Hemp and buckwheat are valuable sources of dietary amino acids, beneficially modulating gastrointestinal hormones and promoting satiety in healthy volunteers (2024)

Subjects

For this study, ten healthy volunteers (males and females), 18–65years old, with BMI 18–35kgm⁻², non-vegetarian, non-smokers, with no known allergies and taking no prescribed medication were recruited. The study started in March 2013. All volunteers completed a medical questionnaire at the Human Nutrition Unit (HNU) assisted by the general practitioner (GP). Volunteers’ height was measured to the nearest 0.1cm with the use of a stadiometer (Holtain Ltd, Crymych, Dyfed, UK) and the weight was measured to the nearest 100g on a digital scale (DIGI DS-410; CMS Weighing Equipment, London, UK). The screening included a Skin Prick Test (SPT), and in vitro specific Immunoglobulin E (IgE) test (for all the flours used) performed at Aberdeen Royal Infirmary Hospital Foresterhill. The study was approved by the North of Scotland Research Ethics Service, with REC reference number: 13/NS/0006. After signing a consent form, the volunteers participated in seven visits to the HNU at The Rowett Institute (RI), Aberdeen, UK, from which, one was the screening visit and six were the 1-day dietary interventions (with a washout of minimum 2 weeks between interventions). The intervention meals were served in the following order: fava bean, lupin, meat, green pea, buckwheat and hemp.

Study intervention visits

Prior to each interventions the volunteers recorded their habitual diet in a 7-day weighed food diary. The volunteers were instructed to come after an overnight fast, and instructed to consume the test meal within 30min. Five hours later, an ad libitum lunch meal was offered. Besides visual analogue scales (VAS) questionnaires, blood samples were taken throughout a 24-h period. Blood samples were collected by using a cannula into lithium heparin tubes at baseline (0min) and after 30, 60, 90, 120, 150, 180, 300min and 24h from the commencement of the intervention meal (Fig.1). The plasma was separated (3300×g; 15min; 4°C) within 5min of collection and stored at − 70°C until analysis. Six different 10-cm VAS were used to assess satiety before the test meal and at 30, 60, 90, 120, 150, 180 and 300min after the test meal, as well as after the consumption of the ad libitum lunch. The VAS questionnaires included the following six common questions [29] as previously described [21].

Open in a separate window

Fig. 1

Intervention day sampling diagram for bloods and questionnaires

Test meals

Each test meal (bread roll weight) was designed to deliver 30g protein from the five different plant-based sources (in flour form of green pea, lupin, hemp, buckwheat or fava beans) and, respectively, meat and used same amount of wheat flour (125g) for each breadroll including the meat meal. Each test meal was served with 30g butter. Commercially available flours were obtained as follows: strong wheat flour was purchased from Tesco Stores Ltd. (Cheshunt, UK); buckwheat flour was purchased from Arrowhead Millls, Inc. (Melville, USA); hemp flour was purchased from Yorkshire Hemp Ltd. (Driffield UK); fava bean flour was purchased from The Barry Farm (Wapakoneta USA); green pea flour was purchased from Bob’s Red Mill Natural Foods (Milkwaukie, USA); lupin flour was purchased from Terrena Lup’ Ingredients (Martigne Ferchaud, France) and the lean beef frying steak Tesco Stores Ltd. The number of rolls served for each intervention meal varied: two rolls (lupin, hemp), and three rolls (fava bean, green pea, buckwheat). The meat meal consisted in a fried meat (containing 30g protein) and a wheat bread roll served with 30g butter. The protein content was measured as crude protein content, %N and multiplied with a 6.25 factor. Same quantity and composition of ad libitum lunch (pasta Bolognese) containing 15% protein, 30% fat and 55% carbohydrate was provided 5 h after the test meal, and instructed to “eat until comfortably full.” The test meals and ad libitum ingredients are presented in Tables S 1 A and B from Online Resource ESM_1.

Composition of test meals

Macronutrients (protein, including amino acid composition (excluding Tryptophan); fibre measured as soluble and insoluble non-starch polysaccharide and fat), micronutrient minerals and phytochemical metabolites were measured by standard published protocols as described in our previous publication [30].

Compliance and metabolic profile

The volunteers attended the HNU on 6intervention days, where they consumed the individual test meals. The blood sampling was performed as described in Fig.1. Blood samples were used to measure the plasma amino acids by isotope dilution (different methodology from the food amino acid determination), with a gravimetric approach (to give μM) described previously [31], for hom*ocysteine (HCys) [32] and arginine [33]. Total amino acid concentrations were used as a biomarker for protein metabolism. All the plasma metabolites lipid profile, glucose, insulin, other phytochemicals, were measured as described before [34–36]. Based on preliminary VAS scores, the total PYY, total ghrelin, and active GLP-1 were analysed from the plasma samples collected (enzyme inhibitors were added to the plasma) from fava bean, meat, buckwheat and hemp meals using ELISA (Millipore) kits as previously described [21].

