograft tumor tissues, scale bars: 50 μm (left). H-score is presented to assess the results of IHC (right). *** p  < 0.001. H-I The overall survival and disease free survival of patients expressing low or high HPRT1 in LUAD based on GEPIA2. J The overall survival of HPRT1 in EGFR-mutant LUAD based on GSE72094 datasets

To further study the effects in vivo, PC9 cells were injected subcutaneously into male balb/c nude mice, then mice were treated with gefitinib and 6-MP alone or in combination. The results showed that mice with the treatment of gefitinib and 6-MP combination had much smaller tumor volumes and weights than in other groups (Fig.  6 E, F). Moreover, IHC staining with Ki67 and γ-H2A.X antibodies revealed that 6-MP combined with gefitinib inhibited tumor growth (Fig.  6 G).

In summary, the inhibition of HPRT1 made tumors more sensitive to gefitinib treatment. Finally, the correlation of HPRT1 with the prognosis of LUAD was evaluated using the GEPIA2 database. The overall survival and disease free survival cures revealed that highly expressed HPRT1 was positively associated with poor prognosis in LUAD (Fig.  6 H, I). Furthermore, based on prognostic information of EGFR-mutant LUAD patients in GSE72094, highly expressed HPRT1 showed a trend toward poor prognosis but there was no significant difference (Fig.  6 J).

In this study, we observed that purine metabolism was enhanced in EGFR-mutant LUAD cells and patient tissues. Purine metabolism is recognized as a significant contributor to cellular growth and proliferation. In tumor cells, purine metabolism usually occurs abnormally, resulting in an increase or decrease in the contents of intracellular purine nucleotides, which in turn affects the proliferation and drug resistance of tumor cells [ 21 ]. Purine-metabolizing enzymes can disrupt the balance of purine pools, thereby interfering with proliferation and migration of tumor cells. The enzymes involved in purine metabolism are overexpressed in various cancer, such as adenosine deaminase (ADA), cytoplasmic-5’-nucleotidase-II (CN-II) and inosine monophosphate dehydrogenase (IMPDH1) [ 22 , 23 , 24 ]. In this study, we found that HPRT1 was overexpressed in EGFR-mutant LUAD cells and tissues, and HPRT1 promoted cell proliferation in vitro and tumor growth in vivo. Similarly, Wang et al. demonstrated that the overexpression of HPRT1 promotes proliferation and migration in head and neck squamous cell carcinoma [ 25 ]. HPRT1 has also been found to promote cisplatin resistance by activating phosphoinositide 3-kinase (PI3K)/AKT pathway in oral squamous cell carcinoma [ 16 ]. Furthermore, salvage purine synthesis pathway mediated by HPRT1 supports cell growth in SCLC [ 17 , 26 ]. Our results are consistent with theirs [ 17 , 26 ], we found that HPRT1-mediated purine metabolism was essential for cell proliferation of EGFR-mutant LUAD. However, our results confirmed that HPRT1 had no effect on the proliferation and purine metabolism of EGFR-WT LUAD cells, possibly because other enzymes in purine metabolism or other pathways antagonize the role of HPRT1 in EGFR-WT LUAD. Similarly, Li et al. reported that PAICS (a de novo purine metabolic enzyme) knockdown could reduce the cell viability of EGFR-WT cells, but had no effect on EGFR-Mut cells. Moreover, after the PAICS gene was knocked out, the p-AKT protein level in EGFR-Mut NSCLC cells was significantly higher than that in EGFR-WT NSCLC cells. Furthermore, BKM120 (a pan-PI3K inhibitor) treatment resulted in a significant decrease in the proliferative capacity in PAICS knockdown EGFR-Mut NSCLC cells, suggesting that activated PI3K-AKT signaling may impair the effects of PAICS knockdown [ 27 ].

MuEGFR upregulates HIF-1α expression by stabilizing HIF-1α protein in a hypoxia-independent manner [ 20 , 28 ]. Consistently, our study verified that the upregulation of HIF-1α was regulated by muEGFR by enhancing protein stability of HIF-1α. HIF-1α acts a transcription factor to regulate the transcription of multiple metabolic enzymes, thereby causing tumor metabolic reprogramming. Specifically, HIF-1α enhances glycolysis and PPP by regulating the expression of pyruvate kinase M2 (PKM2) and glucose-6-phosphate dehydrogenase (G6PD) to impart 5-fluorouracil resistance in colorectal cancer [ 29 ]. HIF-1 drives lipid deposition by suppressing carnitine palmitoyltransferase 1 A (CPT1A) to promote tumor growth in clear cell renal cell carcinoma [ 30 ]. We found that HIF-1α directly bound to the promoter of HPRT1, and transcriptionally activated the expression of HPRT1 to promote purine nucleotides synthesis in EGFR-mutant LUAD cells.

6-MP is an HPRT1 inhibitor and has been widely used clinically as an antileukemia drug and an immunosuppressive drug [ 31 , 32 ]. In this study, 6-MP combined with gefitinib was used to study the effect on tumor growth of EGFR-mutant LUAD. The results revealed that EGFR-mutant LUAD cells treated with HPRT1 inhibitor were more sensitive to gefitinib in vitro and in vivo. Wang et al. demonstrate that the inhibition of HPRT1 increases the anticancer effect of EGFR-TKI in head and neck squamous cell carcinoma [ 25 ]. In addition, combination treatment of 6-MP and temozolomide has been found to inhibit brain tumor growth [ 33 ]. 6-MP combined with methotrexate and methionine sulfoximine suppress the tumor growth of SCLC [ 17 ]. Whether 6-MP can increase the effectiveness of EGFR-TKIs-resistant LUAD patients and prolong the survival of patients requires further in-depth study.

Our study reveals disorders of purine metabolism in EGFR-mutant LUAD from the perspective of metabolism reprogramming. It is confirmed that purine metabolism catalyzed by HPRT1 promotes the proliferation of EGFR-mutant LUAD in vitro and in vivo. Furthermore, the study of the mechanism shows that HIF-1α transcriptionally regulates HPRT1 to accelerate purine nucleotides synthesis to promote cell proliferation and tumorigenesis. Finally, inhibition of HPRT1 coupled with EGFR-TKIs significantly inhibits the tumor growth of EGFR-mutant LUAD (Fig.  7 ). The study indicates that intervening purine metabolism of tumors may be a new target for clinical treatment of EGFR-mutant LUAD.

Schematic diagram of muEGFR-HIF-1α-HPRT1 function axis. In EGFR-mutant lung adenocarcinoma, mutated EGFR upregulates HIF-1α expression through protein stability. Furthermore, HIF-1α binds to the promoter of HPRT1 and promotes its transcription. Upregulated HPRT1 enhances purine metabolism to promote purine nucleotide synthesis, ultimately accelerating cell proliferation. Finally, 6-MP combined with EGFR-TKI inhibits cell proliferation and induces DNA damage and cell apoptosis

Data availability

The authors declare that all data supporting the findings of this study are available in this article and its supplementary files.

Abbreviations

Epidermal growth factor receptor

Lung adenocarcinoma

Hypoxanthine phosphoribosyltransferase 1

Hypoxia-inducible factor-1 alpha

Tyrosine kinase inhibitors

Small cell lung cancer

Gene set enrichment analysis

The Cancer Genome Atlas

Cancer Cell Line Encyclopedia

6-mercaptopurine

Eukaryotic Promoter Database

Adenosine deaminase

Cytoplasmic-5’-nucleotidase-II

Inosine monophosphate dehydrogenase

Phosphoinositide 3-kinase

Pyruvate kinase M2

Glucose-6-phosphate dehydrogenase

Carnitine palmitoyltransferase 1 A

Chromatin immunoprecipitation

Immunohistochemistry

Shi Y, Au JS, Thongprasert S, Srinivasan S, Tsai CM, Khoa MT, et al. A prospective, molecular epidemiology study of EGFR mutations in Asian patients with advanced non-small-cell lung cancer of adenocarcinoma histology (PIONEER). J Thorac Oncol. 2014;9(2):154–62.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Yoneda K, Imanishi N, Ichiki Y, Tanaka F. Treatment of non-small cell Lung Cancer with EGFR-mutations. J UOEH. 2019;41(2):153–63.

Article   CAS   PubMed   Google Scholar  

Wu SG, Shih JY. Management of acquired resistance to EGFR TKI-targeted therapy in advanced non-small cell lung cancer. Mol Cancer. 2018;17(1):38.

Article   PubMed   PubMed Central   Google Scholar  

Oxnard GR, Hu Y, Mileham KF, Husain H, Costa DB, Tracy P, et al. Assessment of Resistance mechanisms and clinical implications in patients with EGFR T790M-Positive Lung Cancer and Acquired Resistance to Osimertinib. JAMA Oncol. 2018;4(11):1527–34.

Article   PubMed   Google Scholar  

Dyrstad SE, Lotsberg ML, Tan TZ, Pettersen IKN, Hjellbrekke S, Tusubira D et al. Blocking aerobic glycolysis by targeting pyruvate dehydrogenase kinase in combination with EGFR TKI and Ionizing Radiation increases therapeutic effect in Non-small Cell Lung Cancer cells. Cancers (Basel). 2021;13(5).

Zhang KR, Zhang YF, Lei HM, Tang YB, Ma CS, Lv QM, et al. Targeting AKR1B1 inhibits glutathione de novo synthesis to overcome acquired resistance to EGFR-targeted therapy in lung cancer. Sci Transl Med. 2021;13(614):eabg6428.

Xu C, Zhang L, Wang D, Jiang S, Cao D, Zhao Z, et al. Lipidomics reveals that sustained SREBP-1-dependent lipogenesis is a key mediator of gefitinib-acquired resistance in EGFR-mutant lung cancer. Cell Death Discov. 2021;7(1):353.

Eltayeb K, La Monica S, Tiseo M, Alfieri R, Fumarola C. Reprogramming of lipid metabolism in Lung Cancer: an overview with Focus on EGFR-Mutated Non-small Cell Lung Cancer. Cells. 2022;11(3).

Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646–74.

De Vitto H, Arachchige DB, Richardson BC, French JB. The intersection of Purine and mitochondrial metabolism in Cancer. Cells. 2021;10(10).

Di Virgilio F, Adinolfi E. Extracellular purines, purinergic receptors and tumor growth. Oncogene. 2017;36(3):293–303.

Liu J, Hong S, Yang J, Zhang X, Wang Y, Wang H, et al. Targeting purine metabolism in ovarian cancer. J Ovarian Res. 2022;15(1):93.

Townsend MH, Robison RA, O’Neill KL. A review of HPRT and its emerging role in cancer. Med Oncol. 2018;35(6):89.

Choudhari MJSEIR, Subramani RALH, Lakshmanaswamy R. R, Hypoxanthine Phosphoribosyl Transferase 1 is upregulated, predicts clinical outcome and controls Gene expression in breast Cancer. Cancers (Basel). 2020;12(6).

Townsend MH, Ence ZE, Cox TP, Lattin JE, Burrup W, Boyer MK, et al. Evaluation of the upregulation and surface expression of hypoxanthine guanine phosphoribosyltransferase in acute lymphoblastic leukemia and Burkitt’s B cell lymphoma. Cancer Cell Int. 2020;20:375.

Wu T, Jiao Z, Li Y, Su X, Yao F, Peng J et al. HPRT1 promotes chemoresistance in oral squamous cell Carcinoma via activating MMP1/PI3K/Akt signaling pathway. Cancers (Basel). 2022;14(4).

Kodama M, Toyokawa G, Sugahara O, Sugiyama S, Haratake N, Yamada Y, et al. Modulation of host glutamine anabolism enhances the sensitivity of small cell lung cancer to chemotherapy. Cell Rep. 2023;42(8):112899.

Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science. 2004;304(5676):1497–500.

Chen PH, Cai L, Huffman K, Yang C, Kim J, Faubert B, et al. Metabolic diversity in Human Non-small Cell Lung Cancer cells. Mol Cell. 2019;76(5):838–51. e5.

Nilsson MB, Robichaux J, Herynk MH, Cascone T, Le X, Elamin Y, et al. Altered regulation of HIF-1alpha in naive- and drug-resistant EGFR-Mutant NSCLC: implications for a vascular endothelial growth factor-dependent phenotype. J Thorac Oncol. 2021;16(3):439–51.

Yin J, Ren W, Huang X, Deng J, Li T, Yin Y. Potential mechanisms connecting Purine Metabolism and Cancer Therapy. Front Immunol. 2018;9:1697.

Cividini F, Cros-Perrial E, Pesi R, Machon C, Allegrini S, Camici M, et al. Cell proliferation and drug sensitivity of human glioblastoma cells are altered by the stable modulation of cytosolic 5’-nucleotidase II. Int J Biochem Cell Biol. 2015;65:222–9.

Rocha R, Torres A, Ojeda K, Uribe D, Rocha D, Erices J et al. The Adenosine A(3) receptor regulates differentiation of Glioblastoma Stem-Like Cells to endothelial cells under Hypoxia. Int J Mol Sci. 2018;19(4).

Valvezan AJ, Turner M, Belaid A, Lam HC, Miller SK, McNamara MC, et al. mTORC1 couples nucleotide synthesis to Nucleotide demand resulting in a targetable metabolic vulnerability. Cancer Cell. 2017;32(5):624–38. e5.

Wang L, Wang Y, Han N, Wang X, Ruan M. HPRT promotes proliferation and metastasis in head and neck squamous cell carcinoma through direct interaction with STAT3. Exp Cell Res. 2021;399(1):112424.

Tabata S, Umemura S, Narita M, Udagawa H, Ishikawa T, Tsuboi M, et al. Metabolic Hallmarks for Purine Nucleotide Biosynthesis in Small Cell Lung Carcinoma. Mol Cancer Res. 2024;22(1):82–93.

LI Y, ZHU L, MAO J et al. Genome-scale CRISPR-Cas9 screen identifies PAICS as a therapeutic target for EGFR wild-type non-small cell lung cancer [J]. MedComm (2020), 2024, 5(3): e483.

Xu L, Nilsson MB, Saintigny P, Cascone T, Herynk MH, Du Z, et al. Epidermal growth factor receptor regulates MET levels and invasiveness through hypoxia-inducible factor-1alpha in non-small cell lung cancer cells. Oncogene. 2010;29(18):2616–27.

Dong S, Liang S, Cheng Z, Zhang X, Luo L, Li L, et al. ROS/PI3K/Akt and Wnt/beta-catenin signalings activate HIF-1alpha-induced metabolic reprogramming to impart 5-fluorouracil resistance in colorectal cancer. J Exp Clin Cancer Res. 2022;41(1):15.

Du W, Zhang L, Brett-Morris A, Aguila B, Kerner J, Hoppel CL, et al. HIF drives lipid deposition and cancer in ccRCC via repression of fatty acid metabolism. Nat Commun. 2017;8(1):1769.

Atkinson MR, Murray AW. Inhibition by 6-Mercaptopurine of Purine Phosphoribosyltransferases from Ehrlich Ascites-Tumour cells that are resistant to the drug. Biochem J. 1965;94(1):71–4.

Sahasranaman S, Howard D, Roy S. Clinical pharmacology and pharmacogenetics of thiopurines. Eur J Clin Pharmacol. 2008;64(8):753–67.

Yin J, Wang X, Ge X, Ding F, Shi Z, Ge Z, et al. Hypoxanthine phosphoribosyl transferase 1 metabolizes temozolomide to activate AMPK for driving chemoresistance of glioblastomas. Nat Commun. 2023;14(1):5913.

Download references

Acknowledgements

Not applicable.

This work was supported by the foundations from the National Natural Science Foundation of China (No. 21934006, No. 22074144), Strategic Priority Research Program (B) of Chinese Academy of Sciences (XDB38020200), foundation from the Youth Innovation Promotion Association CAS (2021186) and the innovation program (DICP I202334) of science and research from the DICP, CAS.

Author information

Authors and affiliations.

State Key Laboratory of Medical Proteomics, CAS Key Laboratory of Separation Science for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, Liaoning Province, 116023, China

Pengyu Geng, Peng Dou, Chunxiu Hu, Jinhui Zhao, Qi Li, Xinyu Liu & Guowang Xu

Liaoning Province Key Laboratory of Metabolomics, Dalian, Liaoning Province, 116023, China

The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan Province, 450052, China

Fei Ye & Xiangnan Li

Clinical Laboratory, The Second Hospital of Dalian Medical University, Dalian, Liaoning Province, 116023, China

Department of Medical Oncology, The First Affiliated Hospital of Xi’an Medical University, Xi’an, Shanxi Province, 710082, China

You can also search for this author in PubMed   Google Scholar

Contributions

GPY peformed cell and animal experiments, and writed the manuscript. HCX provided the technical guideance on the CE-TOF/MS metabolic analysis. DP revised the manuscript and provided important suggestions. HJR performed some cell experiments. ZJH performed the bioinformatics analysis. LQ performed metabolic analysis. BM peformed cell viability assay. YF and LXN provided patients tissue samples. LXY and XGW designed and supervised the entire study, and revised the manuscript. The authors read and approved the final manuscript.

