The effects of substituting red and processed meat for mycoprotein on biomarkers of cardiovascular risk in healthy volunteers: an analysis of secondary endpoints from Mycomeat

Observational studies suggest that red and processed meat consumption is associated with an increased risk of cardiovascular disease (CVD) [1‐5]. In contrast, adherence to plant-based dietary patterns appears to confer a cardioprotective effect [6, 7]. Notably, findings from one prospective cohort study suggest a 10% increase in energy intake from animal protein translates into an 8% increased risk of CVD mortality, while a 3% increase in energy from plant protein reduces risk by 12% [8]. These observations are reciprocated by findings from randomised controlled trials that report favourable effects in CVD risk markers when meat consumption is curtailed [9‐12], including lower low-density lipoprotein (LDL) cholesterol and apolipoprotein B levels [13]; as well as reduced urinary excretion of microbially produced trimethylamine N-oxide (TMAO) [14].

Public health approaches have had limited success in inducing meaningful population level transitions to plant-based eating, due to culinary traditions, taste preferences, and social and cultural norms [15‐18]. Meat alternatives—i.e., vegetarian and/or vegan foods designed to mimic meat, yet devoid of animal meat (vegetarian) or any animal derivative (vegan)—offer an effective approach to curtail meat consumption without drastically altering meal patterns and dietary habits [19, 20]. Mycoprotein is a sustainable protein derived from the continuous cultivation of Fusarium venanatum, which is high in protein and fibre, while low in fat [21‐23]. With processing, it can resemble the texture and flavour of meats.

Previous randomised controlled trials have found replacing total meat with mycoprotein based foods elicits reductions in total and LDL cholesterol. These trials report different responses in high density lipoprotein (HDL), and this may be due to differences in the study population demographics and baseline cardiometabolic disease risk status [24‐26]. It is important therefore to better characterise the effect of displacing meat with mycoprotein on CVD risk markers in wider study populations. In addition, the comparative influence on CVD markers of mycoprotein with that of red and processed meat have not been investigated. Given that epidemiological evidence suggests a greater risk of CVD from red and processed meat versus other animal protein sources [27], such investigations would be informative to nutritional science and public health.

Here, we present an analysis of secondary endpoints from the Mycomeat study. Mycomeat was a randomised crossover controlled dietary intervention trial comprising metabolically healthy males with faecal genotoxicity as the primary endpoint; a surrogate marker of colorectal cancer risk by evaluating the DNA damaging potential of foodstuff and dietary patterns within the colon [28]. The participants consumed a variety of red and processed meat products or the weight equivalent of mycoprotein based products over 2 week diet phases. Blood biochemical parameters related to CVD risk were included as a priori defined secondary endpoints.

Here, we report the effects of replacing red and processed meat with the weight equivalent of mycoprotein based foods over two weeks, on markers of CVD risk in metabolically healthy male adults. We note that these observations are of a priori defined secondary endpoints in a study designed to evaluate measures of gut health, thus we caution the potential for type 1 error. Nevertheless, our observation that mycoprotein consumption elicits reductions in total (− 6.74%) and LDL cholesterol (− 12.3%) in metabolically healthy and relatively young British men are of value; these findings reinforce similar observations from a limited number of other relatively small intervention trials in different study populations. Notably, in two parallel feeding studies, using mycoprotein based test foods purposefully designed for the intervention, Turnbull et al. noted that, (1) 191 g daily mycoprotein for 3 weeks reduced both total (− 13%) and LDL cholesterol (− 9%) in 9 subjects with slightly elevated baseline cholesterol status [26]; and (2) that 130 g/day over 8 weeks reduced total (− 8%) and LDL cholesterol (− 13%) in a similar study population (n = 11) [25]. Using commercially available products, Ruxton and McMillan [37] recorded no change in total and LDL cholesterol in 21 British adults consuming, by product preference, the equivalent of 88 g mycoprotein daily, however this was a lightly controlled open label trial in a free-living population, and they did note evidence of an effect with higher levels of compliance, and a reduction in cholesterol in those with higher baseline values. More recently, Coelho et al. [24] reported a reduction in total (− 14.3%) and LDL (− 19.33%) cholesterol in metabolically healthy young adults (n = 20) consuming the equivalent of 181 g wet weight mycoprotein per day for 1 week. The magnitude of change in LDL (− 12.3%) that we observed with mycoprotein exceeds that achievable through consuming 2.4 g/day of plant sterols and stanols, which is deemed clinically meaningful by the European Food Standards Agency [38]. However, probably due to the nature of our healthy study population (with baseline total cholesterol < 5 mmol/l), the intervention did not have a significant effect on a panel of CVD risk scores [39], and we recommend that future studies target higher risk groups.