Statistical analysis

The sample size of ten volunteers was predicted to detect a treatment effect 1.4 times the magnitude of the within volunteer spread, with 80% power at the 5% significance level. For glucose and insulin, the within volunteer spread was approximately 10% (based on previous studies), which gave the study power to detect a treatment shift of ca. 14%. Data were analysed by analysis of variance (ANOVA) of area under the curve (AUC) and iAUC of data collected over multiple time points following consumption of the test meal, as well as the data at each time point individually. The ANOVA included terms for volunteer, baseline value and diet. A version of the ANOVA in which the diet term was split into a contrast between the meat diet and the plant diet, and the variation among the plant diets was also examined. Differences among the diets were assessed by post hoc tests with Tukey's adjustment for multiple comparisons. Data were checked for the assumptions of normality and constant variance, and were log transformed before analysis where appropriate. All analyses were carried out using Genstat 17 (Lawes Agricultural Trust, VSN international Ltd, Hemel Hempstead, UK). The plant metabolites (non-nutrients) from test meals, plasma phenylpropanoid pathway and products of protein and carbohydrate metabolism data was analysed by principal component analysis (PCA), unit variance (UV)-scaled using SIMCA 14.1 (Umetrics, Cambridge). The correlation between plasma metabolites and gastrointestinal hormones was analysed by partial least squares-discriminant analysis (PLS-DA) SIMCA 14.1 (Umetrics, Cambridge). The effect of test meals over time and between test meals (plasma) as well as the differences between the amounts of amino acids eaten for each interventions were assessed by two‐sided post hoc t tests. A version of the analysis which accounted for the varying protein intake was also conducted, by including a term for amount consumed and its interaction with the protein type.

Subjects baseline characteristics

Ten subjects (n = 3 men; n = 7 women) with a mean age of 42 ± 11.8 (years) and BMI of 26 ± 5.8 (kg m⁻²) (reflecting the average BMI of the Scottish adult population) were recruited for this study.

Composition of the study test meals

Non-nutrient plant metabolites

The ten most abundant phytophenols found in each test meal are presented in Fig.2. Buckwheat meal had the highest amount of total plant metabolites measured (164.6mg/portion), followed by hemp and fava bean meals (82.6 and 81.1mg/portion, respectively) and PCA (Fig.2) of all metabolites showed clear discriminations of buckwheat and hemp from the rest of the meals. The clustering of meat with the green pea, lupin and fava bean meals in the top left quadrant is likely to be due to the similarity of the profiles with the wheat portion of the meals; specifically the presence of ferulic, benzoic and sinapic acid. However, ferulic acid was the most abundant phytophenol in the fava bean meal and was significantly higher (p < 0.001) in comparison with the meat meal. The clear segregation of buckwheat and hemp meal is likely to be due to the presence of anthocyanins (pelargonidin) being present only in the buckwheat meal and syringaresinol being significantly higher in the hemp meal (p < 0.001). The complete profile of plant metabolites is presented in Table S2 A-D from supplementary data ESM_1.

Open in a separate window

Fig. 2

Principal component analysis (PCA), (unit variance (UV) scaled) of plant non-nutrients (measured in triplicate) from test meals was carried out and the first two component scores were plotted. Test meal plant metabolites composition (10 most abundant phytophenols) in mg/test meal ± SD, n = 3

Plasma metabolites

For the meat meal, a higher average intake (93.5% of the meal served) was recorded. This was followed by lupin (80.9%) and hemp (69.7%). The least quantity consumed of all the intervention meals was fava bean, with an average intake of 65.0% of the meal served; green pea and buckwheat meals had similar (p > 0.05) average intakes of 67.1 and 69.7%, respectively.

Overall, the quantity of individual amino acids consumed is significantly higher in the case of the meat meal in comparison with other meals, with very few exceptions where there is no significant difference (Fig.3). Despite the significantly higher quantity of amino acids being consumed, the amount of individual amino acids measured in plasma over 300min following meals consumption was not significantly higher for the meat meal compared to the plant-based meals, with few exceptions (see Fig.3). These included significantly lower levels of His for the fava bean and buckwheat meals, Lys for the fava bean, hemp and buckwheat meals, Tyr, Ile, leu, Phe and Pro for the buckwheat and Met for all meals except hemp (see Fig.3).

Open in a separate window

Fig. 3

The average (n = 10 ± SEM) of individual amino acid eaten by the volunteers from the intervention meals and a statistical comparison (t test) of each amino acid from the individual plant-based meals vs meat meal (LEFT side). The plasma iAUC over 300min for individual amino acids following consumption of test meals and a statistical comparison (ANOVA) of each amino acid from the individual plant-based meals vs meat meal (RIGHT side)

The average (n = 10) plasma concentration for each of the amino acids (in nmoles/g plasma) following consumption of test meals at 30, 60, 90, 120, 150, 180 and 300min together with the AUC and iAUC as well as the plasma amino acid concentrations over 300min are presented in Table S3 from Online Resource ESM_1. Overall ANOVA found a significant effect of the diet on iAUC for the plasma amino acids with few exceptions such as Ala, Gly, Asp, Glu, Gln, and Trp where no significant effect was found. When ANOVA compared iAUC for meat vs the other plant-based meals, no significant difference was found for the amino acid plasma concentrations over 300min with the exception of Leu, Met, Asp, Lys, Arg, His and Tyr, where significant differences were found. When ANOVA compared iAUC between the plant-based meals, again significant differences were found in the amino acid plasma concentrations over 300min with the exception of Ala, Gly, Thr, Asp, Glu, Gln and Trp where no significant differences were found. ANOVA analysis showed that iAUC for the plasma concentrations of all BCAAs (Val, Leu Ile) over 300min only after consumption of buckwheat meal from the plant-based meals was significantly lower (p < 0.001) when compared with plasma distribution after meat meal consumption, exception was iAUC for Leu (p < 0.05) after hemp consumption (see Fig.4). In case of AAAs, ANOVA analysis showed that iAUC for plasma concentrations of Phe was significantly lower after buckwheat (p = 0.05) and green pea (p < 0.05); Tyr was significantly lower after buckwheat (p < 0.05) and lupin (p < 0.001) consumption when compared with meat. ANOVA analysis showed no significant differences in Trp plasma concentrations following plant-based meal when compared with meat (Fig.4).The concentrations of di- and polyamines (spermidine, cadaverine, putresine) as well as tyramine and histamine were found to be significantly increased (p < 0.001) in 24h plasma samples after fava bean meal consumption (Table S5).

