Post-prandial muscle protein synthesis rates following the ingestion of pea-derived protein do not differ from ingesting an equivalent amount of milk-derived protein in healthy, young males

A total of 24, healthy, recreationally active males aged 18–35 years were recruited to participate in this parallel-group, double-blind, randomized controlled trial to compare the impact of ingesting 30 g pea and 30 g milk-derived protein on post-prandial muscle protein synthesis rates in vivo in humans. As we provided the same absolute amount of protein (30 g), we decided to select only a single sex in the present study, to limit the range of the amount of protein provided when expressed per kg muscle mass. Participants were recreationally active and generally performed between 2 and 4 exercise sessions per week in various sports (e.g., soccer, basketball, weight lifting, running, and cycling), but were not involved in any structured progressive exercise training regimen. This study was part of a larger trial registered at the International Clinical Trials Registry Platform (NTR6548) and was conducted between June 2017 and April 2019 at Maastricht University in Maastricht, The Netherlands (see Supplemental Fig. 1 for the CONSORT (Consolidated Standards of Reporting Trials) flow diagram, indicating the specific comparison that the current study was based on). The data of the milk-derived protein group were used in various comparisons and, as such, have been published previously, as well as the procedures applied in this trial [23, 33]. All participants were informed about the purpose of the study, the experimental procedures, and possible risks before providing informed written consent to participate. The procedures followed were in accordance with the ethical standards of the medical ethics committee of Maastricht University Medical Centre + (METC 173001), and in accordance with the Helsinki Declaration of 1975 as revised in October 2013. The study was independently monitored and audited by the Clinical Trial Centre Maastricht.

Participants aged 18–35 years, with BMI > 18.5 and < 27.5 kg∙m⁻² underwent an initial screening session to assess eligibility. Height, weight, blood pressure, and body composition (by dual-energy X-ray absorptiometry; Discovery A, Hologic; (National Health and Nutrition Examination Survey—Body composition analysis (NHANES BCA) enabled) were determined. Participants were deemed healthy based on their responses to a medical questionnaire. The screening sessions and experimental trials were separated by at least 3 days.

Participants were randomly assigned to ingest a 400 mL beverage containing either 30 g milk-derived protein concentrate (MILK), or 30 g pea-derived protein concentrate (PEA). After beverage ingestion, the bottle was rinsed with 150 mL of water, which was also ingested by the participants. Milk-derived protein concentrate (Refit MPC80) was obtained from FrieslandCampina (Wageningen, the Netherlands), and pea-derived protein concentrate (Nutralys S85F) was supplied by Kellogg (Battle Creek, MI, USA). Participants were allocated to a treatment according to a block randomization list performed using a computerized randomizer (http://​www.​randomization.​com/​). An independent researcher was responsible for random assignment (n = 12 per group) and preparation of the study treatment beverages, which were sequentially numbered according to subject number. The beverages were provided in identical, non-transparent protein shakers.

Participants refrained from sports and strenuous physical activities (e.g., lifting heavy weights), and alcohol consumption for 3 days prior to the experimental trial. In addition, all participants were instructed to complete a food and activity record for 3 days prior to the experimental trial (see Supplemental Table 1 for an overview of participants’ habitual food intake in the 3 days prior to the experimental trial). The evening before the trial, all participants consumed a standardized meal containing 2.8 MJ, with 20% energy provided as carbohydrate, 65% as fat, and 15% as protein, before 10:00 PM after which they remained fasted.

The procedures applied in this trial have previously been described elsewhere [23]. At ~ 7:30 AM, participants arrived at the laboratory in an overnight post-absorptive state. A cannula was inserted into an antecubital vein for stable isotope amino acid infusion. A second cannula was inserted retrogradely into a dorsal hand vein on the contralateral arm for arterialized blood sampling. To obtain arterialized blood samples, the hand was placed in a hot box (60 °C) for 10 min prior to blood sample collection.