Overall, there is now consistent evidence across several, albeit small, intervention studies demonstrating that displacement of meat with mycoprotein decreases cholesterol [40, 41]. Whether or not mycoprotein is uniquely cholesterol lowering in the context of a high-fibre low-meat diet is uncertain, for example, SWAP-MEAT was an 8-week randomised controlled trial with no washout in which healthy volunteers consumed ≥ 2 portions per day of a variety of meat or meat substitutes, none of which were mycoprotein based [42]. In that study, volunteers on the meat substitute arm reported consuming an additional ~ 6 g of fibre per day and showed a difference of 0.27 mmol/l LDL between study arms. In Mycomeat, our volunteers consumed a significantly higher amount of total fibre (+ 16.74 g/day) on the mycoprotein arm, leading to a greater net reduction in LDL compared to the meat arm (0.34 mmol/l) than the reduction observed in SWAP-MEAT. A study comparing meat alternatives with different protein constituents, whilst controlling total fibre intake would further the understanding about their unique beneficial properties.

Whilst a definitive intervention study demonstrating that mycoprotein, independently of meat displacement, lowers cholesterol is still needed, there are several mechanisms that might explain the observed effect. (1) In a previous health claim assessment, an EFSA panel concluded that the hypocholesteroleamic properties of mycoprotein were simply a reflection of its β-glucan content [43]. We do note however that the β-glucan (β1-3, β1-6) in mycoprotein has a different bonding arrangement to β-glucan found in oats and barley (β1-3, β1-4). It also has different sugar chain lengths and a different degree of branching which affect its viscosity [44, 45]. The hypocholesterolaemic effects of oat β-glucan is considered a function of its viscosity, and subsequent ability to bind bile and cholesterol. In support of such a mechanism for mycoprotein, Colosimo et al. [46] observed that mycoprotein may both inhibit intestinal lipases and sequester bile in an in vitro model. However, the authors argue that the bile sequester is a function of the fungal hyphae and the food matrix as opposed to its β-glucan content. Notably, we were not able to identify any increase in faecal cholestenone or bile acids in our metabolomic analysis that would suggest increased cholesterol clearance (Supplementary material, Fig. S1). (2) The intestinal microbiota also influences enterohepatic circulation, through its bile and cholesterol hydrolase activity [47]. We have previously described the effects of this intervention on the composition of the microbiome where we did note an increase in the abundance of Lactobacilli which are known to be bile and cholesterol hydrolase capable [47, 48]. Further, in support of this potential explanation, in an exploratory sub-analysis of the microbiome data, we noted differences in baseline microbiome between those we classified as ‘LDL responders’ versus ‘non-responders’ to the mycoprotein diet (Supplementary material, Figs. S2–4). (3) Microbially produced short chain fatty acids (SCFA) influence endogenous cholesterol synthesis [49]. The principal SCFA, acetate, is a substrate for both lipid and cholesterol synthesis, however, its incorporation into cholesterol may be inhibited by the presence of propionate [50], and the overall balance of acetate relative to propionate has been suggested as a predictor of cholesterol status [51]. Propionate and butyrate are also putative inhibitors of hepatic cholesterol synthesis through disruption of sterol regulatory element-binding protein (SREBP) signalling and down regulation of key enzymes of the β-hydroxy β-methylglutaryl-CoA (HMG-CoA) pathway [49, 52, 53]. In support of this potential mechanism, we have previously described the effect of this intervention on SCFA excretion and note higher propionate production with mycoprotein. Thus, future studies should seek to explore these potential mechanisms in respect to mycoprotein and cholesterol status.