Corresponding authors

Correspondence to Xiangnan Li , Xinyu Liu or Guowang Xu .

Ethics declarations

Ethics approval and consent to participate.

For animal studies, all the experimental procedures were performed in accordance with protocols approved by the Ethics Committee of Dalian Institute of Chemical Physics, Chinese Academy of Sciences. The ethical review approval number was DICPEC2308. For tissue studies, human tissue samples were approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University. The ethical review approval number was 2019-ky-30.

Consent for publication

All authors agree to publish this manuscript.

Competing interests

The authors declare no conflict of interest.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Geng, P., Ye, F., Dou, P. et al. HIF-1α-HPRT1 axis promotes tumorigenesis and gefitinib resistance by enhancing purine metabolism in EGFR-mutant lung adenocarcinoma. J Exp Clin Cancer Res 43 , 269 (2024). https://doi.org/10.1186/s13046-024-03184-8

Download citation

Received : 11 May 2024

Accepted : 05 September 2024

Published : 30 September 2024

DOI : https://doi.org/10.1186/s13046-024-03184-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Purine metabolism

Journal of Experimental & Clinical Cancer Research

ISSN: 1756-9966

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

• Open access
• Published: 23 September 2024

It takes two peroxisome proliferator-activated receptors (PPAR-β/δ and PPAR-γ) to tango idiopathic pulmonary fibrosis

• Eistine Boateng 1   nAff8 ,
• Rocio Bonilla-Martinez 1 ,
• Barbara Ahlemeyer 1 ,
• Vannuruswamy Garikapati 1 , 6   nAff9 ,
• Mohammad Rashedul Alam 1 ,
• Omelyan Trompak 2 ,
• Gani Oruqaj 1   nAff10 ,
• Natalia El-Merhie 1   nAff11 ,
• Michael Seimetz 3 ,
• Clemens Ruppert 3 , 4 ,
• Andreas Günther 3 , 5 ,
• Bernhard Spengler 6 ,
• Srikanth Karnati 1   na1   nAff7 &
• Eveline Baumgart-Vogt 1   na1  

Respiratory Research volume  25 , Article number:  345 ( 2024 ) Cite this article

436 Accesses

3 Altmetric

Metrics details

Idiopathic pulmonary fibrosis (IPF) is characterized by aberrant lung epithelial phenotypes, fibroblast activation, and increased extracellular matrix deposition. Transforming growth factor-beta (TGF-β)1-induced Smad signaling and downregulation of peroxisomal genes are involved in the pathogenesis and can be inhibited by peroxisome proliferator-activated receptor (PPAR)-α activation. However, the three PPARs, that is PPAR-α, PPAR-β/δ, and PPAR-γ, are known to interact in a complex crosstalk.

To mimic the pathogenesis of lung fibrosis, primary lung fibroblasts from control and IPF patients with comparable levels of all three PPARs were treated with TGF-β1 for 24 h, followed by the addition of PPAR ligands either alone or in combination for another 24 h. Fibrosis markers (intra- and extracellular collagen levels, expression and activity of matrix metalloproteinases) and peroxisomal biogenesis and metabolism (gene expression of peroxisomal biogenesis and matrix proteins, protein levels of PEX13 and catalase, targeted and untargeted lipidomic profiles) were analyzed after TGF-β1 treatment and the effects of the PPAR ligands were investigated.

TGF-β1 induced the expected phenotype; e.g. it increased the intra- and extracellular collagen levels and decreased peroxisomal biogenesis and metabolism. Agonists of different PPARs reversed TGF-β1-induced fibrosis even when given 24 h after TGF-β1. The effects included the reversals of (1) the increase in collagen production by repressing COL1A2 promoter activity (through PPAR-β/δ activation); (2) the reduced activity of matrix metalloproteinases (through PPAR-β/δ activation); (3) the decrease in peroxisomal biogenesis and lipid metabolism (through PPAR-γ activation); and (4) the decrease in catalase protein levels in control (through PPAR-γ activation) and IPF (through a combined activation of PPAR-β/δ and PPAR-γ) fibroblasts. Further experiments to explore the role of catalase showed that an overexpression of catalase protein reduced collagen production. Additionally, the beneficial effect of PPAR-γ but not of PPAR-β/δ activation on collagen synthesis depended on catalase activity and was thus redox-sensitive.

Our data provide evidence that IPF patients may benefit from a combined activation of PPAR-β/δ and PPAR-γ.

IPF is a severe restrictive interstitial lung disease with patient median survival of 2.5–3.5 years [ 1 ]. Concerning the pathogenesis of IPF, it is being discussed that an excessive injury response results in persistent overproduction of extracellular matrix (ECM) components by activated and proliferating fibroblasts. In addition, oxidative stress remains a major mechanism associated with the progression of this disease [ 2 ]. Today, only limited treatment options for IPF are available. Evidence-based recommendations for the pharmacological management of the disease are the tyrosine kinase inhibitor nintedanib [ 3 , 4 ] and pirfenidone [ 4 , 5 ], an inhibitor of TGF-β1-stimulated collagen synthesis. Both drugs increase quality of life, attenuate symptoms and slow down IPF progression, but only nintedanib influences mortality. Some of the novel medications targeted pentraxin (involved in endogenous tissue repair), lysophosphatidic acid, or connective tissue growth factor (mediates TGF-β1 downstream signaling), but failed the clinical endpoints [ 6 , 7 ]. Other substances in the pipeline are nerandomilast (a tyrosine kinase inhibitor) which successfully completed phase II clinical trials [ 8 ] and inhaled treprostinil, a prostacyclin analogue. Treprostinil showed beneficial effects in the initial INCREASE trial [ 9 ] and ongoing TETON study [ 10 ] and has meanwhile been approved for the therapy of WHO group 1 pulmonary hypertension with an additional positive impact in IPF. Nevertheless, extensive research is still required to develop new therapeutic modalities.

To find therapeutic interventions for IPF, several studies explored the anti-fibrotic potentials of natural and synthetic PPAR ligands. For example, PPAR-α activation was demonstrated to attenuate fibrosis in the liver [ 11 ], heart [ 12 ] and lung [ 13 , 14 ], while PPAR-β agonists exhibited anti-proliferative effects [ 15 ], but increased the secretion of TGF-β1 and ECM [ 16 ]. Ligands of PPAR-γ are most promising [ 17 , 18 , 19 , 20 ] and were thought to inhibit fibroblast trans-differentiation [ 21 , 22 ] and to strengthen the anti-oxidative defense system [ 23 ]. In addition, pan-PPAR agonists, such as lanifibranor [ 24 ] and IVA337 [ 25 ] attenuated fibrosis. In all these studies, however, the anti-fibrotic mechanism of PPAR agonists remained unclear and was supposed to be mainly due to their anti-inflammatory activities [ 26 ]. Another drawback was the time schedule of the drug treatment. Typically, drugs were added before or together with TGF-β1, but these approaches do not reflect the patient situation where drugs can be given only after the diagnosis of the disease, years after its initiation. In two studies, PPAR-γ agonists were applied after bleomyin-induced lung injury in the mouse. Zeng et al. [ 27 ] added the PPARγ ligand asarinin 15–28 days after bleomycin administration, which reduced the severity of fibrosis. Speca et al. [ 22 ] applied GED-0507, a PPARγ modulator with strong anti-inflammatory effects, to mice on day 14 after bleomycin administration and reported resolution of fibrosis with 50% mortality rate. This post-treatment schedule reduced collagen deposition, but to a lesser extent than in the prevention approach used in the same study. Thus, we thought that a post-treatment with a combination of PPAR ligands may further increase the anti-fibrotic effect. Moreover, we aimed to use a human model and human cultured fibroblasts as the latter in vitro model better guaranties the drug availability and allows a selective (biochemical) analysis of changes in fibroblasts, the main players in fibrosis.

In this study, we investigated whether activation of each of the three PPARs alone or in various combinations influenced collagen synthesis and release of lung fibroblasts from control and IPF patients when given 24 h after TGF-β1, the endogenous stimulator of fibrosis. Moreover, we attempted to explore the mechanism of the anti-fibrotic effect of PPAR agonists by analyzing changes in members of matrix metalloproteinases (MMPs) [ 28 ], biogenesis and metabolism of peroxisomes [ 13 , 14 ], and the protein level and activity of catalase, the major anti-oxidative enzyme in peroxisomes [ 29 ] with the highest turnover numbers of all enzymes [ 30 ].

Study approval

Biospecimen collection (i.e. lung tissues and fibroblasts from organ donors) was approved by the Ethics Committee of the Justus Liebig University Giessen (Az58/15 and Az111/08, JLU).

Cell culture and drug treatment

Lung fibroblasts from control and IPF patients (Additional file: Table S1) and catalase-deficient fibroblast cell lines were cultured in Dulbecco´s Modified Eagle Medium (DMEM) with penicillin/streptomycin or puromycin, respectively. For the experiments, cells were serum-starved for 3 h, stimulated with vehicle or rhTGF-β1 for 24 h (except for Figs. 2 B, C, E, 3 B), followed by the addition of vehicle or drugs either alone or in combinations for another 24 h.

Knockdown of catalase in human lung fibroblasts

Knockdown of catalase was done with CAT siRNA using ScreenFectA transfection reagent. Stable catalase knockdown was achieved by transduction with pGIPZ-shCatalase and pGIPZ-non-silencing control lentivirus vectors as described earlier [ 31 ].

Overexpression of catalase in human lung fibroblasts

Transfection with catalase overexpression plasmid (pGL 4.14- Catalase ) and promoter reporter plasmids COL1A2 -luc and PPAR response element ( PPRE) -luc were done as described earlier [ 13 , 32 ]. Data from pRL-SV40 vector served to normalize results of the luciferase reporter plasmid.

Human TGF-β1 immunoassay and sircol collagen assay

The collected culture media of control and IPF fibroblasts were used for Sircol collagen assays and TGF-β1 ELISA assay according to the manufacturers´ instructions.

Measurements of catalase activity, hydrogen peroxide (H 2 O 2 ) production and cell proliferation

Determination of catalase activity with a redox dye assay kit based on the degradation of H 2 O 2 . H 2 O 2 produced by cultured cells was quantified using a fluorometric detection kit. The incorporation of BrdU into proliferating cells was detected with an ELISA kit. For all the aforementioned kits, we followed the manufacturers´ instructions.

Western blotting

Proteins of total cell lysates were separated on 10% SDS-PAGE gels and blotted on polyvinylidene difluoride membranes. Specific proteins were detected using primary and horseradish peroxidase (HRP)-labelled secondary antibodies followed by chemiluminescent detection of the HRP substrate. ImageJ was used for semi-quantitative analysis of signal intensities.

Immunofluorescence staining

Thin sections of paraffin-embedded lung tissues were incubated with primary and secondary fluorophore-labelled antibodies. Immunofluorescence images were acquired by confocal laser scanning microscopy.

Isolation of total RNA and RT-qPCR

Total RNA was isolated using RNAzol and mRNA levels were analyzed by RT-qPCR.

Targeted quantification of fatty acids

Arachidonic acid (AA), docosahexaenoic acid (DHA), and eicosapentaenoic acid (EPA) were analyzed in the culture medium by solid phase extraction and a targeted liquid chromatography tandem mass spectrometry (LC–MS/MS) approach as described previously [ 32 ].

Untargeted lipidomics

Lipids were extracted from cell lysates using a biphasic methyl- tert -butyl ether (MTBE) extraction protocol [ 33 ] and analyzed using an untargeted LC–MS/MS method as described previously [ 34 ].

Analysis was done using GraphPad Prism software. Data were expressed as means ± SEM. For comparisons between two groups, the F-test was applied to compare their variances followed by Mann–Whitney U test (unequal variances) or unpaired t -test (equal variances). For multiple comparisons, ANOVA was used with post-hoc Tukey´s multiple comparisons test. P values < 0.05 were considered as statistically significant.

Characterization of the fibrosis markers COL1 and α-SMA, as well as of PPARs in lung tissues and cultured fibroblasts from control and IPF patients

The fibrosis marker collagen type I (COL1) and myofibroblast marker α-smooth muscle actin (α-SMA) were first assessed in lung biopsy samples from control and IPF patients. Lung tissues from IPF patients showed comparatively higher levels of COL1 and α-SMA than those from control subjects (Fig.  1 A). Although increased mRNA levels of COL1A1 and ACTA2 were detected in cultured lung fibroblasts from IPF compared to control patients (Additional file: Fig. S1A, B), their protein levels and that of transforming growth factor-beta receptor 1 (TGFBR1) were higher in most cases in fibroblasts from control compared to IPF patients (Additional file: Fig. S1C, Table  1 ). Although unexpected at first glance, it is noteworthy that IPF lung tissue contains a much higher number of fibroblasts than controls. Probably, the higher number of fibroblasts in the lungs of IPF patients and to a minor extent their individual properties contribute to the differences in tissue pathology. Moreover, the reduced level of TGFBR1 in IPF fibroblasts suggests that they are less sensitive to TGF-β1 presumably due to their chronic exposure to the cytokine in vivo. Accordingly, analysis of extracellular collagen revealed no significant difference between control and IPF fibroblasts (Fig.  1 B, Additional file: Fig. S1D). IPF is characterized by elevated levels of TGF-β1 mRNA and protein in the lung tissues of patients [ 35 , 36 ]. Interestingly, the amount of released active TGF-β1 was higher in the culture media from control than IPF fibroblasts (Fig.  1 C). We demonstrated an anti-fibrotic role of peroxisomes in the progression of IPF via PPAR-α signaling [ 13 , 14 ]. Since all three PPARs crosstalk with each other [ 37 ], we next analyzed their protein levels in fibroblasts from control and IPF patients at basal conditions (no treatment in vitro). Collectively, IPF fibroblasts showed increased mRNA and protein levels of PPAR-α, but not of the ones of PPAR-β/δ and PPAR-γ compared to control fibroblasts (Fig.  1 D, Table  1 , Additional file: Fig. S1E).