After taking a baseline blood sample (t = -180 min), the plasma phenylalanine pool was primed with a single dose of L-[ring-¹³C₆]-phenylalanine (2.25 µmol∙kg⁻¹). Thereafter, a continuous intravenous infusion of L-[ring-¹³C₆]-phenylalanine (0.05 µmol∙kg⁻¹∙min⁻¹) was initiated (t = − 180 min) using a calibrated IVAC 598 pump (San Diego, CA, USA). Subsequently, arterialized blood samples were collected at t = − 90, − 60 and − 30 min. At t = 0 min, an arterialized blood sample was obtained as well as a muscle biopsy from the M. vastus lateralis. Immediately following the muscle biopsy, participants ingested a 400 mL beverage corresponding to their randomized treatment allocation, i.e., MILK (n = 12), or PEA (n = 12). To minimize dilution of the steady-state plasma L-[ring-¹³C₆]-phenylalanine precursor pool, the phenylalanine content of the protein drink was enriched with 3.85% L-[ring-¹³C₆]-phenylalanine. Arterialized blood samples were then collected at t = 15, 30, 45, 60, 90, 120, 150, 180, 210, 240, and 300 min after protein ingestion in the post-prandial period. Blood samples were collected into EDTA-containing tubes and centrifuged at 1200 g for 10 min at 4 °C. Aliquots of plasma were frozen in liquid nitrogen and stored at − 80 °C. Second and third muscle biopsies from the M. vastus lateralis were collected at t = 120 and t = 300 min to determine post-prandial skeletal muscle protein synthesis rates over the 0–120, 120–300, and 0–300 min post-prandial periods. Muscle biopsy collection was alternated between legs and obtained with the use of a 5 mm Bergström needle [34], custom-adapted for manual suction. Samples were obtained from separate incisions from the middle region of the M. vastus lateralis, ~ 15 cm above the patella and ~ 3 cm below entry through the fascia. Local anesthetic (1% xylocaine with adrenaline 1:100,000) was applied to numb the skin and fascia. Muscle samples were freed from any visible non-muscle material, immediately frozen in liquid nitrogen, and stored at − 80 °C until further processing. When the experimental protocol was complete, cannulae were removed and participants were provided with food and monitored for ~ 30 min before leaving the laboratory. For a schematic representation of the infusion protocol, see Fig. 1.

Batch-specific nitrogen contents for milk- and pea-derived protein concentrates were provided by the manufacturers. The protein content of the milk-derived protein was determined as nitrogen content × 6.38, and the protein content of pea-derived protein was determined as nitrogen × 6.25 [35, 36]. Amino acid contents of the protein powders were determined by acid hydrolysis in triplicate, and subsequent analysis of the free amino acids using ultra-performance liquid chromatography–mass spectrometry (UPLC–MS; ACQUITY UPLC H-Class with QDa; Waters, Saint-Quentin, France), as previously described [23]. The amino acid composition of the protein powders are presented in Table 2.

Plasma glucose and insulin concentrations were analyzed using commercially available kits (ref. no. A11A01667, Glucose HK CP, ABX Diagnostics, Montpellier, France; and ref. no. HI-14 K, Millipore, St. Louis, MO, respectively). Plasma amino acid concentrations were determined by UPLC–MS, as previously described [23].

Plasma L-[ring-¹³C₆]-phenylalanine enrichments were determined by gas chromatography–mass spectrometry (GC–MS; Agilent 7890A GC/5975C MSD; Agilent Technologies), as previously described [23]. In short, the free amino acids from deproteinized plasma samples were purified using cation exchange resin columns (AG 50W-X8, mesh size: 100–200, ionic form: hydrogen (Bio-Rad Laboratories, Hercules, CA, USA)), and subsequently converted to their tert-butyl dimethylsilyl (TBDMS) derivative before analysis by GC–MS.