Open in a separate window

Fig. 4

Aromatic amino acids (phenylalanine, tyrosine and tryptophan) and branch chain amino acids (valine, leucine and isoleucine) post-prandial destitutions over 300min following the consumption of test meals, as average n = 10 ± SEM and iAUC of each amino acid from the test meals vs meat meals (ANOVA analysis)

The major phenylpropanoid‐derived metabolites have been shown to arise from microbial fermentation of protein [37]. Some of these metabolites have been shown to have beneficial effects for human health, including anti‐oxidant, anti‐inflammatory and anti‐cancer activities [38, 39], however, some of these compounds can be toxic, including acting as co‐carcinogens [40, 41]. On this study, the volunteers were fed high-protein meals, so it was also important to measure the main metabolites derived from the aromatic amino acids to assess potential differences in their plasma concentrations. Moreover, the 24-h plasma sample gives an indication of products produced by microbial metabolism. It was observed that only after consumption of the fava bean meal that there was a significant increase (p < 0.001) of phenylpropionic acid, phenylacetic acid and phenylpyruric acid in the 24-h plasma sample suggesting higher levels of microbial metabolism of phenylalanine (Table S4 A, ESM_1). After the consumption of the buckwheat meal, a significant increase (p < 0.05) of phenylpropionic acid and phenylacetic acid was observed in plasma at 3- and 5 h (Table S4 A, ESM_1). Again, only after consumption of the fava bean meals was a significant increase (p < 0.001) of plasma 4-hydroxyphenylacetic acid, 4-hydroxyphenylpyruric acid, 4-hydroxyphenyllactic acid and p-cresol in the 24-h plasma sample observed, indicating high microbial metabolism of tyrosine (Table S4 B, ESM_1). There was a significant effect of diet (p = 0.009) and time (0 vs 24h, p = 0.025) on indole-3-propionic acid (IPA). Specifically, there were significant increases in indole-3-propionic acid concentrations in the 24-h plasma after lupin (p = 0.028), green pea (p = 0.026), and buckwheat (p = 0.034) meal consumption (Table S4 C, ESM_1). Following the consumption of the fava bean meal, there was a significant increase (p < 0.001) of indole-3-acetic acid, indole-3-carboxylic acid and indole-3-lactic acid in the 24-h plasma sample compared with baseline (Table S4 C, ESM_1). Following the consumption of the meat meal, there was significant increase (p < 0.01) of plasma indole-3-carboxylic acid in the 24-h plasma sample and indole-3-pyruvric acid throughout the 5-h postprandial period (Table S4 C, ESM_1).

The highest number of overall plasma metabolites with a significant increase in concentration (within 5 h and mainly at 3 h) was observed following consumption of the buckwheat meal (Table S5). Following the consumption of fava bean meals was observed the highest number of plasma metabolites with a significant change after 24h (Table S5) suggesting that fava bean meal was metabolised late on GI tract at colon level. Ferulic acid was the metabolite with a significant increase at the 3 h plasma samples following the consumption of all the test meals apart from lupin meal, suggesting that it could be bioavailable from the wheat flour ingredient of the meals. In case of the lupin meals, the concentrations of ferulic acid in plasma at 180min postprandially is comparable with the concentrations measured in the other test meals, but the increase is not significant (0 vs 180min).

Gastrointestinal hormones and their correlation with plasma plant metabolites

Overall ANOVA analysis found no significant effects of the test meals on the ghrelin plasma concentration for AUC (p = 0.266), and mild significance for iAUC (p = 0.037). ANOVA found no significant changes in ghrelin concentration with time and between plant-based meals. However, following consumption of the hemp meal, the lowest ghrelin concentrations (lowest AUC) in the volunteers’ plasma was recorded and a significant effect on iAUC (p = 0.014, ANOVA) when plant-based meals vs meat meal were compared (Fig.5).

Open in a separate window

Fig. 5

Plasma average concentrations (pg/ml ± SEM), n = 10 for ghrelin, PYY, GLP1 and insulin before (0min) and 30, 60, 90, 120, 180 and 300min after the fava bean, meat, buckwheat and hemp test meals. (*) GLP1 represents significant plasma concentration differences between fava bean and buckwheat (p = 0.017 at 90min, p = 0.050 at 120min and p = 0.037 at 150min), hemp and buckwheat (p = 0.007 at 90min, p = 0.011 at 120min, and p = 0.051 at 150min), and meat and buckwheat (p = 0.005 at 300min). (*) Insulin represents significant plasma concentration differences at 90min between hemp and buckwheat (p = 0.009), fava bean (p = 0.005) and meat (0.0122). AUC and iAUC for the gut hormones for the test meals for 300min, overall ANOVA p values and ANOVA p values for the meat vs plants meals and plants vs plants meals

Overall ANOVA showed no significant time effect on PYY (p = 0.753), and no diet effect (p = 0.793), AUC (p = 0.400) and iAUC (p = 0.325). ANOVA analysis also found no significant changes in PYY concentration with time. After consumption of the hemp meal, plasma PYY had the lowest AUC in average and after consumption of buckwheat meal had the highest AUC amongst all the test meals (Fig.5).