Characterization of the fibrosis markers COL1 and α-SMA, and PPARs in lung tissue and cultured fibroblasts from control and IPF patients. A Lung tissue sections from control (left side) and IPF (right side) patients were incubated with antibodies to detect collagen (COL1, green) and α-SMA (red), and counterstained with DAPI (blue). Negative controls (NC) were done by omitting the primary antibody . B There was no difference in the release of collagen between fibroblasts from control and IPF patients. The release of collagen into culture media was measured using Sircol assay. Data represent 5 control and 5 IPF patients across six independent fibroblast cultures. C The release of active TGF-β1 is higher in control than in IPF fibroblasts. The amount of active human TGF-β1 was analyzed in the culture media of fibroblasts from 5 controls and 7 IPF patients by ELISA. D The protein levels of PPAR-α were higher in IPF compared to control fibroblasts, whereas there was no difference with regard to PPAR-β/δ and PPAR-γ. Cultured fibroblasts from 5 control and 7 IPF patients were collected and their protein levels were analyzed by Western blot analysis with GAPDH as reference protein

TGF-β1 induced a fibrotic response in fibroblasts from control and IPF patients. A TGF-β1 induced proliferation in control and IPF fibroblasts. Fibroblasts were serum-starved for 3 h and then incubated for 24 h with vehicle or TGF-β1. Thereafter, proliferation was analyzed using BrdU cell proliferation assay. B Treatment with different concentrations of TGF-β1 showed no difference between control and IPF fibroblasts with regard to the release of collagen into culture media. Control and IPF fibroblasts were serum-starved for 3 h and then treated with vehicle (Control) or 2.5, 5, 10 and 20 ng/ml TGF-β1 for 24 h. Cell culture media were collected and extracellular collagen was analyzed using Sircol assay. C , D TGF-β1 increased the level of intracellular COL1 in control and IPF fibroblasts in a time-dependent manner. Control and IPF fibroblasts were serum-starved for 3 h and then treated with vehicle or 5 ng/ml TGF-β1 for 12, 24, 36 and 48 h. Cell lysates were used to detect COL1 and α-SMA by Western blot analysis using GAPDH as reference protein ( C ). Data for a time period of 24 h from 5 control (patients A–E) and 5 IPF (patients F–J) patients is shown in ( D ). E TGF-β1 increased the protein level of PPAR-β/δ, whereas the ones of the other PPARs remained unchanged. Control and IPF fibroblasts were treated for 24, 48 and 72 h with TGF-β1 (5 ng/ml) or vehicle. Cell lysates were used for Western blot analysis of the PPARs using GAPDH as reference protein

Activation of PPAR-β/δ induced anti-fibrotic responses in TGF-β1-stimulated fibroblasts from control and IPF patients

As already noted, the number of fibroblasts in the lungs of IPF patients might be crucial for the disease progression. To confirm this, we analyzed the proliferation of vehicle- and TGF-β1-treated control and IPF fibroblasts since the cytokine was used to mimic part of the disease condition in vitro. As expected, TGF-β1 stimulated cell proliferation in control and IPF fibroblasts (Fig.  2 A). Next, we analyzed time-dependent changes in α-SMA and COL1 protein levels of control and IPF fibroblasts treated with TGF-β1. Control and IPF fibroblasts did not show differences after stimulation with different concentrations of TGF-β1 (2.5–20 ng/ml; Fig.  2 B) in the extracellular collagen released into the culture media. Though 2.5 ng/ml of TGF-β1 was already sufficient to reach the maximal effect for collagen values 24 h after treatment (Fig.  2 B), 5 ng/ml TGF-β1 was used to obtain maximal effects in all following experiments with distinct parameters. TGF-β1 increased intracellular COL1 and α-SMA protein levels from 12 to 48 h in control fibroblasts and from 24 h up to 48 h in IPF fibroblasts (Fig.  2 C). Moreover, the treatment with TGFβ-1 for 24 h in control and IPF fibroblasts from 10 different patients showed a homogenous and stable increase in the protein levels of COL1, but an inconsistent reaction in the case of α-SMA (Fig.  2 D, Table  1 ). To investigate the role of peroxisomes in IPF, their proliferation was induced using different PPAR ligands. Interestingly, TGF-β1 upregulated the protein level of PPAR-β/δ especially after 48 h of treatment (Fig.  2 E). Following 24 h TGF-β1 stimulation, treatment with PPAR-β/δ agonist alone or in combination with the two other members of the PPAR protein family inhibited the TGF-β1-mediated increase in COL1 and—to a lesser extent—α-SMA protein levels in control and IPF fibroblasts (Fig.  3 A). As already noted, anti-fibrotic properties of PPAR-γ have been reported in the past. In our study, the post-treatment with a PPAR-β/δ agonist (GW0742) alone or combined with a PPAR-γ agonist (rosiglitazone) strongly decreased the amount of TGF-β1-mediated increase in intracellular COL1 (Fig.  3 A–C) by affecting COL1A2 promotor activity (Fig.  3 D) as well as extracellular collagen (Fig.  3 E) in both, fibroblasts from control and IPF patients. Lesser effects were observed in the case of activation of PPAR-γ alone (Fig.  3 A, C, E). The decrease in the amount of COL1 as a result of the dual treatment of PPAR-β/δ and PPAR-γ agonists was stable over time (Fig.  3 B) and between patients (Fig.  3 C). Furthermore, the anti-fibrotic effects of a combined activation of PPAR-β/δ and PPAR-γ were blocked in the presence of PPAR-β/δ (GSK0660) and PPAR-γ (GW9662) antagonists (Fig.  3 F). In addition, we thought to use the compound STK 648389 (ZINC ID: 31,775,965), a putative dual agonist for PPAR-β/δ and PPAR-γ. However, analysis of the STK 648389 for its effect on collagen showed adverse effects and even increased extracellular collagen levels released by control and IPF fibroblasts after TGF-β1 exposure (Additional file: Fig. S2). Altogether, these findings suggest that although TGF-β1 increases the PPAR-β/δ protein as a protective adaptive mechanism, endogenous PPAR-β/δ activating ligands are probably diminished to prevent fibrosis in patients.

Activation of PPAR-β/δ induced anti-fibrotic responses in TGF-β1-stimulated fibroblasts. A – C , E Control and IPF fibroblasts were serum-starved for 3 h, treated with TGF-β1 (5 ng/ml) for 24 h, followed by the addition of the PPAR-α agonist WY14643 (100 μM, α; A ), PPAR-β/δ agonist GW0742 (10 μM, β; A–C , E ), and PPAR-γ agonist rosiglitazone (10 μM, γ; A – C , E ) either for 24 h ( A , C , E) or different time periods (12, 24, 36 and 48 h; B ). A PPAR-β/δ activation reversed TGF-β1-induced increase in COL1. Cell lysates were used to detect COL1 and α-SMA by Western blot analysis using GAPDH as reference protein. B , C Reverse of fibrosis phenotype by PPAR-β/δ and PPAR-γ activation was stable for up to 48 h. Cell lysates at 12 to 24 h ( B ) and 48 h from two other control and IPF patients ( C ) were used for Western blot analysis using β-actin (β-ACTIN) as reference protein. D Combined activation of PPAR-β/δ and PPAR-γ abolished TGF-β1-induced increase in COL1A2 promoter activity. IPF fibroblasts were transfected with a plasmid containing the luciferase firefly reporter gene adjacent to COL1A2 promoter and Renilla luciferase as second reporter for normalization. At 72 h after transfection, cells were treated with vehicle (Vector) or TGF-β1 (5 ng/ml) for 24 h followed by the addition of the PPAR-β/δ agonist GW0742 (10 μM, β) combined with the PPAR-γ agonist rosiglitazone (10 μM, γ) or vehicle for another 24 h. Cells were lysed and collected for dual luciferase activity measurements. E Ligand activation of PPAR-β/δ together with PPAR-γ strongly decreased the release of collagen produced by TGF-β1-stimulation in control and IPF fibroblasts. Culture media were collected and extracellular collagen was analyzed using Sircol assay. F Combined activation of PPAR-β/δ and PPAR-γ decreased TGF-β1-stimulated release of collagen by control and IPF fibroblasts—this effect was blocked using the respective antagonists. Cells were serum-starved for 3 h, stimulated with vehicle (Control) or TGF-β1 (5 ng/ml) for 24 h, followed by the addition of the PPAR-β/δ agonist GW0742 (10 μM, β) and PPAR-γ agonist rosiglitazone (10 μM, γ) either combined with vehicle or the PPAR-β/δ antagonist GSK0660 (10 nM, β ant) and PPAR-γ antagonist GW9662 (10 μM, γ ant) for another 24 h. Culture media were collected and extracellular collagen was analyzed by Sircol assay

PPAR-β/δ triggers anti-fibrotic responses by activating MMP-1 in control and IPF fibroblasts

Extracellular collagen is degraded by proteinases, e.g. MMPs. The mRNA levels of selected MMPs in control fibroblasts at basal condition (without treatment) were measured, showing the highest value for MMP1 in comparison to the lower mRNA values for MMP2 , MMP3 , MMP10 , and MMP16 (Fig.  4 A). Interestingly, the mRNA level of MMP7 which is associated with disease severity [ 28 ] was below detectable levels in our samples of control and IPF fibroblasts (ct values > 35). Comparing the mRNA levels between control and IPF fibroblasts, no differences were observed in the case of MMP1 and MMP16 (Fig.  4 B, F), but higher levels were found for MMP2 , MMP3 and MMP10 (Fig.  4 C–E). Individual mRNA values for MMP1 , but also for MMP3 and MMP10 , varied strongly within the IPF sample group (Fig.  4 B, D, E). Due to the much higher mRNA levels for MMP1 compared to the other MMPs (Fig.  4 A), we analyzed MMP-1 protein as the dominant enzyme for collagen degradation in subsequent experiments. As expected, the protein level of active MMP-1 was reduced by TGF-β1 and restored in the presence of PPAR-β/δ agonist alone or in combination with PPAR-α or PPAR-γ agonists (Fig.  4 G). This suggests that PPAR-β/δ might be a key regulator of the protein level of active MMP-1. Therefore, we analyzed the effect of the PPAR-β/δ agonist in TGF-β1-stimulated fibroblasts at the mRNA levels of all detectable MMPs . The mRNA levels of MMP1 in IPF fibroblast were increased (> fivefold) by the PPAR-β/δ agonist in comparison to TGF-β1 stimulation alone (Fig.  4 H). The MMP16 mRNA levels were elevated > fivefold in both types of fibroblasts and that of MMP10 about threefold in control fibroblasts only (Additional file: Fig. S3). To explore the anti-fibrotic potential of increased levels of MMP s , we used a broad-spectrum inhibitor for MMPs, primarily influencing the amount of extracellular collagen. Simultaneous treatment with the MMP inhibitor and PPAR-β/δ agonist after TGF-β1 stimulation increased extracellular collagen in the culture media released by control fibroblasts, but not in the case of IPF fibroblasts (Fig.  4 I). Since the MMP inhibitor only partly blocked the effect of the PPAR-β/δ agonist, we speculate that activated PPAR-β/δ also regulates other proteins involved in fibrosis attenuation.

PPAR-β/δ triggers anti-fibrotic responses by activating MMP-1 in control and IPF fibroblasts. A The transcript of MMP1 is the highest among the different MMPs in control fibroblasts. Analysis of MMP1, MMP2, MMP3, MMP10 and MMP16 of control fibroblasts was done using isolated total RNA and RT-qPCR with HPRT1 as reference gene. B – F Comparative gene expression profile of MMPs was done by RT-qPCR with HPRT1 as reference gene. G PPAR-β/δ attenuated TGF-β1-induced decrease in the amount of active MMP-1. Control and IPF fibroblasts were serum-starved for 3 h, treated with vehicle or TGF-β1 (5 ng/ml) for 24 h, followed by the addition of the PPAR-α agonist WY14643 (100 μM, α), PPAR-β/δ agonist GW0742 (10 μM, β), and PPAR-γ agonist rosiglitazone (10 μM, γ) as well as various combinations thereof for another 24 h. Cell lysates were used to detect active MMP-1 by Western blot analysis using β-actin (β-ACTIN) as reference protein. H Ligand activation of PPAR-β/δ strongly increased the mRNA level of MMP1 in TGF-β1-treated control and IPF fibroblasts. Cells were serum-starved, treated with vehicle (Control) or TGF-β1 (5 ng/ml) for 24 h followed by the addition of the PPAR-β/δ agonist GW0742 (10 μM, β) or vehicle for another 24 h. The mRNA levels were measured by RT-qPCR with HPRT1 as reference gene. I Inhibition of MMPs increased TGF-β1-induced release of collagen. Control and IPF fibroblasts were serum-starved for 3 h, treated with vehicle or TGF-β1 (5 ng/ml) for 24 h, followed by the addition of the PPAR-β/δ agonist GW0742 (10 μM, β) and MMP inhibitor (MMP inh., 4-aminobenzoyl-Gly-Pro-D-Leu-D-Ala hydroxamic acid, 20 μM) for another 24 h. The release of collagen into the culture media was measured by Sircol assay

Activation of PPAR-β/δ and PPAR-γ in TGF-β1-treated fibroblasts increased peroxisomal biogenesis and lipid metabolism, and the inhibited fibrotic response

Previously, we showed that pretreatment with PPAR-α agonists could inhibit fibrosis phenotypes [ 13 , 14 ]. In the present study, we treated control and IPF fibroblasts with TGF-β1 before the addition of agonists of all three PPARs, an experimental setup that more accurately recapitulates the clinical setting. We first investigated the mRNA levels of several peroxisomal genes involved in the organelle biogenesis ( PEX13, PEX14 ), plasmalogen synthesis ( AGPS, GNPAT ), and fatty acid β-oxidation ( ACOX1, ACAA1 ) in control and IPF fibroblasts. The mRNA levels of PEX13 , ACOX1 and AGPS were higher in IPF compared to control fibroblasts, whereas those of PEX14, ACAA1 and GNPAT were not significantly different (Additional file: Fig. S4A–F). Next, we stimulated peroxisomal proliferation with different PPAR agonists (alone or in combination) in TGF-β1-treated control and IPF fibroblasts. Combined activation of PPAR-β/δ and PPAR-γ increased mRNA (Additional file: Fig. S4G) and protein levels (Fig.  5 A) of PEX13 in TGF-β1-stimulated control and IPF fibroblasts compared to TGF-β1 treatment only. Since the combined activation of PPAR-β/δ and PPAR-γ reversed the TGF-β1-induced trans-differentiation of fibroblasts into myofibroblasts (as indicated by changes in the level of α-SMA, Fig.  3 A–C), decreased the protein level of COL1 (Fig.  3 A–C) and increased PEX13 (Fig.  5 A), we focused on these two PPAR agonists in the following experiments. First, the intracellular lipidomic profile was assessed in control and IPF fibroblasts to ascertain possible differences in the lipid metabolism under basal conditions and after treatments with TGF-β1 alone and PPAR-β/δ and PPAR-γ agonists. In total, 1003 lipid ion species covering 5 major lipid categories (glycerophospholipids, sphingolipids, glycerolipids, fatty acyls, and sterols) belonging to 22 lipid classes were identified based on high mass accuracy (5 ppm) and their fragmentation patterns (Additional file: Fig. S5A). Basal levels of all classes of lipids analyzed were lower in IPF fibroblasts except for the triglycerides (TG; Fig.  5 B). TGF-β1 decreased the levels of phosphatidylcholine (PC), phosphatidylethanolamine (PE), sphingomyelin (SM) and TG in IPF fibroblasts. The levels of PC, SM and TG were partially restored by a post-treatment with PPAR-β/δ and PPAR-γ agonists (Fig.  5 B). Furthermore, activation of PPAR-β/δ and PPAR-γ strongly increased the synthesis of peroxisome-derived AA, DHA, and EPA (Fig.  5 C), which are all endogenous activators of PPARs. In the absence of TGF-β1, PPAR agonists either increased or decreased the levels of PC in control and IPF fibroblasts (Additional file: Fig. S5B) and increased the levels of AA, DHA and EPA with PPAR-γ exhibiting the strongest effect on DHA (Additional file: Fig. S5C). This suggests that, the PPAR-γ agonist was the driving factor for the increase and release of AA, DHA and EPA in fibroblasts treated with TGF-β1 followed by combined PPAR-β/δ and PPAR-γ agonists treatment (Fig.  5 C). Collectively, activation of PPAR-β/δ and PPAR-γ potentially regulates the fibrosis phenotype by modulating peroxisomal lipid metabolism, but differently in control and IPF fibroblasts.

Activation of PPAR-β/δ and PPAR-γ in TGF-β1-treated fibroblasts increased peroxisomal biogenesis and lipid metabolism. A - C Control and IPF fibroblasts were serum-starved for 3 h, treated with vehicle or TGF-β1 (5 ng/ml) for 24 h, followed by the addition of the PPAR-α agonist WY14643 (100 μM, α; A ), PPAR-β/δ agonist GW0742 (10 μM, β; A – C ), and PPAR-γ agonist rosiglitazone (10 μM, γ; A-C ) as well as various combinations thereof for another 24 h. A Activation of PPAR-β/δ and PPAR-γ reversed TGF-β1-induced decrease in the protein levels of the peroxisomal biogenesis protein PEX13. Cell lysates were used for Western blot analysis of PEX13 using GAPDH as reference protein. B Heatmap of the lipidomic profile of control and IPF fibroblasts. Cells were collected in PBS for lipid analysis using LC–MS/MS. C Activation of PPAR-β/δ and PPAR-γ increased the synthesis of endogenous activators of these receptors in line with a positive feedback loop. Fibroblasts from control and IPF patients were serum-starved for 3 h, treated with vehicle (Control) or TGF-β1 (5 ng/ml) for 24 h, followed by the addition of vehicle or the PPAR-β/δ agonist GW0742 (10 μM, β) combined with the PPAR-γ agonist rosiglitazone (10 μM, γ) for another 24 h. The releases of AA, DHA, and EPA were analyzed in the culture media by LC–MS/MS

Activation of PPAR-β/δ in combination with PPAR-γ restored TGF-β1-induced decrease in catalase mRNA and protein levels

Though not significant, TGF-β1 decreased CAT mRNA level in control and IPF fibroblasts, which was restored by the combined activation of PPAR-β/δ and PPAR-γ (Additional file: Fig. S4G). We therefore speculated that this anti-oxidative enzyme might be involved in regulation of fibrogenesis. We first analyzed catalase and glutathione peroxidase (GPX)1/2 in human lung tissue samples. The protein level of catalase was markedly decreased in alveolar epithelial type II cells in the lungs of IPF compared to control patients (Fig.  6 A), whereas that of GPX1/2 was increased (Fig.  6 B), probably to compensate catalase deficiency. Moreover, we detected a gradual decrease in catalase protein level in mouse lungs after bleomycin-induced fibrosis, remarkably from day 14 after treatment (Additional file: Fig. S6A). When we analyzed the fibroblasts from control and IPF patients, we found no differences in the mRNA levels of CAT and GPX1/2 (Additional file: Fig. S6B, C). Protein level of catalase was lower in IPF compared to control fibroblasts (isolated from 5 patients each, Fig.  6 C, Table  1 ). Apart from catalase and GPX1/2, peroxiredoxins (PRDXs) were measured as they also support the anti-oxidant defense system. The mRNA levels of different peroxiredoxin family members varied strongly (Additional file: Fig. S6D) with PRDX6 showing the highest and PRXD2 and PRXD3 the lowest gene expression levels. Only the mRNA levels of PRDX4 and PRDX6 were significantly higher in IPF compared to control fibroblasts (Additional file: Fig. S6E–J). To confirm the regulatory effects of TGF-β1 on catalase, we treated control and IPF fibroblasts with TGF-β1 at various concentrations. Increasing concentrations of TGF-β1 gradually decreased the protein level of catalase in both fibroblast groups (Fig.  6 D). Catalase activity was reduced by TGF-β1 in control and IPF fibroblasts, but not in the same manner since IPF fibroblasts were less sensitive towards lower concentrations of TGF-β1 (2.5 and 5 ng/ml; Fig.  6 E). Activation of PPAR-γ increased the protein level of catalase in the absence of TGF-β1 (Additional file: Fig. S6K) and reversed the TGF-β1-induced decrease in catalase in control fibroblasts (Fig.  6 F). The level of catalase increased in both groups when PPAR-β/δ and PPAR-γ were activated 24 h after TGF-β1 treatment (Fig.  6 F), but not when added together with TGF-β1 (Fig.  6 G).