Basal (post-absorptive) muscle protein synthesis rates were assessed to confirm that protein ingestion increases muscle protein synthesis rates. The single biopsy approach was applied to assess post-absorptive muscle protein synthesis rates without the need to collect an additional muscle biopsy [37]. In short, plasma protein obtained prior to tracer infusion (t = − 180 min) was used to determine background L-[ring-¹³C₆]-phenylalanine enrichments. For this purpose, the plasma sample was precipitated by adding perchloric acid. Subsequently, similarly as for the myofibrillar protein fraction, the denaturized plasma protein pellet was hydrolyzed, passed over a cation exchange resin column (AG 50W-X8, mesh size: 100–200, ionic form: hydrogen (Bio-Rad Laboratories, Hercules, CA, USA)), and the resulting amino acid samples were derivatized to their N(O,S)-ethoxycarbonyl-ethylesters before being measured by gas chromatography-combustion-isotope ratio mass spectrometry (GC-IRMS; Mat 253, Thermo Scientific, Bremen, Germany) using a DB5MS (30 m) column (Agilent technologies, Santa Clara, Ca, USA), as previously described [23].

Muscle analysis for the determination of muscle protein-bound L-[ring-¹³C₆]-phenylalanine enrichments has previously been explained in detail [23]. In short, a piece of wet muscle (~ 50–70 mg) was homogenized and a myofibrillar protein-enriched fraction was obtained by removal of the collagen enriched fraction. Subsequently, the amino acids from the resulting dried myofibrillar protein-enriched fractions were liberated by adding 2 mL of 6 M HCl and heating to 110 °C for 16 h, passed over a cation exchange resin column (AG 50W-X8, mesh size: 100–200, ionic form: hydrogen (Bio-Rad Laboratories, Hercules, CA, USA)), and derivatized to their N(O,S)-ethoxycarbonyl-ethylesters. The ratio of ¹³C/¹²C of myofibrillar protein-bound phenylalanine was determined using GC-IRMS.

The plasma free and muscle protein-bound L-[ring-¹³C₆]-phenylalanine enrichments were used to calculate fractional myofibrillar protein synthesis rates (%∙h⁻¹). This calculation was performed by the standard precursor-product equation [38]:where E_b is the increment in myofibrillar protein-bound L-[ring-¹³C₆]-phenylalanine enrichment (mole% excess, MPE) during the tracer incorporation period, and t is the tracer incorporation time in h. Weighted mean plasma L-[ring-¹³C₆]-phenylalanine enrichments were calculated by taking the measured enrichment between consecutive time points and correcting for the time between these sampling time points (E_precursor). For calculation of post-prandial FSR, skeletal muscle biopsy samples at t = 0, 120 and 300 min were used. For the calculation of basal FSR, E_b2 represented the protein-bound L-[ring-¹³C₆]-phenylalanine enrichments in muscle at t = 0 min, and E_b1 represented the protein-bound L-[ring-¹³C₆]-phenylalanine enrichments in plasma protein at t = − 180 min.

A sample size calculation was performed with differences in post-prandial myofibrillar FSRs between the 2 treatments as primary outcome measure. Based on previous work in this area, a sample size of 12 participants per treatment, including a 10% dropout rate was calculated using a power of 80%, a significance level of 0.05, a difference in FSR of 0.008%∙h⁻¹ (or ~ 20% when expressed as relative difference, e.g., 0.040 vs 0.048%∙h⁻¹) [39], and a within-group standard deviation of 0.0065%∙h⁻¹ (or ~ 16%) [40, 41].

The primary outcome, post-prandial (0–300 min) muscle protein synthesis rates between the two treatments, was analyzed by independent samples t-test. Likewise, basal post-absorptive, (− 180–0 min) and post-prandial myofibrillar protein synthesis rates during the early (0–120 min) and late (120–300 min) post-prandial period were analyzed by independent samples t-test. As secondary analyses, a two-way repeated measures ANOVA was performed to evaluate changes over time and the increase in post-prandial muscle protein synthesis rates above basal post-absorptive rates. Plasma glucose, insulin, and amino acid concentrations and amino acid enrichments over time were compared between groups using a two-way (time x treatment) repeated measures ANOVA, with time as within-subjects factor, and treatment as between-subjects factor. In case a significant time x treatment interaction was observed, post-hoc analyses were performed to determine significant differences between treatments for each time point. Participants’ characteristics, plasma glucose, insulin, and amino acid concentrations, expressed as peak values, time to peak and iAUC, were analyzed by independent samples t-test to locate differences between groups. Statistical analyses were performed with a software package (IBM SPSS statistics for Windows, version 26.0, IBM Corp., Armonk, NY, USA). Means were considered to be significantly different for P values < 0.05. Data are expressed as means ± SD. Except for plasma insulin concentrations (n = 11 for MILK), no missing values were present for any of the outcome parameters.