For GLP-1, overall ANOVA analysis found significant effect of diet (p = 0.0045) and time (p < 0.001) and between diet effects (p = 0.0120), with buckwheat vs hemp (p = 0.049) and buckwheat vs fava bean (p = 0.011). Significant effect on overall AUC iAUC (p = 0.02). ANOVA analysis found significant plasma concentration differences between the fava bean and buckwheat meals (p = 0.017 at 90min, p = 0.050 at 120min and p = 0.037 at 150min), hemp and buckwheat meals (p = 0.007 at 90min, p = 0.011 at 120min, and p = 0.051 at 150min), and the meat and buckwheat meals (p = 0.005 at 300min). There were significant effects on AUC and iAUC (p = 0.01, ANOVA), with the AUC (51,228.20) after the hemp meal being significantly higher (Fig.5).

Overall ANOVA analysis found a time effect (p = 0.007) for insulin but no effect on AUC and iAUC (p = 0.157 and p = 0.173). ANOVA analysis found a mild diet effect at 30min (p = 0.048) and significant insulin plasma concentration differences at 90min. Insulin concentration after the hemp meal was significantly lower when compared with the buckwheat (p = 0.009), fava bean meal (p = 0.005) and meat (p = 0.012) meals and had the lowest, but not significant, AUC and iAUC.

PLS DA analysis highlighted several plasma metabolites potentially associated with the gastrointestinal hormones measured (Fig.6 and Tables S6 A–D). Plasma 4-hydroxyphenylpyruvic acid, indole-3-pyruvic acid, 5-hydoxy tryptophan, genistein and biochanin A were present in significantly higher amounts (p < 0.001) after hemp consumption (at 1 and 3 h) compared with meat and were positively associated with increased GLP-1 and PYY and negatively associated with increased insulin. Plasma 3-hydroxymandelic acid and luteolidin were positively associated with the effect of hemp on GLP-1 and ghrelin.

Open in a separate window

Fig. 6

See Also

The Benefits of Buckwheat - A Nutritionist's View

PLS DA analysis of each gastrointestinal hormone and the plant metabolites measured in plasma using LCMS/MS over 300min highlighting in each case the group of metabolites identified as associated with the effect of hemp meal on the gut hormones

Secoisoresinol, benzoic acid, 4-hydroxy mandelic acid were found to be present in significantly higher amounts (p < 0.001) after hemp consumption (at 1 and 3 h) when compared with meat and were positively associated with the decrease of insulin and ghrelin (Tables S6 A–D). Similarly, significantly higher plasma concentrations (p < 0.001) of coumestrol were observed following hemp consumption when compared with meat and was associated with increased PYY and decreased Insulin concentrations in plasma. Furthermore, p-coumaric acid, daidzein, matairesinol (p < 0.01), phenyl acetic acid, luteolin, apigenin, and pinoresinol (p < 0.001) were detected at significantly higher amounts in plasma at 1 and 3 h after consumption of the hemp vs meat meals and were associated with decreased ghrelin.

Plasma lipids, glucose, urea and hom*ocysteine

There was no significant diet effect on HDL (p = 0.359) and LDL (p = 0.280) when analyzed with ANOVA, there was no significant difference between meals for the AUC and iAUC values for HDL, p = 0.865, and p = 0.291, respectively (ANOVA). There was no significant difference between meals for the AUC, and iAUC values for LDL, p = 0.064, and p = 0.298, respectively (ANOVA). Overall ANOVA analysis found significant diet and time effect on postprandial plasma NEFA (p < 0.001, ANOVA) over 300min, plasma concentrations being significantly higher after consumption of hemp and lupin meals than other test meals. ANOVA found significant effect of diet on AUC for NEFA (p < 0.001, ANOVA), AUC for hemp and lupin meals being significantly higher than rest of the test meals and no significant effect on iAUC (p = 0.055). Significant effect on NEFA concentration at 90min (p = 0.005, ANOVA), NEFA plasma concentration after the hemp meals was significantly higher than after meat and fava bean meal, and NEFA plasma concentration after the lupin meal were significantly higher than after the green pea, fava bean and meat meals; NEFA plasma concentration after the lupin and hemp meals was significantly higher than after the green pea, fava bean, buckwheat and meat meals at 120min (p = 0.004, ANOVA), at 150min (p < 0.001, ANOVA) and 180min (p < 0.001, ANOVA); at 300min, NEFA plasma concentration after the lupin meal were significantly higher than after the green pea, fava bean and buckwheat meals and NEFA plasma concentration after the hemp meal was significantly higher than the buckwheat and green pea meals (p = 0.018, ANOVA). Overall ANOVA analysis found no significant diet effect on triglycerides (p = 0.060), AUC (p = 0.409) but a significant time effect (p < 0.001) and a mild effect on iAUC (0.041), significantly higher after the lupin meal iAUC (174.0) in comparison with fava bean, meat and buckwheat meals. The hemp and lupin meals are the highest in fat, however, we consider that the supplementation of the diet with these meals which are also high in insoluble fibre (NSP) should be done for a prolonged period in order to see any effect of this nutrient or their metabolites on the metabolic health biomarkers such as NEFA.