TGF-β1 caused a decrease in catalase mRNA and protein levels. A , B The immunoreactivity of catalase was lower, and that of GPX1/2 higher in IPF (right) compared to control (left) lung tissues. Immunofluorescence staining was performed using antibodies to detect catalase ( A , red) and GPX1/2 ( B , red) and DAPI to counterstain nuclei. C The protein level of catalase is lower in IPF than in control fibroblasts. Cell lysates of fibroblasts from 5 control and 5 IPF patients were used for Western blot analysis of catalase (CAT) with β-actin (β-ACTIN) as reference protein. D TGF-β1 decreased catalase protein levels in control and IPF fibroblasts. Cells were serum-starved for 3 h, and treated with various concentrations of TGF-β1 or vehicle for 48 h. Cell lysates were used for Western blot analysis of catalase with GAPDH as reference protein. E – G Activation of PPAR-β/δ in combination with PPAR-γ restored TGF-β1-induced decrease in catalase protein levels and activity. E TGF-β1 decreased catalase activity in control and IPF fibroblasts. Cells were serum-starved for 3 h, and treated with vehicle (Control) or various concentrations of TGF-β1 for 12 h. Cell lysates were used for measuring catalase activity. F , G Activation of PPAR-β/δ in combination with PPAR-γ inhibited TGF-β1-induced decrease in catalase protein levels in control and IPF fibroblasts. Cells were serum-starved for 3 h, stimulated with vehicle ( F , G ) or TGF-β1 (5 ng/ml, F , G ) or for 24 h, followed by the addition of the PPAR-β/δ agonist GW0742 (10 μM, β) and the PPAR-γ agonist rosiglitazone (10 μM, γ) for another 24 h ( F ). In ( G ), the PPAR agonists were added together with TGF-β1 for 48 h. Cell lysates were used to detect catalase (CAT) by Western blot analysis using α-tubulin (α-TUB) as reference protein

Catalase contributes to collagen reduction in pulmonary fibrosis

To confirm the anti-fibrotic role of catalase in IPF, we intended to generate stable catalase-deficient fibroblast cell lines by RNAi using two independent shRNAs against catalase (CAT sh1 RNA and CAT sh2 RNA). Knockdown efficiency of catalase was high and stable in control fibroblasts, whereas IPF fibroblasts died after a few passages probably because the catalase protein level was already low prior to shRNA transduction (see Fig.  6 C) and a further decrease in this protein was detrimental. Successful reduction of catalase is shown on the protein (Fig.  7 A) and activity (Fig.  7 B) levels, resulting in an increase in H 2 O 2 concentration (Fig.  7 C). The decrease in catalase protein in control fibroblasts expressing either of the two independent catalase shRNAs was accompanied with increased extracellular collagen (Fig.  7 D) and intracellular COL1 (Fig.  7 A) levels. Using siRNA technology, a transient catalase knockdown was achieved in control and IPF fibroblasts (Additional file: Fig. S7A). In IPF fibroblasts, we detected higher levels of collagen released into the culture medium compared to those transfected with scrambled control siRNA (Additional file: Fig. S7B). Moreover, catalase overexpression in control and IPF fibroblasts decreased COL1 and α-SMA protein levels even after TGF-β1 stimulation (Fig.  7 E). Lastly, we analyzed whether the reduction in collagen synthesis by activation of PPAR-β/δ and PPAR-γ depends on catalase activity. In both fibroblast cell lines, the reduction in collagen by the PPAR-γ agonist, but not by PPAR-β/δ was reversed in the presence of 3-amino-1,2,4-triazole (AT, Fig.  7 F, lane 5 versus lanes 7 and 8). Interestingly, AT inhibited the beneficial effect of a combined activation of PPAR-β/δ and PPAR-γ in control, but not in IPF fibroblasts (Fig.  7 F, lane 5 versus lane 6). We suggest that during TGF-β1 treatment either the protein level, sensitivity or signaling of PPAR-β/δ dominates in IPF and that of PPAR-γ in control fibroblasts with regard to catalase protein content and its activity.

Catalase contributes to collagen reduction in pulmonary fibrosis. A , B Stable knockdown of catalase decreased catalase protein and activity. Cell lines transfected with catalase shRNA (CAT sh1, CAT sh2) were serum-starved for 3 h. Cell lysates were used for measuring catalase (CAT), COL1 and α-SMA protein levels by Western blot analysis using GAPDH as reference protein ( A ) and catalase activity by catalase activity assay kit ( B ). C , D Stable knockdown of catalase increased the cellular H 2 O 2 production and extracellular collagen levels. Culture media from catalase-deficient (CAT sh1, CAT sh2) and mock-transfected (CAT sc) control fibroblasts were used to detect the release of H 2 O 2 using the hydrogen peroxide assay ( C ) and of extracellular collagen by Sircol assay ( D ). E Overexpression of catalase decreased the protein level of COL1 in control and IPF fibroblasts under basal condition (no treatment) and after TGF-β1 treatment. Control and IPF fibroblasts were transfected with pGL 4.14-Catalase (CAT overexpr.) or a mock vector for 48 h, followed by the addition of vehicle or TGF-β1 (5 ng/ml) for another 48 h. Cell lysates were analyzed for catalase (CAT), α-SMA, and COL1 protein levels by Western blot analysis using GAPDH as reference protein. F The catalase activity inhibitor AT does not increase COL1 in control and IPF fibroblasts. Cells were serum-starved for 3 h, treated with vehicle or TGF-β1 (5 ng/ml) or for 24 h, followed by the addition of the PPAR-β/δ agonist GW0742 (10 μM, β), the PPAR-γ agonist rosiglitazone (10 μM, γ) and AT (25 µM) as well as various combinations thereof for another 24 h. Cell lysates were used to analyze catalase (CAT), COL1, and α-SMA protein levels by Western blot analysis using GAPDH as reference protein

In the present study, cultured human lung fibroblasts were treated with TGF-β1 to mimic fibrosis and were then analyzed to evaluate the role of PPARs during disease progression. Human lung tissue samples from control and IPF patients (Figs. 1 , 6 ) were used in parallel. Traditional animal models of experimental lung fibrosis were carried out by radiation or intratracheal administration of asbestosis fibers and silica, but the latter two induce rather asbestosis and silicosis than fibrosis [ 38 ]. Since high levels of TGF-β1 were shown to initiate and support fibrosis [ 35 , 36 ], a rat model of adenoviral overexpression of TGF-β1 has been established, however, the adenovirus vector itself already induced fibrosis [ 38 , 39 ]. Most commonly, mice were treated with bleomycin which induced a rapid fibrosis within 2–4 weeks via intra-tracheal instillation or 4–12 weeks by systemic administration [ 38 ]. The injury first triggers an inflammatory response which leads to wound healing. The infiltrating immune cells produce pro-fibrotic cytokines, e.g. TGF-β1, which stimulates fibroblast-to-myofibroblast transition. A dysregulated wound healing process could moreover lead to excessive deposition of ECM and finally resulting in fibrosis. However, this mouse model does not represent all aspects of the histopathological phenotype of the disease as observed in humans, for example, honeycomb pattern, thick scars at the alveolar region and fibroblastic foci [ 40 , 41 , 42 ], probably because these features take time to develop in humans. In addition, bleomycin-induced fibrosis is often reversible and contains a strong inflammatory component in the beginning which is not true for the disease in humans [ 38 ].

To mimic fibrosis in vitro, pro-fibrotic cytokines were added to cultured lung fibroblasts such as platelet-derived growth factor, connective tissue growth factor, interleukin-1β, tumor necrosis factor-α (TNF-α) and TGF-β1 [ 43 ]. Interleukin-1β and growth factors induced a marked inflammation and fibrosis with aberrant wound healing, TNF-α induced a strong inflammation and mild fibrosis, and TGF-β1 solely caused minor inflammation together with a marked fibrosis. Thus, TGF-β1-induced changes reflected the pathogenesis found in human IPF patients and was therefore used in our experiments. In vitro models, as an advantage, allow drug treatments to block TGF-β1-induced fibrosis signaling pathways and cell transfection to knockdown proteins of interest, which is difficult to establish in vivo. On the other hand, analysis of cultured lung fibroblasts neglects the in vivo situation where they interact with themselves and other cell types such as alveolar epithelial cells type I and type II, endothelial cells and macrophages. Interestingly, alveolar epithelial type II cells restrict the number of fibroblasts [ 44 ], and thus, control fibroblasts in vitro (and in the absence of alveolar epithelial type II cells) might re-start proliferation together with an increased collagen synthesis reaching similar levels as found in IPF fibroblasts. Moreover, TGF-β1 in IPF is mainly produced by macrophages [ 45 ]. Therefore, TGF-β1 (at least 5 ng/ml) had to be added to induce fibrosis in cultures of pure fibroblasts (which secrete 0.15 ng/ml TGF-β1, Fig.  1 C). In this study, tissues and an in vitro model established with fibroblasts from control and IPF patients were used in parallel.

To study the pathophysiology of lung fibrosis, we measured the two fibrosis markers associated with IPF such as collagen [ 46 , 47 , 48 ], and α-SMA, although the latter has been currently debated as a sole marker for studying fibrosis [ 49 ] as its expression doesn´t mean that a cell produces high amounts of collagen [ 50 ]. Interestingly, IPF is characterized by excessive accumulation of collagen-rich ECM produced by activated fibroblasts and myofibroblasts [ 51 , 52 ]; thus the degree of fibrosis is strongly dependent on their number and proliferation. Our data showed that fibroblasts from control and IPF patients were not different with regard to (1) the intracellular level of α-SMA and ω-fatty acids such as AA, DHA and EPA; (2) the release of collagen into the extracellular space; (3) the activity of collagen-degrading enzyme MMP-1; and (4) cell proliferation rate under basal conditions. Instead, fibroblasts from IPF compared to control patients showed significantly lower protein levels of PEX13, catalase, and of the TGFBR1 and are thus less sensitive towards TGF-β1. They secrete less active TGF-β1 into the culture medium. Contrarily, higher protein levels were found in IPF compared to control fibroblasts for intracellular GPX1/2 and PPAR-α. For IPF, the number and proliferation of fibroblasts/myofibroblasts are directly and the level of catalase indirectly related to the disease progression. Nonetheless, individual fibroblasts from control and IPF patients differ strongly even within the group (Figs. 1 D, 6 C , Additional file: Fig. S1A–C). This phenomenon might probably be due to the recently reported spatial heterogeneity of fibroblasts in fibrotic foci containing multiple subtypes such as lipofibroblasts, myofibroblasts, EBF1 + fibroblasts, intermediate fibroblasts, and mesothelial cells, all expressing different amounts of collagen under healthy conditions and during IPF progression [ 50 ]. In addition, the patients differ either with regard to the disease (acute exacerbation versus chronic stages, slow versus rapid decline of lung function), to co-morbidities (hypertension, viral infection, chronic aspiration of gastric content) or to other trigger factors such as age (age-related mitochondrial and peroxisomal dysfunction leading to oxidative stress), environmental exposures, smoking, and genetic factors [ 53 ]. Interestingly, differences between patients in our experiments were mainly observed for protein levels of PPAR-α (Fig.  1 D), PPAR-γ (Fig.  1 D), MMP-1 (Additional file: Fig. S1C) and catalase (Fig.  6 C), whereas the protein levels of PPAR-β/δ (Fig.  1 D), catalase activity (Fig.  6 E), the level of intracellular and secreted collagen with and without TGF-β1 (Figs.  1 B,   2 B–D, Additional file: Fig. S1D) as well as the collagen-reducing effect of a combined treatment with PPAR-β/δ and PPAR-γ agonists (Fig.  3 A–F) were less variable. This gives hope that the observed beneficial effect of PPAR-β/δ and PPAR-γ agonists is applicable to a broad spectrum of IPF patients. However, the strong heterogeneity of the target, namely the fibroblasts of IPF, but also of control patients, will limit the global use of any drug for IPF. Clinical trials discriminating between different subsets of patients may help to find the right drug in this regard.

We demonstrated that among the three PPARs, PPAR-β/δ might be a strong target for lung fibrosis resolution compared to PPAR-α (minor effect) and PPAR-γ (additive effect with PPAR-β/δ under these experimental conditions, Table  2 ). Focusing first on fibrosis pathways, we detected no differences between control and IPF fibroblasts with regard to the synthesis and release of collagen as well as gene expression and activity of MMP-1 (the dominant MMP, Fig.  4 A) either when treated or untreated with TGF-β1, and PPAR-β and PPAR-γ agonists . However, MMPs differ between the diverse lung cell types such as alveolar epithelial type I and type II cells, alveolar macrophages and endothelial cells [ 54 , 55 ]. In addition, MMP-1, -2, -3, -7, -13, -14, and -19, exhibit either anti- or pro-fibrotic [ 28 ] activities. MMP-2, as an example for the latter one, cleaves elastin which is deleterious for the lung. Interestingly, PPAR-β stimulation decreased the secretion of MMP-2 and increased the elastin level in human skin fibroblasts [ 56 ].

Next, we observed a TGF-β1-induced decrease in the peroxisomal biogenesis protein PEX13 which is reversed by stimulation of PPAR-γ. This was accompanied by changes in peroxisomal lipid metabolism, e.g. TGF-β1 increased the level of phosphatidylcholine in control, but decreased it in IPF fibroblasts with no additional effects of the PPAR drugs. The levels of AA, DHA and EPA were not significantly changed by TGF-β1, but increased strongly upon treatment with the PPAR-γ agonist. Metabolites from AA oxidation have been described to mediate inflammatory responses, and DHA is known to be anti-inflammatory [ 57 , 58 ]. A balance between the fatty acids will essentially determine the direction of the drug interventions. The production of DHA was more than that of AA in control and IPF fibroblasts following PPAR-γ activation, whereas the activation of PPAR-β/δ increased levels of AA to a higher extent compared to DHA in control and IPF fibroblasts. However, the strong anti-fibrotic effects of PPAR-β/δ support the combined activation of both receptors during treatments. Thus, with regard to peroxisomes, PPAR-β/δ and PPAR-γ agonists increased the peroxisomal biogenesis protein PEX13, as well as peroxisome lipid metabolism, and the resulting metabolites may further activate PPARs, establishing a positive activation loop [ 59 , 60 ].