Twenty-four healthy, recreationally active males (24 ± 3 years; 1.77 ± 0.06 m; 71.6 ± 8.9 kg) volunteered to participate in this parallel-group, double-blind, randomized controlled trial (Table 1).

Plasma glucose concentrations were maintained following protein ingestion, with no differences between treatments (time x treatment: P = 0.27; Fig. 2A). Plasma insulin concentrations increased following protein ingestion, with no differences between the MILK and PEA treatment group over time (time x treatment: P = 0.32; Fig. 2B). Similarly, peak plasma insulin concentrations (28 ± 8 vs 25 ± 7 mU∙L⁻¹, respectively; independent samples t-test: P = 0.34), and post-prandial plasma insulin availability (iAUC) did not differ following MILK vs PEA ingestion (1058 ± 331 vs 797 ± 498 mU∙300 min∙L⁻¹, respectively; independent samples t-test: P = 0.16).

Plasma EAA concentrations increased following protein ingestion, with a greater rise in circulating EAA concentrations following MILK vs PEA ingestion (time x treatment: P = 0.03; Fig. 3A). Plasma EAA concentrations were increased above basal post-absorptive concentrations for the entire 300 min post-prandial period after MILK and PEA ingestion. In accordance with the significant time x treatment interaction, peak plasma EAA concentrations following MILK vs PEA ingestion were reached at 36 ± 10 min and 56 ± 32 min (independent samples t-test: P = 0.05), reaching levels of 1871 ± 124 and 1601 ± 162 µmol∙L⁻¹ (independent samples t-test: P < 0.001), respectively. The overall increase in plasma EAA availability over the entire 300 min post-prandial period, expressed as iAUC, was ~ 48% greater for MILK vs PEA (151 ± 31 vs 102 ± 15 mmol∙300 min∙L⁻¹; independent samples t-test: P < 0.001; Fig. 3B).

The post-prandial increase in plasma leucine concentrations following protein ingestion (Fig. 3C) differed between MILK vs PEA (time x treatment: P < 0.01). Plasma leucine concentrations increased for the entire 300 min post-prandial period following ingestion of both MILK and PEA. In accordance with the significant time x treatment interaction, peak plasma leucine concentrations were ~ 25% higher for MILK vs PEA (353 ± 45 vs 282 ± 30 µmol∙L⁻¹, respectively; P < 0.001) and were reached 46 ± 43 and 58 ± 31 min after protein ingestion, respectively (independent samples t-test: P = 0.47). The overall increase in plasma leucine availability over the entire 300 min post-prandial period, expressed as iAUC, was ~ 44% greater for MILK vs PEA (36 ± 7 vs 25 ± 4 mmol∙300 min∙L⁻¹; independent samples t-test: P < 0.001; Fig. 3D).

The post-prandial increase in plasma lysine concentrations following protein ingestion was not different following MILK vs PEA ingestion (time x treatment P = 0.33; Fig. 3E). Plasma lysine concentrations increased for 240 and 210 min after MILK and PEA ingestion, respectively. Peak plasma lysine concentrations were not different following MILK vs PEA ingestion (370 ± 29 vs 339 ± 50 µmol∙L⁻¹, respectively; independent samples t-test: P = 0.08), but were reached ~ 16 min earlier (34 ± 7 vs 50 ± 21 min after protein ingestion respectively, independent samples t-test: P = 0.02). Peak plasma lysine concentrations increased ~ 137% above baseline values for MILK, and ~ 106% above baseline for PEA. Consequently, the overall increase in plasma lysine availability over the entire 300 min post-prandial period, expressed as iAUC, was ~ 25% greater for MILK vs PEA (25 ± 8 vs 20 ± 4 mmol∙300 min∙L⁻¹; independent samples t-test: P = 0.03; Fig. 3F).