Overall ANOVA analysis found significant effects on diet and time on plasma glucose concentrations (p < 0.001), these being significantly higher after consuming of the buckwheat meal ANOVA analysis found a significant effect of the test meals on AUC and iAUC for glucose (p = 0.010, p = 0.032, respectively), AUC after the buckwheat meal was significantly higher than all other meals, iAUC was significantly higher after the buckwheat meal in comparison with the green pea, meat and hemp meals. After consumption of the buckwheat meal, plasma concentrations for glucose were significantly higher at 60min (p = 0.015, ANOVA), when compared with the lupin, hemp, green pea, and meat meals; at 120min (p = 0.028, ANOVA), when compared with all the other meals; and at 150min (p = 0.001, ANOVA), when compared with the hemp, fava bean, green pea and meat meals.

Overall ANOVA analysis found a significant diet (p < 0.010) and time (p < 0.001) effect on postprandial plasma urea over 300min. ANOVA analysis found no significant effect of test meals on AUC and significant effect for iAUC for urea (p = 0.091, p = 0.035, respectively), iAUC was significantly higher after the meat and lupin meals in comparison with the green pea, fava bean and buckwheat meals. After consumption of the lupin meal. plasma urea concentrations were significantly higher at 150 and 180min (p = 0.047, p = 0.015, respectively ANOVA), when compared with the buckwheat, fava bean and green pea meals; and at 300min (p < 0.001, ANOVA), plasma concentration of urea were significantly higher after consumption of the meat meals in comparison with the buckwheat, fava bean, green pea, hemp meals.

Overall ANOVA analysis found significant diet and time effect on postprandial plasma HCys (p < 0.001) over 300min, plasma HCys concentrations being significantly higher after the fava bean meal in comparison with all the other test meals. ANOVA found no significant effect of test meals on AUC and significant effect for iAUC for urea (p = 0.536, p < 0.001, respectively), iAUC was significantly higher after the fava bean meals in comparison with all the other meals. After consumption of the fava bean meals, plasma HCys concentration was significantly at 300min (p = 0.031, ANOVA) when compared with all the other test meals.

Satiety and hunger

Significant differences were found for all the VAS measures (hunger, fullness, desire, quantity, thirst, preoccupation) with time, from baseline to post-lunch (p < 0.001), data not shown. There were no significant differences between the intervention meals for most of the variables measured. The largest and only significant difference between the intervention meals occurs with hunger at 300min (p = 0.018), see Fig.7. Hunger score was significantly lower at 300min after the buckwheat meal (35.60) vs meat meal (48.50), lupin meal (51.10) and fava bean meal (51.70). There was no significant difference in consumption of ad libitum lunch (p = 0.977). The results showed mild significance (p = 0.04) for the effect of protein (with a coefficient of − 2.08) for the hunger VAS values for all intervention meals when adjusted for protein intake (using consumed protein amount as a covariate with ANCOVA). The source of the protein did not show significant difference for inducing increased satiety (p = 0.434).

Open in a separate window

Fig. 7

Average hunger score (mm ± SEM), n = 10, before (0min) and 30, 60, 90, 120, 180 and 300min after the test meals and at 330min (after the ad libitum) lunch. ANOVA Fisher’s LSD analysis for hunger mean scores at 300min after the test meals. (*) represents significant differences (p = 0.018) for the hunger scores between test meals

Study limitations

The present study has several limitations. A relatively small sample size of ten participants limits the conclusions that can be drawn from the study, suggesting the importance to replicate our findings in a larger sample size, however, this modest sample size does not weaken the significant differences which have been found. Additionally, the order the meals were consumed was not randomized, however, the interventions were properly spaced out (minimum 2 weeks between each) and, therefore, we do not estimate any carry over between treatments. Another limitation is that the volunteers were not consuming the entire test meals, but our analysis accounted for the varying protein intake by including a term for amount consumed.

Below is the link to the electronic supplementary material.

Conflict of interest

None of the authors had a personal or financial conflict of interest.

Ethics approval

The study was approved by the North of Scotland Research Ethics Service, with REC reference number: 13/NS/0006.

Consent to participate

Written consent was taken for each volunteer at the screening visit prior the intervention visits.

Consent for publication

All the authors have read and approved the manuscript for publication.