Furthermore, the TGF-β1-induced decrease in the protein level and activity of catalase was reversed upon stimulation of PPAR-γ and PPAR-β/δ. Interestingly, in control fibroblasts the anti-fibrotic effect is mediated mainly via the maintenance of catalase protein through a reactive oxygen species (ROS)-dependent stimulation of PPAR-γ, because the effect is blocked by the specific catalase inhibitor AT in the combined treatment group by sustaining catalase levels. In IPF fibroblasts, the anti-fibrotic effect is mainly caused by a combined activation of PPAR-β/δ and PPAR-γ. The collagen-reducing effect is not inhibited by AT and thus ROS-independent. A decreased catalase level has been found in lung homogenates (and especially in the bronchial epithelium) of patients with IPF [ 61 ]. In acatalasemic mice, bleomycin induced a much higher invasion of pro-inflammatory cells together with increased levels of TGF-β1 and collagen and thus a higher degree of fibrosis [ 29 ], suggesting a beneficial role of high catalase levels in IPF disease progression. Interestingly, catalase (low affinity, high turnover) together with PRDX1 and PRDX5 (high affinity, low turnover), breakdown H 2 O 2 generated by multiple pathways inside peroxisomes. While catalase is crucial for safeguarding the organelle at excessive H 2 O 2 , PRDX1 and PRDX5 function as a redox-regulator in cell signaling and H 2 O 2 redox relay factor at low levels of H 2 O 2 , respectively [ 62 ]. In addition, catalase impedes ROS-induced inhibition of peroxisomal β-oxidation including the synthesis of the anti-inflammatory DHA [ 61 ]. With regard to PPARs, the catalase gene promotor region contains PPRE binding sites, e.g. for PPAR-γ (located at nucleotides − 1027 to − 1014; [ 63 ]) and an additional PPAR-γ binding site in humans only (located at nucleotides − 11,710 to − 11,698, [ 64 ]). Activation of PPAR-γ [ 23 ], but also of PPAR-β/δ (at the direct repeat 1 response element, [ 65 ]) increased catalase protein levels [ 65 , 66 ]. We assume that the observed increase in catalase protein in our experiments by PPAR-β/δ and PPAR-γ was similarly due to an induction of the catalase promotor activity. The additive effect by the combined treatment with PPAR-β/δ and PPAR-γ ligands in IPF fibroblasts suggests an importance of the additional human-specific PPRE binding sites and demonstrates that human models are required to analyze the role of PPARs in fibrosis.

We would like to emphasize that in contrast to most of the previous publications we performed a post-treatment (to mimic the clinical situation) with a combination of PPAR-β/δ and PPAR-γ agonists to reverse the TGF-β1-induced fibrotic phenotype of IPF fibroblasts. It is well known that activated PPAR-γ alone is potentially anti-fibrotic [ 17 , 18 , 19 , 20 ]. With regard to PPAR-β/δ, to the best of our knowledge, only one review described an inhibition of the proliferation of normal human lung fibroblasts by its stimulation [ 26 ]. The question arises how an activation of PPAR-β/δ can support PPAR-γ or vice-versa. One possibility is that stimulation of one PPAR might increase the protein level of itself and of the other receptors. For example, agonists for PPAR-α and PPAR-β/δ, but not PPAR-γ, have been shown to increase the protein levels of PPAR-β/δ and PPAR-γ in osteoblasts [ 37 ]. Thus, especially PPAR-β/δ stimulation can end up in a positive activation loop as it increased its own as well as the PPAR-γ receptor [ 60 ]. This offers the possibility for a post-treatment schedule starting with the PPAR-β/δ agonist to increase PPAR-γ levels so that the later given PPAR-γ agonist can work more efficiently. Interestingly, after 48 h treatment with TGF-β1, we observed increases in the protein levels of PPAR-γ and PPAR-β/δ in control and IPF fibroblasts although with varying degrees (Fig.  2 E, Table  1 ). This might explain why the post-treatment with PPAR-β/δ and PPAR-γ agonists is even more beneficial than direct treatment. Moreover, we demonstrated that the test compound STK 648389 (ZINC ID: 31,775,965), which has been suggested to be a dual PPAR-β/δ/PPAR-γ agonist by structure-based virtual screening [ 67 ], did not elicit anti-fibrotic effects (Additional file: Fig. S2). We hypothesized that the dual agonist (which is a single molecule) might be less specific for both receptors than the respective individual agonists and must be applied at a higher concentration which could induce more side effects in lung fibroblasts. Indeed, luciferase transactivation assays have shown EC50 values of 132 µM for PPAR-β/δ and 18 µM for PPAR-γ [ 67 ], and thus STK 648389 activated PPAR-γ only (see Fig.  3 E showing no reduction of the extracellular collagen using 10 µM of the specific PPARγ agonist troglitazone).

In summary, combined activation of PPAR-β/δ and PPAR-γ exerts strong anti-fibrotic effects. Catalase, which is decreased during treatment with TGF-β1, is inverse proportionally involved in collagen production. Catalase protein level and activity can be increased by stimulation of PPAR-β/δ and PPAR-γ in control and IPF human lung fibroblasts. For IPF patients (to refer to the clinical situation), the most beneficial anti-fibrotic effects could possibly be achieved by a combined local treatment with PPAR-β/δ and PPAR-γ agonists via aerosol inhalation.

Availability of data and materials

Raw data of the lipid analyses are available upon request to the corresponding author.

Abbreviations

α-Smooth muscle actin

Arachidonic acid

3-Amino-1,2,4-triazole

Collagen type 1

Dulbecco’s Modified Eagle Medium

Docosahexaenoic acid

Extracellular matrix

Eicosapentaenoic acid

Glutathione peroxidase 1/2

Horse radish peroxidase

• Idiopathic pulmonary fibrosis

Liquid chromatography tandem mass spectrometry

Matrix metalloproteinase

Methyl- tert -butyl ether

Phosphatidylcholine

Phosphatidylethanolamine

Peroxisomal biogenesis protein, peroxin

Peroxisome proliferator-activated receptor

Peroxiredoxin

PPAR response element

Reactive oxygen species

Quantitative reverse transcription polymerase chain reaction

Sphingomyelin

Solid phase extraction

Triglycerides

Transforming growth factor-beta 1

Transforming growth factor-beta receptor 1

Tumor necrosis factor-α

Maher TM. Interstitial lung disease: a review. JAMA. 2024;331:1655–65. https://doi.org/10.1001/jama.2024.3669 .

Article   PubMed   CAS   Google Scholar  

Cheresh P, Kim SJ, Tulasiram S, Kamp DW. Oxidative stress and pulmonary fibrosis. Biochim Biophys Acta. 2013;1832:1028–40. https://doi.org/10.1016/j.bbadis.2012.11.021 .

Richeldi L, du Bois RM, Raghu G, Azuma A, Brown KK, Costabel U, Cottin V, Flaherty KR, Hansell DM, Inoue Y, Kim DS, Kolb M, Nicholson AG, Noble PW, Selman M, Taniguchi H, Brun M, Le Maulf F, Girard M, Stowasser S, Schlenker-Herceg R, Disse B, Collard HR, INPULSIS Trial Investigators. Efficacy and safety of nintedanib in idiopathic pulmonary fibrosis. N Engl J Med. 2014;370:2071–82. https://doi.org/10.1056/NEJMoa1402584 .

Raghu G, Remy-Jardin M, Richeldi L, Thomson CC, Inoue Y, Johkoh T, Kreuter M, Lynch DA, Maher TM, Martinez FJ, Molina-Molina M, Myers JL, Nicholson AG, Ryerson CJ, Strek ME, Troy LK, Wijsenbeek M, Mammen MJ, Hossain T, Bissell BD, Herman DD, Hon SM, Kheir F, Khor YH, Macrea M, Antoniou KM, Bouros D, Buendia-Roldan I, Caro F, Crestani B, Ho L, Morisset J, Olson AL, Podolanczuk A, Poletti V, Selman M, Ewing T, Jones S, Knight SL, Ghazipura M, Wilson KC. Idiopathic pulmonary fibrosis (an update) and progressive pulmonary fibrosis in adults: an official ATS/ERS/JRS/ALAT clinical practice guideline. Am J Respir Crit Care Med. 2022;205:e18–47. https://doi.org/10.1164/rccm.202202-0399ST .

Article   PubMed   PubMed Central   Google Scholar  

King TE Jr, Bradford WZ, Castro-Bernardini S, Fagan EA, Glaspole I, Glassberg MK, Gorina E, Hopkins PM, Kardatzke D, Lancaster L, Lederer DJ, Nathan SD, Pereira CA, Sahn SA, Sussman R, Swigris JJ, Noble PW, ASCEND Study Group. A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N Engl J Med. 2014;370:2083–92. https://doi.org/10.1056/NEJMoa1402582 .

Sofia C, Comes A, Sgalla G, Richeldi L. Promising advances in treatments for the management of idiopathic pulmonary fibrosis. Expert Opin Pharmacother. 2024;25:717–25. https://doi.org/10.1080/14656566.2024.2354460 .

Bonella F, Spagnolo P, Ryerson C. Current and future treatment landscape for idiopathic pulmonary fibrosis. Drugs. 2023;83:1581–93. https://doi.org/10.1007/s40265-023-01950-0 .

Richeldi L, Azuma A, Cottin V, Hesslinger C, Stowasser S, Valenzuela C, Wijsenbeek MS, Zoz DF, Voss F, Maher TM, 1305-0013 Trial Investigators. Trial of a preferential phosphodiesterase 4B inhibitor for idiopathic pulmonary fibrosis. N Engl J Med. 2022;386:2178–87. https://doi.org/10.1056/NEJMoa2201737 .

Waxman A, Restrepo-Jaramillo R, Thenappan T, Engel P, Bajwa A, Ravichandran A, Feldman J, Hajari Case A, Argula RG, Tapson V, Smith P, Deng C, Shen E, Nathan SD. Long-term inhaled treprostinil for pulmonary hypertension due to interstitial lung disease: INCREASE open-label extension study. Eur Respir J. 2023;61:2202414. https://doi.org/10.1183/13993003.02414-2022 .

Article   PubMed   PubMed Central   CAS   Google Scholar  

Nathan SD, Behr J, Cottin V, Lancaster L, Smith P, Deng CQ, Pearce N, Bell H, Peterson L, Flaherty KR. Study design and rationale for the TETON phase 3, randomised, controlled clinical trials of inhaled treprostinil in the treatment of idiopathic pulmonary fibrosis. BMJ Open Respir Res. 2022;9: e001310. https://doi.org/10.1136/bmjresp-2022-001310 .

Toyama T, Nakamura H, Harano Y, Yamauchi N, Morita A, Kirishima T, Minami M, Itoh Y, Okanoue T. PPARalpha ligands activate antioxidant enzymes and suppress hepatic fibrosis in rats. Biochem Biophys Res Commun. 2004;324:697–704. https://doi.org/10.1016/j.bbrc.2004.09.110 .

Ogata T, Miyauchi T, Sakai S, Irukayama-Tomobe Y, Goto K, Yamaguchi I. Stimulation of peroxisome-proliferator-activated receptor alpha (PPAR alpha) attenuates cardiac fibrosis and endothelin-1 production in pressure-overloaded rat hearts. Clin Sci (Lond). 2002;103:S284–8. https://doi.org/10.1042/CS103S284S .

Article   Google Scholar  

Oruqaj G, Karnati S, Vijayan V, Kotarkonda LK, Boateng E, Zhang W, Ruppert C, Günther A, Shi W, Baumgart-Vogt E. Compromised peroxisomes in idiopathic pulmonary fibrosis, a vicious cycle inducing a higher fibrotic response via TGF-beta signaling. Proc Natl Acad Sci USA. 2015;112:2048–57. https://doi.org/10.1073/pnas.1415111112 .

Article   CAS   Google Scholar  

Oruqaj G, Karnati S, Kotarkonda LK, Boateng E, Bartkuhn M, Zhang W, Ruppert C, Günther A, Bartholin L, Shi W, Baumgart-Vogt E. Transforming growth factor-b1 regulates peroxisomal genes/proteins via Smad signaling in idiopathic pulmonary fibrosis fibroblasts and transgenic mouse models. Am J Pathol. 2023;193:259–74. https://doi.org/10.1016/j.ajpath.2022.11.006 .

Foreman JE, Sharma AK, Amin S, Gonzalez FJ, Peters JM. Ligand activation of peroxisome proliferator-activated receptor-beta/delta (PPARbeta/delta) inhibits cell growth in a mouse mammary gland cancer cell line. Cancer Lett. 2010;288:219–25. https://doi.org/10.1016/j.canlet.2009.07.006 .

Ham SA, Kim HJ, Kim HJ, Kang ES, Eun SY, Kim GH, Park MH, Woo IS, Kim HJ, Chang KC, Lee JH, Seo HG. PPARdelta promotes wound healing by up-regulating TGF-beta1-dependent or -independent expression of extracellular matrix proteins. J Cell Mol Med. 2010;14:1747–59. https://doi.org/10.1111/j.1582-4934.2009.00816.x .

Burgess HA, Daugherty LE, Thatcher TH, Lakatos HF, Ray DM, Redonnet M, Phipps RP, Sime PJ. PPARgamma agonists inhibit TGF-beta induced pulmonary myofibroblast differentiation and collagen production: implications for therapy of lung fibrosis. Am J Physiol Lung Cell Mol Physiol. 2005;288:L1146–53. https://doi.org/10.1152/ajplung.00383.2004 .

Milam JE, Keshamouni VG, Phan SH, Hu B, Gangireddy SR, Hogaboam CM, Standiford TJ, Thannickal VJ. Reddy RC PPAR-gamma agonists inhibit profibrotic phenotypes in human lung fibroblasts and bleomycin-induced pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol. 2008;294:L891–901. https://doi.org/10.1152/ajplung.00333.2007 .

Ferguson HE, Kulkarni A, Lehmann GM, Garcia-Bates TM, Thatcher TH, Huxlin KR, Phipps RP, Sime PJ. Electrophilic peroxisome proliferator-activated receptor-gamma ligands have potent antifibrotic effects in human lung fibroblasts. Am J Respir Cell Mol Biol. 2009;41:722–30. https://doi.org/10.1165/rcmb.2009-0006OC .

Lin Q, Fang LP, Zhou WW, Liu XM. Rosiglitazone inhibits migration, proliferation, and phenotypic differentiation in cultured human lung fibroblasts. Exp Lung Res. 2010;36:120–8. https://doi.org/10.3109/01902140903214659 .

El Agha E, Moiseenko A, Kheirollahi V, De Langhe S, Crnkovic S, Kwapiszewska G, Szibor M, Kosanovic D, Schwind F, Schermuly RT, Henneke I, MacKenzie B, Quantius J, Herold S, Ntokou A, Ahlbrecht K, Braun T, Morty RE, Günther A, Seeger W, Bellusci S. Two-way conversion between lipogenic and myogenic fibroblastic phenotypes marks the progression and resolution of lung fibrosis. Cell Stem Cell. 2017;20:261-73.e3. https://doi.org/10.1016/j.stem.2016.10.004 .

Speca S, Dubuquoy C, Rousseaux C, Chavatte P, Desreumaux P, Spagnolo P. GED-0507 attenuates lung fibrosis by counteracting myofibroblast transdifferentiation in vivo and in vitro. PLoS ONE. 2021;16: e0257281. https://doi.org/10.1371/journal.pone.0257281 .

Korbecki J, Bobiński R, Dutka M. Self-regulation of the inflammatory response by peroxisome proliferator-activated receptors. Inflamm Res. 2019;68:443–58. https://doi.org/10.1007/s00011-019-01231-1 .

Derrett-Smith E, Clark KEN, Shiwen X, Abraham DJ, Hoyles RK, Lacombe O, Broqua P, Junien JL, Konstantinova I, Ong VH, Denton CP. The pan-PPAR agonist lanifibranor reduces development of lung fibrosis and attenuates cardiorespiratory manifestations in a transgenic mouse model of systemic sclerosis. Arthritis Res Ther. 2021;23:234. https://doi.org/10.1186/s13075-021-02592-x .

Avouac J, Konstantinova I, Guignabert C, Pezet S, Sadoine J, Guilbert T, Cauvet A, Tu L, Luccarini JM, Junien JL, Broqua P, Allanore Y. Pan-PPAR agonist IVA337 is effective in experimental lung fibrosis and pulmonary hypertension. Ann Rheum Dis. 2017;76:1931–40. https://doi.org/10.1136/annrheumdis-2016-210821 .

Lakatos HF, Thatcher TH, Kottmann RM, Garcia TM, Phipps RP, Sime PJ. The role of PPARs in lung fibrosis. PPAR Res. 2007;2007:71323. https://doi.org/10.1155/2007/71323 .

Zeng Q, Zhou TT, Huang WJ, Huang XT, Huang L, Zhang XH, Sang XX, Luo YY, Tian YM, Wu B, Liu L, Luo ZQ, He B, Liu W, Tang SY. Asarinin attenuates bleomycin-induced pulmonary fibrosis by activating PPARgamma. Sci Rep. 2023;13:14706. https://doi.org/10.1038/s41598-023-41933-5 .

Craig VJ, Zhang L, Hagood JS, Owen CA. Matrix metalloproteinases as therapeutic targets for idiopathic pulmonary fibrosis. Am J Respir Cell Mol Biol. 2015;53:585–600. https://doi.org/10.1165/rcmb.2015-0020TR .

Odajima N, Betsuyaku T, Nagai K, Moriyama C, Wang DH, Takigawa T, Ogino K, Nishimura M. The role of catalase in pulmonary fibrosis. Respir Res. 2010;11:183. https://doi.org/10.1186/1465-9921-11-183 .