TMAO is considered a marker of CVD risk [54] and has been selected as the primary endpoint in recent randomised controlled trials, including SWAP-MEAT [14, 42]. We did not however observe statistically meaningful differences in TMAO in this intervention. Future work could consider a dedicated study design with a primary focus on mycoprotein consumption and TMAO production.

Thus far, three acute feeding studies have reported on the effect of consuming mycoprotein on glycaemic response in healthy volunteers. In particular, Turnbull et al. [55] noted that 75 g of mycoprotein suppresses glycaemic response following a carbohydrate challenge relative to a protein matched soya-based control in healthy volunteers. Bottin et al. [56] observed no influence on post-test-meal glucose but a lower insulinemic response to mycoprotein relative to an iso-energetically matched serving of chicken in overweight but otherwise metabolically healthy volunteers. Most recently, Dunlop et al. [57] demonstrated a more controlled insulinemic response to a mass matched serving of liquefied mycoprotein relative to milk. To our knowledge, only one prior study has considered insulin sensitivity and glucose control over a longer duration. In that study by Coelho et al. [24], 2 servings a day of mycoprotein had no effect on glycaemia after a 1 week intervention. These null observations mirror our own in relation to fasted glucose, however we caution that both studies were carried out in metabolically healthy volunteers, with glucose being measured as secondary endpoints. Mechanistically, Colosimo et al. [58] showed that the cell wall fraction of Fusarium venanatum may entrap α-amylase and thereby reduce starch hydrolysis. Supplementary oat β-glucan consistently lowers the glucose and insulin response to high carbohydrate meals [59], and National Diet and Nutrition Survey data suggest that regular consumers of mycoprotein have lower glycated HbA1c [60]. In the context of metabolic health, we also noted, in our 2-week high-dose intervention, a small but statistically significant difference in waist circumference. This may be consistent with a handful of acute feeding studies noting increased satiation and desire to eat relative to control [61, 62]. However longer interventions in higher metabolic risk groups are needed to fully determine any potential influence of mycoprotein in adiposity and metabolic health.

We observed non-statistically significant, but potentially clinically meaningful trends towards reduced systolic and diastolic blood pressure with mycoprotein (− 4.11 mmHg and − 2.71 mmHg relative to the meat diet). The effects of mycoprotein on blood pressure has not been reported previously. However, a reduction in blood pressure may be consistent with a diet high in fibre, alongside the small observed reduction in waist circumference, and greater excretion of potassium and urinary nitrates relative to the meat arm.

Strengths of this work include the use of a randomised controlled crossover study design, whereby the participants acted as their own control. A parallel design would have required much larger numbers and matching of characteristics to produce equivalent statistical robustness. The investigators were also blinded to the intervention, removing any chance of bias from the analysis. The mycoprotein and meat products included were not exclusively produced for the study and are readily available and consumed amongst the population, The intervention was based on a high but not unrealistic intake of meat amongst the target group, healthy young UK based men, thus the findings may be translatable and of particular relevance to those consuming high meat diets.

Limitations of this work include that this was an analysis of a priori defined secondary endpoints in a study designed to assess gut health. Further, our study population was relatively young and comprised only healthy, male adults. Compliance was only assessed using a 1-day food record, this was primarily due to concerns over participant burden and was balanced against the fact that nutritional status was not a primary outcome, a more thorough interrogation of compliance might have helped explain why some volunteers responded better than others to the intervention.

In conclusion, this work is timely in the context of the explosion of interest in meat substitutes and the push towards meat reduction for ecological and health purposes. It demonstrates in a real-world setting that substituting mycoprotein for meat improves biochemical markers of cardiovascular disease risk. Further work is needed to evaluate dose response as part of normal dietary patterns and to elucidate mechanisms behind the observed responses.