1. Popkin BM, Adair LS, Nq SW. Global nutrition and the pandemic of obesity in developing countries. Nutr Rev. 2012;70:3–21. [PMC free article] [PubMed] [Google Scholar]

2. Tilman D, Clark M. Global diets link environmental sustainability and human health. Nature. 2014;515:518–522. [PubMed] [Google Scholar]

3. Newgard CB, An J, Bain JR, et al. A branched-chain amino acid-related metabolic signature that differentiates obese and lean humans and contributes to insulin resistance. Cell Metab. 2009;9:311–326. [PMC free article] [PubMed] [Google Scholar]

4. Wang TJ, Larson MG, Vasan RS, et al. Metabolite profiles and the risk of developing diabetes. Nat Med. 2011;17(4):448–453. [PMC free article] [PubMed] [Google Scholar]

5. Würtz P, Soininen P, Kangas AJ, et al. Branched-chain and aromatic amino acids are predictors of insulin resistance in young adults. Diabetes Care. 2013;36(3):648–655. [PMC free article] [PubMed] [Google Scholar]

6. Tillin T, Hughes AD, Wang Q, et al. Diabetes risk and amino acid profiles: cross-sectional and prospective analyses of ethnicity, amino acids and diabetes in a South Asian and European cohort from the SABRE (Southall And Brent REvisited) Study. Diabetologia. 2015;58(5):968–979. [PMC free article] [PubMed] [Google Scholar]

7. de Mello VD, Paananen J, Lindstrom J, et al. Indolepropionic acid and novel lipid metabolites are associated with a lower risk of type 2 diabetes in the Finnish Diabetes Prevention. Study Sci Rep. 2017;7:46337. [PMC free article] [PubMed] [Google Scholar]

8. Wlodarska M, Luo C, Kolde R, et al. Indoleacrylic acid produced by commensal Peptostreptococcus species suppresses inflammation. Cell Host Microbe. 2017;22:25–37. [PMC free article] [PubMed] [Google Scholar]

9. Hwang IK, Yoo KY, Li H, et al. Indole-3-propionic acid attenuates neuronal damage and oxidative stress in the ischemic hippocampus. J Neurosci Res. 2009;87(9):2126–2137. [PubMed] [Google Scholar]

10. Karbownik M, Stasiak M, Zygmunt A, et al. Protective effects of melatonin and indole-3-propionic acid against lipid peroxidation, caused by potassium bromate in the rat kidney. Cell Biochem Funct. 2006;24:483–489. [PubMed] [Google Scholar]

11. Chimerel C, Emery E, Summers DK, et al. Bacterial metabolite indole modulates incretin secretion from intestinal enteroendocrine L cells. Cell Rep. 2014;9(4):1202–1208. doi:10.1016/j.celrep.2014.10.032. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

12. le Roux CW, Aylwin SJ, Batterham RL, et al. Gut hormone profiles following bariatric surgery favor an anorectic state, facilitate weight loss, and improve metabolic parameters. Ann Surg. 2006;243:108–114. [PMC free article] [PubMed] [Google Scholar]

13. Neary NM, Small CJ, Druce MR, et al. Peptide YY3–36 and glucagon-like peptide-17–36 inhibit food intake additively. Endocrinology. 2005;146:5120–5127. [PubMed] [Google Scholar]

14. Adrian TE, Ferri GL, Bacarese-Hamilton AJ, et al. Human distribution and release of a putative new gut hormone, peptide YY. Gastroenterology. 1985;89:1070–1077. [PubMed] [Google Scholar]

15. Batterham RL, Heffron H, Kapoor S, et al. Critical role for peptide YY in protein-mediated satiation and body-weight regulation. Cell Metab. 2006;4:223–233. [PubMed] [Google Scholar]

16. Helou N, Obeid O, Azar ST, Hwalla N. Variation of postprandial PYY 3–36 response following ingestion of differing macronutrient meals in obese females. Ann Nutr Metab. 2008;52:188–195. [PubMed] [Google Scholar]

17. Del Olmo-Garcia MI, Merino-Torres JF. GLP-1 receptor agonists and cardiovascular disease in patients with type 2 diabetes. J Diabetes Res. 2018;2018:4020492. [PMC free article] [PubMed] [Google Scholar]

18. Bodnariuc AM, Prud'homme D, Blanchet R, et al. Nutritional modulation of endogenous glucagon-like peptide-1 secretion: a review. Nutr Metab. 2016;13:92. [PMC free article] [PubMed] [Google Scholar]

19. Martínez-González MA, Sánchez-Tainta A, Corella D, et al. A provegetarian food pattern and reduction in total mortality in the Prevención con Dieta Mediterránea (PREDIMED) study. Am J Clin Nutr. 2014;100:320S–328S. [PubMed] [Google Scholar]

20. Satija A, Bhupathiraju SN, Rimm EB, et al. Plant-based dietary patterns and incidence of type 2 diabetes in US men and women: results from three prospective cohort studies. PLoS Med. 2016;13:e1002039. [PMC free article] [PubMed] [Google Scholar]

21. Neacsu M, Fyfe C, Horgan G, et al. Appetite control and biomarkers of satiety with vegetarian (soy) and meat-based high-protein diets for weight loss in obese men: a randomized crossover trial. Am J Clin Nutr. 2014;100:548–558. [PubMed] [Google Scholar]

22. Dinu M, Abbate R, Gensini GF, et al. Vegetarian, vegan diets and multiple health outcomes: a systematic review with meta-analysis of observational studies. Crit Rev Food Sci Nutr. 2017;57:3640–3649. [PubMed] [Google Scholar]

23. Eichelmann F, Schwingshackl L, Fedirko V, Aleksandrova K. Effect of plant-based diets on obesity-related inflammatory profiles: a systematic review and meta-analysis of intervention trials. Obes Rev. 2016;17(11):1067–1079. [PubMed] [Google Scholar]

24. Orlich MJ, Singh PN, Sabaté J, et al. Vegetarian dietary patterns and mortality in Adventist Health study 2. JAMA Intern Med. 2013;173(13):1230–1238. [PMC free article] [PubMed] [Google Scholar]