The post-prandial increase in plasma methionine concentrations following protein ingestion was significantly greater following MILK vs PEA ingestion (time x treatment: P < 0.001; Fig. 3G). Plasma methionine concentrations increased for 240 and 90 min after MILK and PEA ingestion, respectively. After which methionine concentrations became lower when compared to post-absorptive values in the PEA group. In accordance with the significant time x treatment interaction, peak plasma methionine concentrations were ~ 114% greater for MILK vs PEA (60 ± 5 and 28 ± 4 µmol∙L⁻¹, independent samples t-test: P < 0.001), and reached ~ 30 min after protein ingestion (34 ± 9 vs 35 ± 22 min; independent samples t-test: P = 0.86). As a result, peak plasma methionine concentrations increased ~ 190% above baseline values for MILK, but only increased ~ 33% above baseline values for PEA. The overall increase in plasma methionine availability over the entire 300 min post-prandial period, expressed as iAUC, was several fold greater for MILK vs PEA (4.7 ± 1.4 vs − 0.6 ± 0.4 mmol∙300 min∙L⁻¹; independent samples t-test: P < 0.001; Fig. 3H).

In general, post-prandial increases in plasma amino acid concentrations revealed significant differences over time following MILK vs PEA ingestion for most amino acids (Supplemental Fig. 2; time x treatment: P < 0.05). The post-prandial increases in plasma alanine, BCAA, cystine, proline, threonine, tryptophan, tyrosine, and valine availability over the entire 300 min post-prandial period (iAUC) were greater for MILK vs PEA (independent samples t-test: P < 0.05), with an exception for plasma arginine, asparagine, glycine, and ornithine, which were lower for MILK vs PEA (independent samples t-test: P < 0.05, Supplemental Fig. 2).

Plasma L-phenylalanine concentrations and L-[ring-¹³C₆]-phenylalanine enrichments over time are presented in Fig. 4A and 4B, respectively. Plasma L-[ring-¹³C₆]-phenylalanine enrichments over time did not differ following MILK vs PEA ingestion during the post-prandial period (time x treatment: P = 0.18). Mean plasma L-[ring-¹³C₆]-phenylalanine enrichments averaged 7.11 ± 0.65 and 6.63 ± 0.58 MPE during the basal post-absorptive period (independent samples t-test: P = 0.07), and 6.64 ± 0.53 and 6.33 ± 0.27 MPE throughout the 5 h post-prandial period (independent samples t-test: P = 0.08) following MILK and PEA ingestion, respectively. Myofibrillar protein-bound L-[ring-¹³C₆]-phenylalanine enrichments were higher following ingestion of MILK and PEA from 0.0032 ± 0.0032 and 0.0028 ± 0.0029 MPE at t = 0 min, to 0.0115 ± 0.0041 and 0.0104 ± 0.0035 MPE at t = 120 min, reaching 0.0214 ± 0.0049 and 0.0205 ± 0.0047 MPE at t = 300 min after protein ingestion, respectively.

Post-absorptive fractional myofibrillar protein synthesis rates averaged 0.014 ± 0.014 and 0.015 ± 0.017%∙h⁻¹ in the MILK and PEA experiment, with no differences between the groups (independent samples t-test: P = 0.94). The primary outcome, post-prandial muscle protein synthesis rates (0–300 min), did not differ between MILK vs PEA, (0.053 ± 0.013 vs 0.053 ± 0.017%∙h⁻¹, independent samples t-test: P = 0.96, Fig. 5), neither did the increase above basal post-absorptive rates (change scores, independent samples t-test: P = 0.99). In addition, muscle protein synthesis rates did not differ for the early (0–120 min; independent samples t-test: P = 0.71), and late (120–300 min; independent samples t-test: P = 0.55) post-prandial period. Secondary analyses using two-way repeated measure ANOVA showed that protein ingestion increased myofibrillar protein synthesis rates to 0.059 ± 0.024 and 0.054 ± 0.031%∙h⁻¹ during the early post-prandial period (0–120 min) and to 0.049 ± 0.017 and 0.053 ± 0.015%∙h⁻¹ during the late post-prandial period (120–300 min) in MILK and PEA, respectively (main effect of time P < 0.001), with no time x treatment interaction.