Smejkal GB, Kakumanu S. Enzymes and their turnover numbers. Expert Rev Proteomics. 2019;16:543–4. https://doi.org/10.1080/14789450.2019.1630275 .

Filatova A, Seidel S, Böğürcü N, Gräf S, Garvalov BK, Acker T. Acidosis acts through HSP90 in a PHD/VHL-independent manner to promote HIF function and stem cell maintenance in glioma. Cancer Res. 2016;76:5845–56. https://doi.org/10.1158/0008-5472.CAN-15-2630 .

Vijayan V, Srinu T, Karnati S, Garikapati V, Linke M, Kamalyan L, Mali SR, Sudan K, Kollas A, Schmid T, Schulz S, Spengler B, Weichhart T, Immenschuh S, Baumgart-Vogt E. A new immunomodulatory role for peroxisomes in macrophages activated by the TLR4 ligand lipopolysaccharide. J Immunol. 2017;198:2414–25. https://doi.org/10.4049/jimmunol.1601596 .

Matyash V, Liebisch G, Kurzchalia TV, Shevchenko A, Schwudke D. Lipid extraction by methyl-tert-butyl ether for high-throughput lipidomics. J Lipid Res. 2008;49:1137–46. https://doi.org/10.1194/jlr.D700041-JLR200 .

Garikapati V, Colasante C, Baumgart-Vogt E, Spengler B. Sequential lipidomic, metabolomic, and proteomic analyses of serum, liver, and heart tissue specimens from peroxisomal biogenesis factor 11a knockout mice. Anal Bioanal Chem. 2022;414:2235–50. https://doi.org/10.1007/s00216-021-03860-0 .

Khalil N, O’Connor RN, Unruh HW, Warren PW, Flanders KC, Kemp A, Bereznay OH, Greenberg AH. Increased production and immunohistochemical localization of transforming growth factor-beta in idiopathic pulmonary fibrosis. Am J Respir Cell Mol Biol. 1991;5:155–62. https://doi.org/10.1165/ajrcmb/5.2.155 .

Bergeron A, Soler P, Kambouchner M, Loiseau P, Milleron B, Valeyre D, Hance AJ, Tazi A. Cytokine profiles in idiopathic pulmonary fibrosis suggest an important role for TGF-beta and IL-10. Eur Respir J. 2003;22:69–76. https://doi.org/10.1183/09031936.03.00014703 .

Qian G, Fan W, Ahlemeyer B, Karnati S, Baumgart-Vogt E. Peroxisomes in different skeletal cell types during intramembranous and endochondral ossification and their regulation during osteoblast differentiation by distinct peroxisome proliferator-activated receptors. PLoS ONE. 2015;10: e0143439. https://doi.org/10.1371/journal.pone.0143439 .

Moore BB, Lawson WE, Tim Oury TD, Sisson TH, Raghavendran K, Hogaboam CM. Animal models of fibrotic lung disease. Am J Respir Cell Mol Biol. 2013;49:167–79. https://doi.org/10.1165/rcmb.2013-0094TR .

Sime PJ, Xing Z, Graham FL, Csaky KG, Gauldie J. Adenovector-mediated gene transfer of active transforming growth factor-beta1 induces prolonged severe fibrosis in rat lung. J Clin Invest. 1997;100:768–76. https://doi.org/10.1172/JCI119590 .

Fukuda Y, Basset F, Ferrans VJ, Yamanaka N. Significance of early intra-alveolar fibrotic lesions and integrin expression in lung biopsy specimens from patients with idiopathic pulmonary fibrosis. Hum Pathol. 1995;26:53–61. https://doi.org/10.1016/0046-8177(95)90114-0 .

Cavazza A, Rossi G, Carbonelli C, Spaggiari L, Paci M, Roggeri A. The role of histology in idiopathic pulmonary fibrosis: an update. Respir Med. 2010;104:S11-22. https://doi.org/10.1016/j.rmed.2010.03.013 .

Article   PubMed   Google Scholar  

Jones MG, et al. Three-dimensional characterization of fibroblast foci in idiopathic pulmonary fibrosis. JCI Insight. 2016;1: e86375. https://doi.org/10.1172/jci.insight.86375 .

Kelly M, Kolb M, Bonniaud P, Gauldie J. Re-evaluation of fibrogenic cytokines in lung fibrosis. Curr Pharm Des. 2003;9:39–49. https://doi.org/10.2174/1381612033392341 .

Cerny L, Torday JS, Rehan VK. Prevention and treatment of bronchopulmonary dysplasia: contemporary status and future outlook. Lung. 2008;186:75–89. https://doi.org/10.1513/pats.201203-023AW .

Fernandez IE, Eickelberg O. The impact of TGF-beta on lung fibrosis: from targeting to biomarkers. Proc Am Thorac Soc. 2012;9:111–6. https://doi.org/10.1513/pats.201203-023AW .

Nakos G, Adams A, Andriopoulos N. Antibodies to collagen in patients with idiopathic pulmonary fibrosis. Chest. 1993;103:1051–8. https://doi.org/10.1378/chest.103.4.1051 .

Ramos C, Montaño M, García-Alvarez J, Ruiz V, Uhal BD, Selman M, Pardo A. Fibroblasts from idiopathic pulmonary fibrosis and normal lungs differ in growth rate, apoptosis, and tissue inhibitor of metalloproteinases expression. Am J Respir Cell Mol Biol. 2001;24:591–8. https://doi.org/10.1165/ajrcmb.24.5.4333 .

Su Y, Gu H, Weng D, Zhou Y, Li Q, Zhang F, Zhang Y, Shen L, Hu Y, Li H. Association of serum levels of laminin, type IV collagen, procollagen III N-terminal peptide, and hyaluronic acid with the progression of interstitial lung disease. Medicine (Baltimore). 2017;96: e6617. https://doi.org/10.1097/MD.0000000000006617 .

Sun KH, Chang Y, Reed NI, Sheppard D. alpha-Smooth muscle actin is an inconsistent marker of fibroblasts responsible for force-dependent TGFbeta activation or collagen production across multiple models of organ fibrosis. Am J Physiol Lung Cell Mol Physiol. 2016;310:L824–36. https://doi.org/10.1152/ajplung.00350.2015 .

Liu X, Dai K, Zhang X, Huang G, Lynn H, Rabata A, Liang J, Noble PW, Jiang D. Multiple fibroblast subtypes contribute to matrix deposition in pulmonary fibrosis. Am J Respir Cell Mol Biol. 2023;69:45–56. https://doi.org/10.1165/rcmb.2022-0292OC .

Zhang K, Rekhter MD, Gordon D, Phan SH. Myofibroblasts and their role in lung collagen gene expression during pulmonary fibrosis. A combined immunohistochemical and in situ hybridization study. Am J Pathol. 1994;145:114–25.

PubMed   PubMed Central   CAS   Google Scholar  

Zhang B, Berger J, Zhou G, Elbrecht A, Biswas S, White-Carrington S, Szalkowski D, Moller DE. Insulin- and mitogen-activated protein kinase-mediated phosphorylation and activation of peroxisome proliferator-activated receptor gamma. J Biol Chem. 1996;271:31771–4. https://doi.org/10.1074/jbc.271.50.31771 .

Gandhi S, Tonelli R, Murray M, Samarelli AV, Spagnolo P. Environmental causes of idiopathic pulmonary fibrosis. Int J Mol Sci. 2023;24:16481. https://doi.org/10.3390/ijms242216481 .

Wallace AM, Sandford AJ, English JC, Burkett KM, Li H, Finley RJ, Müller NL, Coxson HO, Paré PD, Abboud RT. Matrix metalloproteinase expression by human alveolar macrophages in relation to emphysema. COPD. 2008;5:13–23. https://doi.org/10.1080/15412550701817789 .

Garcia-de-Alba C, Buendia-Roldán I, Salgado A, Becerril C, Ramírez R, González Y, Checa M, Navarro C, Ruiz V, Pardo A, Selman M. Expression of matrix metalloproteases by fibrocytes: possible role in migration and homing. Am J Respir Crit Care Med. 2010;182:1144–52. https://doi.org/10.1164/rccm.201001-0028OC .

Ham SA, Yoo T, Hwang JS, Kang ES, Lee WJ, Paek KS, Park C, Kim JH, Do JT, Lim DS, Seo HG. Ligand-activated PPARδ modulates the migration and invasion of melanoma cells by regulating Snail expression. Am J Cancer Res. 2014;4:674–82.

PubMed   PubMed Central   Google Scholar  

Tallima H, El Ridi R. Arachidonic acid: physiological roles and potential health benefits—a review. J Adv Res. 2018;11:33–41. https://doi.org/10.1016/j.jare.2017.11.004 .

Meital LT, Windsor MT, Perissiou M, Schulze K, Magee R, Kuballa A, Golledge J, Bailey TG, Askew CD, Russell FD. Omega-3 fatty acids decrease oxidative stress and inflammation in macrophages from patients with small abdominal aortic aneurysm. Sci Rep. 2019;9:12978. https://doi.org/10.1038/s41598-019-49362-z .

Kliewer SA, Forman BM, Blumberg B, Ong ES, Borgmeyer U, Mangelsdorf DJ, Umesono K, Evans RM. Fatty acids and eicosanoids regulate gene expression through direct interactions with peroxisome proliferator-activated receptors alpha and gamma. Proc Natl Acad Sci USA. 1997;94:4318–23. https://doi.org/10.1073/pnas.94.9.4318 .

Colasante C, Chen J, Ahlemeyer B, Baumgart-Vogt E. Peroxisomes in cardiomyocytes and the peroxisome/peroxisome proliferator-activated receptor-loop. Thromb Haemost. 2015;113:452–63. https://doi.org/10.1160/TH14-06-0497 .

Hashimoto F, Hayashi H. Significance of catalase in peroxisomal fatty acyl-CoA beta-oxidation: NADH oxidation by acetoacetyl-CoA and H 2 O 2 . J Biochem. 1990;108:426–31. https://doi.org/10.1093/oxfordjournals.jbchem.a123217 .

Fransen M, Lismont C. Peroxisomal hydrogen peroxide signalling: a new chapter in intracellular communication research. Curr Opin Chem Biol. 2024;78:102426. https://doi.org/10.1016/j.cbpa.2024.102426 .

Girnun GD, Domann FE, Moore SA, Robbins ME. Identification of a functional peroxisome proliferator-activated receptor response element in the rat catalase promoter. Mol Endocrinol. 2002;16:2793–801. https://doi.org/10.1210/me.2002-0020 .

Okuno Y, Matsuda M, Miyata Y, Fukuhara A, Komuro R, Shimabukuro M, Shimomura I. Human catalase gene is regulated by peroxisome proliferator activated receptor-gamma through a response element distinct from that of mouse. Endocr J. 2010;57:303–9. https://doi.org/10.1507/endocrj.k09e-113 .

Hur J, Kang ES, Hwang JS, Lee WJ, Won JP, Lee HG, Kim E, Seo HG. Peroxisome proliferator-activated receptor-delta-mediated upregulation of catalase helps to reduce ultraviolet B-induced cellular injury in dermal fibroblasts. J Dermatol Sci. 2021;103:167–75. https://doi.org/10.1016/j.jdermsci.2021.08.003 .

Hwang JS, Kim E, Lee HG, Lee WJ, Won JP, Hur J, Fujii J, Seo HG. Peroxisome proliferator-activated receptor δ rescues xCT-deficient cells from ferroptosis by targeting peroxisomes. Biomed Pharmacother. 2021;143: 112223. https://doi.org/10.1016/j.biopha.2021.112223 .

Maltarollo VG, Togashi M, Nascimento AS, Honorio KM. Structure-based virtual screening and discovery of new PPARd/g dual agonist and PPARd and g agonists. PLoS ONE. 2015;10: e0118790. https://doi.org/10.1371/journal.pone.0118790 .

Download references

Acknowledgements

Many thanks to Petra Hahn-Kohlberger, Bianca Pfeiffer, Andrea Textor and Susanne Pfreimer for their outstanding technical assistance. Sincere gratitude to Dr. Eunsum Jung (Biospectrum Life Science Institute) and Prof. Marc Fransen (Université catholique de Louvain, Belgium) for providing the COL1A2 luciferase and catalase overexpression plasmids, respectively.

Open Access funding enabled and organized by Projekt DEAL. This work was supported by funding from the German Academic Exchange Service (Government of Ghana DAAD, grant number 50015294) to EB, collaborative grant of the German Academic Exchange Service (DAAD) for granting the Graduate School Scholarship Programme "Lipids in Nutrition and Metabolism" (DAAD-GSSP-2015) to EBV, VG is a doctoral scholarship holder within this programme (Personal reference no 91566181), and performance-related resource allocation-funds of the Medical Faculty of the Justus Liebig University Giessen, so-called “ L eistungs o rientierte M ittel” (LOM) to EBV.

Author information

Srikanth Karnati

Present address: Institute for Anatomy and Cell Biology, Julius Maximilians University, 97070, Würzburg, Germany

Eistine Boateng

Present address: Department of Medical Education, College of Medicine and Life Sciences, University of Toledo, Toledo, OH, 43614, USA

Vannuruswamy Garikapati

Present address: Max Planck Institute of Molecular Cell Biology and Genetics, 01307, Dresden, Germany

Gani Oruqaj

Present address: Department of Internal Medicine II, Member of the German Center for Lung Research (DZL), Universities of Giessen and Marburg Lung Center (UGMLC), Justus Liebig University, 35392, Giessen, Germany

Natalia El-Merhie

Present address: Institute for Lung Health (ILH), Member of the German Center for Lung Research (DZL), Universities of Giessen and Marburg Lung Center (UGMLC), Justus Liebig University, 35392, Giessen, Germany

Srikanth Karnati and Eveline Baumgart-Vogt share senior authorship.

Authors and Affiliations

Institute for Anatomy and Cell Biology, Division of Medical Cell Biology, Justus Liebig University, Aulweg 123, 35392, Giessen, Germany

Eistine Boateng, Rocio Bonilla-Martinez, Barbara Ahlemeyer, Vannuruswamy Garikapati, Mohammad Rashedul Alam, Gani Oruqaj, Natalia El-Merhie, Srikanth Karnati & Eveline Baumgart-Vogt

Department of Internal Medicine VIII, Eberhard Karls University, 72076, Tübingen, Germany

Omelyan Trompak

Excellence Cluster Cardio-Pulmonary System, German Center for Lung Research (DZL), Universities of Giessen and Marburg Lung Center, 35392, Giessen, Germany

Michael Seimetz, Clemens Ruppert & Andreas Günther

UGMLC Giessen Biobank, Universities of Giessen and Marburg Lung Center, 35392, Giessen, Germany

Clemens Ruppert

Center for Interstitial and Rare Lung Diseases, Department of Internal Medicine, German Center for Lung Research, Universities of Giessen and Marburg Lung Center, 35392, Giessen, Germany

Andreas Günther

Institute of Inorganic and Analytical Chemistry, Justus Liebig University, 35392, Giessen, Germany

Vannuruswamy Garikapati & Bernhard Spengler

You can also search for this author in PubMed   Google Scholar

Contributions

E.B.-V., S.K. and E.B. conceived and designed the research studies; E.B., R.B, B.A., O.T., V.G. and M.R.A. conducted the experiments; E.B.-V., B.A., B.S., S.K., R.B., V.G., N.E-M. and E.B. acquired and analyzed the data analyzed; E.B.-V., S.K., M.S., B.S., C.R., G.O., and A.G. provided reagents and materials; and E.B.-V., B.A., R.B. and E.B. wrote the manuscript. All authors reviewed the manuscript.

Corresponding author

Correspondence to Eveline Baumgart-Vogt .

Ethics declarations

Ethics approval and consent to participate.

Protocol, data collection and analysis of the human material (isolated lung fibroblasts, lung tissues) were approved by the Local Ethics Committee of the Justus Liebig University Giessen (Az58/15 and Az111/08, JLU).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Eistine Boateng, Vannuruswamy Garikapati, Gani Oruqaj, Natalia El-Merhie, Srikanth Karnati: All experimental work has been done at Institute for Anatomy and Cell Biology, Division of Medical Cell Biology, Justus Liebig University, Aulweg 123, 35392, Giessen, Germany. Vannuruswamy Garikapati: The experimental work has been done in a cooperative project at two places at the JLU Giessen (Institute for Anatomy and Cell Biology, Division of Medical Cell Biology, Justus Liebig University and Institute of Inorganic and Analytical Chemistry, Justus Liebig University).