25. Cicero AFG, Colletti A. Food and plant bioactives for reducing cardiometabolic disease: how does the evidence stack up? Trends Food Sci Technol. 2017;69:192–202. [Google Scholar]

26. Teodoro AJ. Bioactive compounds of food: their role in the prevention and treatment of diseases. Oxid Med Cell Longev. 2019;2019:3765986. [PMC free article] [PubMed] [Google Scholar]

27. Rangel-Huerta OD, Pastor-Villaescusa B, et al. A systematic review of the efficacy of bioactive compounds in cardiovascular disease: phenolic compounds. Nutrients. 2015;7(7):5177–5216. [PMC free article] [PubMed] [Google Scholar]

28. Speer H, D'Cunha NM, Botek M, et al. The effects of dietary polyphenols on circulating cardiovascular disease biomarkers and iron status: a systematic review. Nutr Metab Insights. 2019;12:1178638819882739. [PMC free article] [PubMed] [Google Scholar]

29. Johnstone AM, Faber P, Gibney E, et al. Effect of an acute fast on energy compensation and feeding behaviour in lean men and women. Int J Obes Relat Metab Disord. 2002;26:1623–1628. [PubMed] [Google Scholar]

30. Multari S, Neacsu M, Scobbie L, et al. Nutritional and phytochemical content of high-protein crops. J Agric Food Chem. 2016;64:7800–7811. [PubMed] [Google Scholar]

31. Calder AG, Garden KE, Anderson SE, et al. Quantitation of blood and plasma amino acids using isotope dilution electron impact gas chromatography/mass spectrometry with U-(13)C amino acids as internal standards. Rapid Commun Mass Spectrom. 1999;13:2080–2083. [PubMed] [Google Scholar]

32. Wilson FA, van den Borne JJ, Calder AG, et al. Tissue methionine cycle activity and hom*ocysteine metabolism in female rats: impact of dietary methionine and folate plus choline. Am J Physiol Endocrinol Metab. 2009;296:E702–E713. [PubMed] [Google Scholar]

33. Patterson BW, Carraro F, Wolfe RR. Measurement of 15N enrichment in multiple amino acids and urea in a single analysis by gas chromatography/mass spectrometry. Biol Mass Spectrom. 1993;22:518–523. [PubMed] [Google Scholar]

34. Johnstone AM, Horgan GW, Murison SD, et al. Effects of a high-protein ketogenic diet on hunger, appetite, and weight loss in obese men feeding ad libitum. Am J Clin Nutr. 2008;87:44–55. [PubMed] [Google Scholar]

35. Neacsu M, McMonagle J, Fletcher RJ, et al. Availability and dose response of phytophenols from a wheat bran-rich cereal product in healthy human volunteers. Mol Nutr Food Res. 2017;61:1–15. [PubMed] [Google Scholar]

36. Russell WR, Gratz SW, Duncan SH, et al. High-protein, reduced-carbohydrate weight-loss diets promote metabolite profiles likely to be detrimental to colonic health. Am J Clin Nutr. 2011;93:1062–1072. [PubMed] [Google Scholar]

37. Russell WR, Duncan SH, Scobbie L, et al. Major phenylpropanoid-derived metabolites in the human gut can arise from microbial fermentation of protein. Mol Nutr Food Res. 2013;57(3):523–535. [PubMed] [Google Scholar]

38. Huang WY, Cai YZ, Zhang YB. Natural phenolic compounds from medicinal herbs and dietary plants: potential use for cancer prevention. Nutr Cancer Int J. 2010;62:1–20. [PubMed] [Google Scholar]

39. Hsu CL, Yen GC. Phenolic compounds: evidence for inhibitory effects against obesity and their underlying molecular signalling mechanisms. Mol Nut Food Res. 2008;52:53–61. [PubMed] [Google Scholar]

40. Gill CIR, Rowland IR. Diet and cancer: assessing the risk. Br J Nutr. 2002;88(1):S73–S87. [PubMed] [Google Scholar]

41. Hwang MK, Bode AM, Byun S, et al. Cocarcinogenic effect of capsaicin involves activation of EGFR signalling but not TRPV1. Cancer Res. 2010;70:6859–6869. [PubMed] [Google Scholar]

42. Chen S, Akter S, Kuwahara K, et al. Serum amino acid profiles and risk of type 2 diabetes among Japanese adults in the Hitachi Health study. Sci Rep. 2019;9:7010. [PMC free article] [PubMed] [Google Scholar]

43. Tobias DK, Clish C, Mora S, et al. Dietary intakes and circulating concentrations of branched-chain amino acids in relation to incident type 2 diabetes risk among high-risk women with a history of gestational diabetes mellitus. Clin Chem. 2018;64(8):1203–1210. [PMC free article] [PubMed] [Google Scholar]

44. William R, Wikoff WR, Andrew T, et al. Metabolomics analysis reveals large effects of gut microflora on mammalian blood metabolites. Proc Natl Acad Sci USA. 2009;106:3698–3703. [PMC free article] [PubMed] [Google Scholar]

45. Young SN, Anderson GM, Gauthier S, et al. The origin of indoleacetic acid and indolepropionic acid in rat and human cerebrospinal fluid. J Neurochem. 1980;34:1087–1092. [PubMed] [Google Scholar]