Supplementary Information

Detailed description of the methods, 8 additional files., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Boateng, E., Bonilla-Martinez, R., Ahlemeyer, B. et al. It takes two peroxisome proliferator-activated receptors (PPAR-β/δ and PPAR-γ) to tango idiopathic pulmonary fibrosis. Respir Res 25 , 345 (2024). https://doi.org/10.1186/s12931-024-02935-7

Download citation

Received : 03 May 2024

Accepted : 01 August 2024

Published : 23 September 2024

DOI : https://doi.org/10.1186/s12931-024-02935-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Human lung fibroblasts
• Matrix metalloproteinases

Respiratory Research

ISSN: 1465-993X

• General enquiries: [email protected]

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

The PMC website is updating on October 15, 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

Effects of Protein Supplementation on Performance and Recovery in Resistance and Endurance Training

Harry p. cintineo.

1 Center for Health and Human Performance, Rutgers University, New Brunswick, NJ, United States

Michelle A. Arent

Jose antonio.

2 Department of Health and Human Performance, Nova Southeastern University, Davie, FL, United States

Shawn M. Arent

3 Department of Kinesiology and Health, Rutgers University, New Brunswick, NJ, United States

There is robust evidence which shows that consuming protein pre- and/or post-workout induces a significant rise in muscle protein synthesis. It should be noted, however, that total daily caloric and protein intake over the long term play the most crucial dietary roles in facilitating adaptations to exercise. However, once these factors are accounted for, it appears that peri-exercise protein intake, particularly in the post-training period, plays a potentially useful role in terms of optimizing physical performance and positively influencing the subsequent recovery processes for both resistance training and endurance exercise. Factors that affect the utility of pre- or post-workout feeding include but are not necessarily limited to: training status (e.g., novice vs. advanced, or recreational vs. competitive athlete), duration of exercise, the number of training sessions per day, the number of competitive events per day, etc. From a purely pragmatic standpoint, consuming protein post-workout represents an opportunity to feed; this in turn contributes to one's total daily energy and protein intake. Furthermore, despite recent suggestions that one does not “need” to consume protein during the immediate (1 h or less) post-training time frame, it should be emphasized that consuming nothing offers no advantage and perhaps even a disadvantage. Thus, based on performance and recovery effects, it appears that the prudent approach would be to have athletes consume protein post-training and post-competition.

Introduction

Dietary protein plays a critical role in countless physiological processes in the body. The current Recommended Dietary Allowance (RDA) for healthy individuals is 0.8 g/kg/day ( 1 ). It is increasingly evident, however, that protein intake of at least 1.4–1.6 g/kg/day ( 2 ) would be more appropriate for active individuals attempting to optimize training adaptations. In an effort to meet this threshold, protein supplements are often consumed. In 2015, protein powder sales were valued at 4.7 billion U.S. dollars and were second only to sport drinks in the sports nutrition market ( 3 ). The popularity of protein supplements is likely influenced by the claims of increased muscle mass, increased fat loss, improved performance, and improved markers of recovery.

To date several meta-analyses, reviews, and systematic reviews have attempted to quantify and clarify these claims, but with mixed results ( 2 , 4 – 7 ). However, these efforts are complicated by the fact that the populations studied included trained and untrained, healthy normal weight, overweight or obese individuals, as well as injured, movement impaired, and those with metabolic or other diseases states. Additionally, the emphasis of recent reviews has been largely on impacts on muscle protein synthesis (MPS), hypertrophy, and body composition, with most of the outcomes pertaining solely to resistance training ( 2 , 4 – 7 ). Performance and recovery effects have been given secondary consideration at best, and these are areas that would be of particular interest to most athletes or athletic individuals.

Furthermore, performance and recovery outcomes, as well as physiological adaptations, are unique to the modality of training primarily employed. Anaerobic training refers to short bouts of high intensity movements which are often interspersed with longer recovery periods between efforts, with two of the most popular applications being resistance training or interval training ( 8 ). On the other hand, aerobic or endurance training refers to exercise bouts that primarily rely on oxidative phosphorylation and can last from minutes to hours ( 9 ). This latter type of training has received almost no consideration in recent protein reviews. Whether engaging in resistance or endurance training, protein supplementation may have the potential to enhance or complement exercise-induced physiological responses. The purpose of this review is to examine these potential performance and recovery applications of protein supplementation for both resistance and endurance training, with emphasis placed on studies utilizing various “peri-exercise” supplementation protocols within ~60 min pre- or post-training in healthy, exercising individuals.

Protein supplementation and resistance training

A recent comprehensive review by Jager et al. ( 2 ) identified a number of key issues related to protein intake in healthy, exercising individuals. Of particular note, the importance of protein intake during and around a training session for recovery and performance appears to be dependent on total daily protein intake, as well as presence or absence of an energy deficit. While findings do support the effect of post-exercise protein intake on increases in fat free mass (FFM), individuals consuming adequate daily calories and a minimum daily protein intake of 1.6 g/kg may not see any added benefit of immediate post-training protein consumption on muscular strength ( 2 ). However, Morton et al. ( 7 ) suggested that the strength (and hypertrophy) effects of additional post-resistance training protein supplementation may be greater in those with previous resistance training experience and that the magnitude of this effect is somewhat mitigated with aging. Furthermore, it is important to note that resistance-trained individuals in a caloric deficit require significantly more protein to offset any potential loss of lean body mass, with optimal daily protein intake for these individuals potentially being in the range of 2.3–3.1 g/kg FFM ( 10 ). While this recommendation increases total caloric intake from protein, resulting in the necessity to decrease energy intake from fat and carbohydrate, protein appears to have unique characteristics, and overfeeding with protein has been shown to have no negative effects on body composition in trained individuals ( 11 ). Similarly, healthy, older adults also require a greater quantity of total daily protein (0.61 g/kg FFM) compared to their younger counterparts (0.25 g/kg FFM) ( 12 ). Additionally, as a percentage of total daily energy intake, older adults must increase the contribution from protein due to decreases in energy intake, as well as protein's ability to attenuate sarcopenia by increasing muscle hypertrophy, subsequently maintaining or increasing muscular strength and power ( 13 ).

It has previously been demonstrated that ingestion of milk-based protein following a damaging eccentric resistance protocol helps to attenuate the expected decrements in strength and repeated sprint ability from 24 to 72 h following the bout ( 14 – 16 ). Recently, a group of researchers found that whey protein can facilitate muscle recovery following an intense isotonic exercise bout as well and that it is more than just an issue of caloric replacement ( 17 ). They compared the effects of a whey protein supplement (25 g protein, 2.5 g fat, and 3 g CHO) to a calorie-equated carbohydrate drink (32.5 g CHO) in resistance-trained young men performing an acute, total body resistance training protocol, and assessed performance variables at 10- and 24-h post-exercise. A moderate beneficial effect on acute anaerobic power and strength was found in the group that consumed the protein supplement, suggesting that there may have been improvements in rate of recovery over those who consumed the carbohydrate drink ( 17 ). This is particularly notable given that the subjects were already habitually consuming 1.9 g/kg/d of protein and may hold particular relevance for athletes engaging in high-intensity, explosive sports.

It has been suggested that protein quality may have an effect on both acute and chronic adaptations to exercise ( 2 , 18 , 19 ). Protein quality is a measure of a given protein source's ability to provide adequate quantities of the essential amino acids required for protein synthesis ( 20 ). Additionally, leucine, a branched-chain amino acid (BCAA), has been shown to be a prerequisite stimulator of skeletal MPS, which is critical for both the recovery and adaptive processes following a training bout ( 21 ). Given some of the favorable outcomes seen with ingestion of certain complete proteins, particularly milk-based and, more specifically, whey proteins ( 2 ), questions have been raised about the possible application of other protein sources that may be lower in leucine content. Two recent investigations have studied the effects of the quality of a post-exercise protein source on performance and recovery ( 22 , 23 ). Each of these studies took a unique approach to determining the differences in physiological changes following exercise and protein supplementation. Fabre et al. ( 22 ) compared the effects of 20 g of whey protein, 10 g of whey protein plus 10 g of casein protein, and 4 g of whey protein plus 16 g of casein protein consumed post-exercise in 31 recreationally resistance-trained males. Following 9 weeks of resistance training 4 days per week, no differences in changes in body composition, muscular strength, or muscular endurance were found, suggesting all three protein supplements were equally effective. When comparing 16 g of beef protein, 18 g of whey protein, and a calorie-equated carbohydrate drink consumed post-resistance training 3 days per week for 8 weeks in 42 recreationally resistance-trained males, no differences in changes in body composition, muscle thickness, or performance variables were found ( 23 ). One limitation of each of these studies is that they failed to control for total daily energy and macronutrient intake; therefore, subjects may have already been consuming adequate total daily calories and protein so the additional protein, regardless of its source, failed to result in any additional improvements in performance or body composition.

While most protein supplement resistance training studies have used a “post-exercise” administration protocol, it is possible that timing effects extend to the entire peri-workout period. Schoenfeld et al. ( 24 ) examined the effects of consuming 25 g of hydrolyzed whey protein immediately prior to a resistance training session with a 3-h fast post-exercise vs. consuming the same quantity and source of protein immediately following the same training session after having fasted for 3 h in 21 resistance-trained males. All subjects were consuming a 500-kcal surplus and 1.8 g/kg of protein daily. No differences in changes in body composition or one-rep max back squat or bench press were found between the groups after the 8-week intervention. Along with the findings from other studies ( 22 , 23 , 25 ), these data support the idea that protein intake post-workout may not be critical as long as protein is consumed prior to training or total daily protein intake is adequate. However, this does not preclude the possibility that pre- and post-exercise supplementation would be even more beneficial depending on dose.

To interpret the disparate effects of protein supplementation on resistance training performance, a few issues should be taken into account. The training stimulus must be adequate to result in strength improvement, regardless of protein timing, total protein intake, or nutritional status. Protein supplementation by individuals participating in ineffective resistance training programs will be less impactful. The beginning training status of individuals also appears to play a significant a role in any potential benefit seen as a result of protein consumption on strength, hypertrophy, and body composition ( 7 ). While the main focus of this paper is the healthy, trained individual, it is worth noting that protein supplementation for novice individuals may not confer any additional benefit above and beyond that of the training intervention ( 5 ). However, as training status increases, so does the potential effect of protein supplementation for improving performance and recovery.

Alternatively, Reidy and Rasmussen ( 6 ) have proposed the existence of a “protein paradox” wherein well-trained individuals may require less dietary protein due to the increased efficiency of protein turnover in this population. However, it should be noted that this is speculative and has not been fully substantiated by the available research, particularly for performance-related outcomes. Even taking this into account, one factor that appears to be just as important as total daily protein intake in well-trained individuals is the utilization of a specific protein dosing strategy based on body weight or FFM. Additionally, Thomson et al. ( 12 ) showed that healthy, older adults may also benefit from a higher protein intake in addition to a protein dosing strategy to adequately stimulate MPS. Thus, the appropriate timing or pacing of protein intake throughout the day may optimize results from resistance training ( 26 ). While recent critical or meta-analytic reviews have argued that protein timing is inconsequential after accounting for total protein intake ( 6 , 27 ), there are two factors that must be taken into account when considering these conclusions. First, very few “timing” studies have actually been conducted. In most cases, the studies were not designed to compare time of administration, but rather type or quantity of nutrient (or placebo) ingested post-exercise. Second, only a few of the included studies used trained subjects. Most employed novice exercisers. One of the studies that has found a benefit of protein timing ( 28 ) was conducted in experienced resistance-trained males. Again, this may lend credence to the notion that training status matters when considering protein supplementation strategies. Additionally, it should be noted that strength improvements not reaching statistical significance may prove to be significant in areas of individual competition or performance. Very few studies have actually utilized highly trained individuals or athletes, so translating the current findings to this population should be done with caution. Finally, it is worth noting that several studies have shown the addition of carbohydrate and creatine monohydrate to a protein supplement, typically whey protein, results in greater strength and hypertrophy improvements from resistance training programs ( 26 ). Though a detailed discussion of these other macronutrients is beyond the scope of this review, these results do point to an overall “nutrient” impact as well as possible synergistic effects.

Perhaps a driving factor in performance (i.e., strength or power) improvements with peri-workout protein supplementation could be enhanced recovery, which would potentially translate to enhanced capacity for an increased training load stimulus. Recovery from exercise has been measured through many different methods in previous research. Delayed onset muscle soreness (DOMS), which is defined as an aching pain in a given muscle following a novel exercise bout, has been measured subjectively ( 29 ). Though the cause of DOMS is multifaceted and tied to a cascade of events linked to muscle damage, it is not necessarily an indicator of the magnitude of muscle damage and, therefore, cannot be used by itself to determine muscular recovery and adaptations from exercise ( 29 ). Specific biomarkers and MPS rates appear to be the most efficient and widely used methods of objectively determining muscle breakdown, recovery, and adaptation from exercise. Acute elevations of cortisol and creatine kinase (CK) are two biological indicators of muscle damage and the subsequent recovery processes that can be measured through blood sample analysis ( 30 , 31 ). Post-exercise muscle biopsies can be used to determine rates of MPS, which directly measure the magnitude of the recovery process immediately following exercise ( 32 ).

West et al. ( 17 ) measured recovery variables following a total-body resistance training session and found that those subjects who consumed a whey protein supplement (25 g protein, 2.5 g fat, and 3 g CHO) had lower rates of whole body protein breakdown, while those who consumed a carbohydrate supplement (32.5 g CHO) actually had higher rates of whole body protein synthesis. The protein group, however, appeared to improve whole body net protein balance over 24 h post-exercise. As noted previously, the subjects were already consuming 1.9 g/kg/d protein, so additional protein through supplementation may have been less impactful. Interestingly, there was no difference between total body net protein balance between the groups. It should be noted that whole body protein synthesis is not necessarily a reflection of skeletal muscle protein synthesis ( 33 ). Kim et al. ( 33 ) discovered that net protein (whole body) balance was superior with a 70 vs. 40 g dose consumed prior to a resistance-training protocol. However, no differences were found in muscle protein synthesis between the 40 and 70 g dose. Thus, one must not conflate measures of whole body protein metabolism with those of skeletal muscle.

Nevertheless, the recovery of muscle function has been demonstrated in other studies ( 15 , 16 ) of milk protein supplementation after eccentric exercise, perhaps due to myofibrillar protein remodeling. The results of these studies further support the idea that protein consumed post-exercise is crucial for maximizing rates of protein synthesis in skeletal muscle. The effect on total body protein balance, however, is still a bit unclear. Carbohydrates have been shown to have a protein sparing effect, therefore the combination of protein and carbohydrate to decrease rates of muscle protein breakdown (MPB) and increase rates of MPS may be the best strategy for shifting total body protein balance to the net anabolic side ( 34 ), even if carbohydrate itself does not necessarily enhance MPS ( 35 , 36 ). This may partially explain the benefits of the milk supplement used by Cockburn et al. ( 15 ) and Cockburn and Stevenson ( 16 ) as it also contained carbohydrate. Perhaps there is a synergistic effect.

In addition to the investigations discussed earlier regarding post-exercise protein quality and training adaptations, Burd et al. ( 25 ) also measured markers of recovery through protein synthesis. The researchers collected muscle biopsies and measured rates of MPS following resistance training. In the 0–2 h post-exercise window, the group that consumed 30 g of protein in the form of skim milk expressed higher rates of MPS than the group that consumed 30 g of protein from beef ( 25 ). However, rates of MPS in the 2–5 h post-exercise window did not differ. This may be explained by the rate of digestion and absorption of these protein sources. Protein from dairy, specifically the whey portion, appears to be absorbed faster, and elicit a faster MPS response than beef.

The difference between whole egg and protein-equated egg white consumption post-exercise was also studied recently ( 37 ). The researchers measured rates of MPS through muscle biopsies and found that the group that consumed the whole egg exhibited higher rates of MPS. One limitation to this study was the lack of control for total calories and macronutrients. The whole egg treatment consisted of 18 g of protein, 17 g of fat, and 223 kcal, while the egg white treatment consistent of 18 g of protein, 0 g of fat, and only 73 kcal ( 37 ). While the discrepancy in calories between treatment groups may have impacted total daily calories, thus impacting MPS, one cannot ignore the possibility of the role that differences in macronutrients may play.