46. Tuomainen M, Lindström J, Lehtonen M, et al. Associations of serum indolepropionic acid, a gut microbiota metabolite, with type 2 diabetes and low-grade inflammation in high-risk individuals. Nutr Diabetes. 2018;8:35. [PMC free article] [PubMed] [Google Scholar]

47. Abildgaard A, Elfving B, Hokland M, et al. The microbial metabolite indole-3-propionic acid improves glucose metabolism in rats but does not affect behaviour. Arh Physiol Biochem. 2018;124:306–312. [PubMed] [Google Scholar]

48. Heisler LK, Jobst EE, Sutton GM, et al. Serotonin reciprocally regulates melanocortin neurons to modulate food intake. Neuron. 2006;51(2):239–249. [PubMed] [Google Scholar]

49. Cangiano C, Laviano A, Del Ben M, et al. Effects of oral 5-hydroxy-tryptophan on energy intake and macronutrient selection in non-insulin dependent diabetic patients. Int J Obes Relat Metab Disord. 1998;22(7):648–654. [PubMed] [Google Scholar]

50. Russell WR, Drew JE, Scobbie L, et al. Inhibition of cytokine-induced prostanoid biogenesis by phytochemicals in human colonic fibroblasts. Biochim Biophys Acta Mol Basis Dis. 2006;1762:124–130. [PubMed] [Google Scholar]

51. Russell WR, Burkitt MJ, Scobbie, , et al. An EPR investigation into the effects of substrate structure on peroxidase-catalysed phenylpropanoid oxidation. Biomacromol. 2006;7:268–273. [PubMed] [Google Scholar]

52. Russell WR, Scobbie L, Duthie GG, et al. Inhibition of 15-lipoxygenase-catalysed oxygenation of arachidonic acid by substituted benzoic acids. Bioorg Med Chem. 2008;16:4589–4593. [PubMed] [Google Scholar]

53. Goldfine AB, Silver R, Aldhahi W, et al. Use of salsalate to target inflammation in the treatment of insulin resistance and type 2 diabetes. Clin Transl Sci. 2008;1(1):36–43. [PMC free article] [PubMed] [Google Scholar]

54. Yu Z, Yin Y, Zhao W, et al. Anti-diabetic activity peptides from albumin against α-glucosidase and α-amylase. Food Chem. 2012;135(3):2078–2085. [PubMed] [Google Scholar]

55. Alakolanga A, Kumar N, Jayasinghe L, et al. Antioxidant property and α-glucosidase, α-amylase and lipase inhibiting activities of flacourtia inermis fruits: characterization of malic acid as an inhibitor of the enzymes. J Food Sci Technol Mys. 2015;52(12):8383–8388. [PMC free article] [PubMed] [Google Scholar]

56. Koska J, Ortega E, Buntl C, et al. The effect of salsalate on insulin action and glucose tolerance in obesenon-diabetic patients: results of a randomized double-blind placebo-controlled study. Diabetologia. 2009;52:385–393. [PMC free article] [PubMed] [Google Scholar]

57. Rumore MM, Kim KS. Potential role of salicylates in type 2 diabetes. Ann Pharm. 2010;44:1207–1221. [PubMed] [Google Scholar]

58. Clarke R, Bennett DA, Parish S, et al. hom*ocysteine and coronary heart disease: meta-analysis of MTHFR case-control studies, avoiding publication bias. PLoS Med. 2012;9:e1001177. [PMC free article] [PubMed] [Google Scholar]

59. Meigs JB, Jacques PF, Selhub J, et al. Fasting plasma hom*ocysteine levels in the insulin resistance syndrome—the Framingham offspring study. Diabetes Care. 2001;24:1403–1410. [PubMed] [Google Scholar]

60. Tinelli C, Di Pino A, Ficulle E, et al. Hyperhom*ocysteinemia as a risk factor and potential nutraceutical target for certain pathologies. Front Nutr. 2019;6:49. [PMC free article] [PubMed] [Google Scholar]

61. Huo Y, Qin X, Wang J, et al. Efficacy of folic acid supplementation in stroke prevention: new insight from a meta-analysis. Int J Clin Pract. 2012;66:544–551. [PubMed] [Google Scholar]

62. Xie Y, Bowe B, Li T, Xian H, Al-Aly Z. Blood urea nitrogen and risk of insulin use among people with diabetes. Diabetes Vasc Dis Res. 2018;15(5):409–416. [PubMed] [Google Scholar]

63. Nielsen LV, Kristensen MD, Klingenberg L, et al. Protein from meat or vegetable sources in meals matched for fiber content has similar effects on subjective appetite sensations and energy intake—a randomized acute cross-over meal test study. Nutrients. 2018;10:96. [PMC free article] [PubMed] [Google Scholar]

64. Kristensen MD, Bendsen NT, Christensen S, et al. Meals based on vegetable protein sources (beans and peas) are more satiating than meals based on animal protein sources (veal and pork)—a randomized cross-over meal test study. Food Nutr Res. 2016;60:326–334. [PMC free article] [PubMed] [Google Scholar]

66. Sharafi M, Alamdari N, Wilson M, et al. Effect of a high-protein, high-fiber beverage preload on subjective appetite ratings and subsequent ad libitum energy intake in overweight men and women: a randomized, double-blind placebo-controlled, crossover study. Curr Dev Nutr. 2018;2(6):nzy022. [PMC free article] [PubMed] [Google Scholar]