Lastly, a 2017 investigation looked at the differences between protein-equated native whey protein, whey protein concentrate, and milk ( 38 ). Native whey protein is produced through the filtration of raw milk, while whey protein concentrate is a byproduct of cheese production. Native whey protein consists of undenatured proteins and has a higher leucine content ( 38 ). Each treatment consisted of ~20 g of protein, ~6 g of fat, and ~40 g of carbohydrates but contained 2.7, 2.2, and 2.0 g of leucine, respectively. The supplements were ingested immediately after and again 2 h post-exercise following a moderate intensity lower body resistance training session. Results showed higher blood amino acids concentrations in native whey and whey protein concentrate than in milk. MPS was elevated in the whey protein condition from 1 to 3 h post, while it was elevated 1–5 h post in the native whey condition. There was no difference in MPS 1–5-h post-workout between native whey and whey protein concentrate, though MPS was higher from 1 to 5 h post-workout in the native whey condition compared to milk ( 38 ). Collectively, these data support that whey protein, regardless of its levels of processing (i.e., native whey protein vs. whey protein concentrate), increase MPS by similar magnitudes that are greater than those of milk alone. How this translates to long-term differences remains to be determined.

Protein supplementation and endurance training

While the majority of the literature regarding the effects of protein intake on performance has focused on anaerobic activities, more recent work has examined its role on endurance activities, but this has mostly been absent from the most recent reviews. Similar to resistance training, the impact appears to be at least somewhat dependent on the presence or absence of other nutrients, particularly carbohydrate. A 2010 systematic review and meta-analysis compared 11 studies investigating the effects of consumption of protein and carbohydrate vs. consumption of carbohydrate alone during a bout of cycling on performance during a subsequent bout of cycling ( 39 ). Across the 11 studies, consumption of protein and carbohydrate resulted in an average improvement of 9% in performance (defined as time to exhaustion and time trial performance) compared to consumption of carbohydrate alone ( 39 ). To investigate if the increased caloric intake due to inclusion of protein was responsible for this improved performance, a further analysis of isocarbohydrate and isocaloric conditions was performed. Examination of isocarbohydrate conditions yielded a 10.5% improvement in overall performance, while isocaloric conditions resulted in 3.4% improvement ( 39 ), suggesting that the improvements due to protein inclusion were not simply due to increased calories. When considering only those studies measuring performance by time trial (3), improvements were not statistically significant. However, studies utilizing time to exhaustion protocols (8) did result in statistically significant improvements. It is worth noting that in all studies showing statistically significant improvement, whey protein was the source of protein utilized, though differences between concentrate vs. isolate were not quantified. Again, it is prudent to highlight that performance improvements not reaching statistical significance may have clinical or practical relevance, specifically for athletes. For example, a 1% improvement in performance would have been the difference in winning the Gold Medal instead of the Silver Medal in the men's marathon in the 2016 Olympic Games in Rio. Therefore, even seemingly “trivial” differences do indeed have a significant effect on performance and outcomes at the elite level.

When discussing the impact of protein on performance, it is imperative to include the impact that protein may have on glycogen replenishment and subsequent exercise performance. Standard discussions of glycogen replenishment focus solely on carbohydrate consumption. Recommendations for adequate post-exercise carbohydrate consumption are to consume 0.6–1.0 g/kg carbohydrate within 30 min of cessation of exercise and again every 2 h for the next 4–6 h ( 40 , 41 ). Carbohydrate consumption of 1.2 g/kg every 30 min over 3.5 h also resulted in maximal glycogen replenishment ( 41 , 42 ). In cases of suboptimal post-exercise carbohydrate consumption, the addition of protein can improve glycogen replenishment and decrease symptoms of muscle damage ( 43 ). Practical applications of standard post-exercise carbohydrate consumption recommendations may be limited in real world situations. Moreover, athletes training multiple times daily may have fewer opportunities to consume recovery meals or have an elevated need for “rapid” recovery, including rehydration, to facilitate the subsequent training session. Beyond just glycogen replenishment aspects, it has also been shown that the presence of protein in rehydration beverages can enhance intestinal fluid uptake, aiding in rehydration ( 44 ) and that BCAA consumption during endurance exercise may improve time trial performance and peak power output while improving markers of immune health ( 45 ) and attenuate serotonin levels, subsequently resulting in a delay of central fatigue ( 46 ).

A systematic review by Pasiakos et al. ( 5 ) investigated the relationship between protein, muscle function, and recovery. The authors included studies that measured markers of muscle damage followed by a test of physical performance or muscle function. Populations of the review included healthy individuals with daily dietary protein intake at or above the current RDA of 0.8 g/kg per day. While some of the endurance exercise studies included showed decreases in markers of muscle damage, such as CK, or decreases in muscle soreness in groups consuming protein after initial exercise bout ( 47 – 49 ), many did not ( 50 – 52 ). This may have resulted from the inclusion of studies utilizing both trained and untrained subjects, as well as individuals consuming suboptimal daily protein intakes. Despite the reduced plasma CK levels and muscle soreness, consumption of protein did not result in improvements in subsequent performance measures when repeat performance was tested < 24 h following the initial bout. This evidence suggests that plasma CK levels, perceived level of muscle soreness, and muscle function may only be modestly related or perhaps utilizing a single method of measure paints an inadequate picture of recovery due to individual variability ( 5 ). Without additional studies to clarify these relationships, developing guidelines based on these markers as representing recovery may be ill-advised. Individuals must be cautious when attempting to measure recovery from exercise based on these metrics alone. For example, a recent study of 20 high-level soccer players tested the effects of a milk protein concentrate supplement (80% casein and 20% whey) compared to an isocaloric carbohydrate-containing placebo on high intensity running performance, knee extensor and flexor strength, and antioxidative capacity over the course of a 1-week in-season microcycle consisting of two games separated by 2 days ( 53 ). On game days (days 1 and 4), the supplement was consumed immediately post-, 3 h post-, and 6 h post-match in three different doses of 25, 30, and 25 g, respectively, resulting in a total of 80 g. On training days (days 2, 3, 5, and 6), 20 g of the supplement was consumed with breakfast. High intensity running performance, measured as distance covered at speeds >19 km/h, was greater during the last 15 min of game two following protein supplementation. Additionally, knee extensor concentric strength recovered quicker after the first game following protein supplementation. Endogenous antioxidant concentrations were greater following game two only in the protein-supplemented condition. Though soccer is a “power-endurance” sport rather than simply an endurance sport, these findings hold relevance for understanding the impact of protein intake on recovery and repeated performance in actual athletes.

Since 2014, additional work investigating the impact of protein consumption on biochemical markers of metabolic status, physiological fatigue, and recovery in endurance-trained athletes has been performed ( 54 ). For 5 weeks, elite or experienced marathon runners received either 33.5 g/day of whey protein or maltodextrin 30 min following the completion of each training session leading up to a race covering marathon distance. Blood samples were collected to assess biochemical markers of metabolism, muscle damage, and fatigue and took place prior to beginning the intervention, 1 day following the marathon, and 1 week following the marathon. These markers included CK, lactate dehydrogenase (LDH), AST, and ALT. Runners who supplemented with whey protein displayed decreased AST and ALT compared to maltodextrin-supplemented runners. CK and LDH, biochemical indicators of muscle damage, were significantly greater in the maltodextrin group post-marathon compared to the whey protein-supplemented group. Elevations in CK and LDH were still significant 1-week post-marathon in the maltodextrin group compared to the whey protein group ( 54 ). The whey protein group also showed significantly decreased triglycerides (TG) and total cholesterol (TC) compared to the maltodextrin group post-marathon. The maltodextrin group actually showed increased TC levels. Only the whey protein group showed significant decreases in LDL post-marathon and at 1 week post-marathon ( 54 ). The authors suggested that the decrease in TC seen in whey-supplemented runners may indicate that cholesterol was more efficiently converted to steroid hormones, resulting in improved physiological recovery and adaptations from the strenuous exercise bout. One week post-marathon, most biomarkers of damage and stress were still significantly lower in the whey protein group compared to the maltodextrin group ( 54 ). In addition to the more favorable biomarker profiles in the protein supplemented group, performance in the 12-min run/walk test was also greater in the whey protein-supplemented group 1-week post-marathon. Together, these results indicate that whey protein supplementation during marathon preparation and recovery, and that the supplement aids in attenuating metabolic and muscular damage. Daily dietary assessments were not included in this study ( 54 ), thus limiting possible practical applications or recommendations. As we have addressed previously, caloric deficit or daily protein consumption <1.4–1.6 g/kg may potentiate the effect of peri-workout protein consumption on recovery and subsequent performance. Further studies are necessary to elucidate the potential contribution of peri-workout whey protein ingestion on makers of muscle damage, recovery, and subsequent performance measures in endurance athletes.

In real-world sport performance situations, recovery and performance must be evaluated in the context of an accumulated effect. The ability to train consistently while remaining healthy is critical for continued progression and optimal performance. Endurance athletes in particular are at increased risk for upper respiratory tract infections ( 55 ). Factors contributing to this increased risk may include reduced immune function through low circulation of certain T-lymphocytes, especially during periods of increased volume and/or intensity of training. A diet providing a daily protein intake of 3 g/kg, including 60 g/day of casein protein, has been shown to be sufficient in returning circulating immune cell levels to those seen during lighter training periods, while a diet providing a daily protein intake 1.5 g/kg did not result in enhanced immune cell levels ( 56 , 57 ). Kephart et al. ( 45 ) have also found this beneficial effect on the immune system to extend to BCAA supplementation in doses of 12 g/d in trained cyclists.

Additionally, Rowlands et al. ( 58 ) found that consumption of ~64 g protein over 3 h following intense endurance exercise resulted in gene expression favorable for improving substrate, specifically fatty acid, mobilization and mitochondrial proteins for oxidation, especially in the electron transport chain. Post-exercise consumption of protein at levels thought to maximally stimulate MPS would potentially not have this same impact. Post-exercise protein consumption affects other systems and pathways and should not be considered only in terms of stimulating MPS. As further evidence of this notion, Levenhagen et al. ( 59 ) demonstrated that 10 g of casein protein enhanced MPS following 60 min of moderate intensity endurance exercise. Although this supplementation protocol stimulated MPS, subjects were found to be in negative whole-body protein balance. Because prolonged bouts of endurance exercise (i.e., >2 h) result in considerable oxidation of amino acids, specifically leucine, and intense or prolonged bouts of endurance exercise result in hypoxia-mediated small intestinal injury, negative whole-body protein balance may be common in endurance athletes ( 60 – 62 ). Because of this, protein requirements and recommendations for endurance athletes must consider more than MPS, especially since short-term increases in MPS do not fully explain the dynamics of long-term whole-body net protein balance and various training adaptations.

Conclusions and future direction

Overall, total daily energy and protein intake over the long term play the most crucial dietary roles in facilitating adaptations to exercise. However, once these factors are accounted for, it appears that peri-exercise protein intake plays a potentially useful role in optimizing physical performance and positively influencing the subsequent recovery processes. Challenges surround the definition of “performance” and the appropriate metrics by which to measure it based on desired outcomes. Difficulties also arise in attempting to define and quantify the concept of recovery. Additionally, both performance and recovery must be viewed in context depending on whether the emphasis is an immediate, short-term effect (i.e., 24 h or less) or a long-term training response.

It should also be noted that protein timing, whether it is pre-, during, or post-workout, is often framed within the context of bodybuilding (i.e., the singular goal of increasing skeletal muscle mass). It is evident that to use such a narrow frame of reference ignores the potential utility of protein timing within the context of endurance events (i.e., running, cycling, rowing, swimming, triathlon, etc.), as well as the vast majority of individual and team sports in which skeletal muscle hypertrophy is not a pre-eminent concern. For instance, if one competes in a weight-class sport (e.g., boxing, mixed martial arts, weightlifting, powerlifting, etc.), gains in body weight or lean body mass are often avoided; otherwise, the individual athlete would need to compete in a heavier weight class. In these situations, protein timing in particular may serve a useful role in recovery.

Translating research into practical application requires differentiation between novice or trained individuals, healthy normal weight or healthy overweight individuals, special populations, or those with certain metabolic or disease states. Here, we specifically focus on healthy, exercising individuals and limit our conclusions to these individuals. It is important moving forward that the study populations used are appropriate for the goals of the study and desired applications. For example, it is of little use to have a sample of recreationally-trained individuals if the goal is to understand performance in high-level athletes.

Though protein-containing meals result in increase of MPS on their own, as does resistance training, the timing of ingestion of protein around exercise further enhances this increase of MPS ( 63 , 64 ). It is worth noting that an upper limit for this acute dosing has not really been established, though there is evidence that 40 g of protein stimulates MPS to a greater degree than 20 g following whole-body resistance training ( 65 ). A dose higher than this, however, has not been included using the same timing paradigm. In reality, the “ideal” amount of peri-exercise protein consumption depends on many factors, including total caloric intake, total daily protein intake, training status of the individual, age of the individual, FFM, type of protein consumed, type and amount of other nutrients consumed, and the composition and timing of the most recent pre-training meal.

Much attention has been given to daily protein consumption and thresholds that must be met for peri-training protein consumption to exert additional benefit (>1.6–2.2 g/kg/d). As such, pre-, intra-, and post-training nutrient consumption present additional opportunities for athletes to contribute to their daily protein intake total and can be viewed in the context of ways to meet these “larger” daily needs by optimizing intake.

With regard to endurance exercise, protein consumption during exercise may not confer an immediate ergogenic benefit, especially when carbohydrate consumption is adequate. It may, however, aid in delaying central fatigue, reducing MPB, and contributing to a more positive, whole-body nitrogen balance. Additionally, protein consumption in and around intense or prolonged endurance activity may aid in reduction of upper respiratory tract infection incidence and improved immune system function. It may also aid in upregulating gene expression of proteins necessary for improving bioenergetic pathways. The impact of this on subsequent training sessions should not be dismissed and is an important part of improving performance.

The effect of protein consumption on resistance training is highly dependent on many variables not related to protein. The combination of peri-training protein consumption with inadequate or ineffective resistance training protocols will not maximize improvements in strength or hypertrophy. Resistance training protocol interventions must be of adequate intensity, volume, and frequency with an emphasis on progressive overload to produce results. Additionally, adequate training interventions coupled with calorie-restricted nutrition protocols may require increased protein intake of 2.3–3.1 g/kg FFM to yield desired improvements in strength, hypertrophy, or maintenance of FFM ( 10 ). Consideration must also be made for the age of resistance-trained individuals, as older adults require protein intake over and above that of their younger counterparts to receive the same benefits noted above ( 66 ).

In order to fully understand the role of protein (or any substrate for that matter) on performance, the practical application beyond the contrived training or recovery interventions presented must be addressed. Daily training schedules of athletes require an ongoing ability to recover and perform. As an example, most of the studies included in this area utilized a training protocol that took ~3–4 h per week, typically in moderately-trained individuals. For comparative purposes, a competitive athlete may spend 3–10 times this amount of time training per week (if not more). For this reason, the “window” for recovery should be considered to encompass each and every hour between training and competition. Protein dosing strategies need to take this into account. This becomes even more apparent when considering that the uniform distribution of protein throughout the day results in greater MPS than an uneven distribution even when total daily protein intake is equal ( 67 ). Arciero et al. ( 64 ) demonstrated the combination of resistance training and consumption of 4–6 meals per day containing 20–40 g of protein per meal resulted in positive changes in body composition and physical performance. These results suggest that the pattern of daily protein ingestion may also impact results from resistance training protocols and provides further evidence that we must look beyond the few hours following training to determine the impact that protein may have on performance and recovery. Further evidence in support of extending the “recovery window” concept are results from nighttime protein ingestion studies. Madzima et al. ( 68 ) found that consumption of 30 g of casein, 30 g of whey, or 33 g of carbohydrate 30 min prior to sleep resulted in increased resting energy expenditure and improved VO 2 the following morning. While no statistically significant changes were observed between groups, protein groups trended toward greater increases when compared to the carbohydrate group while morning fat oxidation was greatest in the casein supplemented group.

Taken together, these data demonstrate the need for a more comprehensive view and methods of measuring recovery. Increased sensitization of muscle to protein and nutrients for 24–72 h following training coupled with multiple weekly training sessions results in an on-going state of recovery. Because of this, we need to begin considering this longer stimulus window as an opportunity to maximize feeding, rather than as a reason why immediate post-workout ingestion may not be particularly important. In other words, consuming nothing post-workout would be an unwise strategy if the goal is to potentially optimize the adaptive response to exercise training.

Overall, there appears to be no adaptive advantage to avoiding protein intake in the peri-workout period. Stimulation of MPS in the acute period following training may not result in improvements in strength, hypertrophy, body composition, or performance without deliberate implementation of additional strategies during the prolonged recovery period. As such, this much broader view should be considered with regard to future investigations.

Author contributions

JA and SA conceived the topic. HC, MA, JA, and SA wrote the paper.

Conflict of interest statement

SA is on the Advisory Panel for Dymatize. JA is the CEO of the International Society of Sports Nutrition—an academic non-profit that receives grants in part from companies that sell dietary protein.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer CK declared a past co-authorship with several of the authors SA and JA to the handling Editor.