
==== Front
Gut Microbiome (Camb)
Gut Microbiome (Camb)
GMB
Gut Microbiome
2632-2897
Cambridge University Press Cambridge, UK

10.1017/gmb.2022.1
S2632289722000019
Research Article
Higher total faecal short-chain fatty acid concentrations correlate with increasing proportions of butyrate and decreasing proportions of branched-chain fatty acids across multiple human studies
LaBouyer Maria 1Formal analysis Methodology
https://orcid.org/0000-0002-1488-6610
Holtrop Grietje 2Data curation Formal analysis Methodology Software Visualization
https://orcid.org/0000-0002-6048-1374
Horgan Graham 2Data curation Formal analysis Methodology Visualization Writing – review & editing
https://orcid.org/0000-0002-7367-4511
Gratz Silvia W. 1Conceptualization Funding acquisition Investigation Project administration Writing – review & editing
https://orcid.org/0000-0002-9186-3463
Belenguer Alvaro 3Formal analysis Investigation Methodology Writing – review & editing
https://orcid.org/0000-0002-5137-1384
Smith Nicola 1Formal analysis Investigation Methodology Visualization Writing – review & editing
https://orcid.org/0000-0001-5099-8495
Walker Alan W. 1Funding acquisition Project administration Supervision Writing – review & editing
https://orcid.org/0000-0002-4903-0978
Duncan Sylvia H. 1Conceptualization Funding acquisition Investigation Methodology Supervision Writing – original draft Writing – review & editing
https://orcid.org/0000-0002-5484-292X
Johnstone Alexandra M. 1Conceptualization Funding acquisition Methodology Project administration Supervision Writing – review & editing
https://orcid.org/0000-0003-2115-2399
Louis Petra 1Conceptualization Funding acquisition Project administration Supervision Writing – review & editing
https://orcid.org/0000-0003-2116-0188
Flint Harry J. 1Conceptualization Funding acquisition Project administration Supervision Writing – original draft Writing – review & editing
https://orcid.org/0000-0002-4608-0013
Scott Karen P. 1*Conceptualization Funding acquisition Methodology Project administration Supervision Writing – original draft Writing – review & editing
1 Gut Health Group, Rowett Institute, University of Aberdeen, Foresterhill, Aberdeen, Scotland, UK
2 Biomathematics and Statistics Scotland (BioSS), Aberdeen, Scotland, UK
3 Nutrigenomics and Fish Growth Endocrinology Group, Institute of Aquaculture of Torre de la Sal, (IATS - CSIC), Castellon, Spain
* Corresponding author. Email: k.scott@abdn.ac.uk
Harry J. Flint and Karen P. Scott share joint last author status.

2022
30 3 2022
3 e213 10 2021
25 2 2022
18 3 2022
© The Author(s) 2022
2022
The Author(s)
https://creativecommons.org/licenses/by/4.0/ This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.

Metabolites produced by microbial fermentation in the human intestine, especially short-chain fatty acids (SCFAs), are known to play important roles in colonic and systemic health. Our aim here was to advance our understanding of how and why their concentrations and proportions vary between individuals. We have analysed faecal concentrations of microbial fermentation acids from 10 human volunteer studies, involving 163 subjects, conducted at the Rowett Institute, Aberdeen, UK over a 7-year period. In baseline samples, the % butyrate was significantly higher, whilst % iso-butyrate and % iso-valerate were significantly lower, with increasing total SCFA concentration. The decreasing proportions of iso-butyrate and iso-valerate, derived from amino acid fermentation, suggest that fibre intake was mainly responsible for increased SCFA concentrations. We propose that the increase in % butyrate among faecal SCFA is largely driven by a decrease in colonic pH resulting from higher SCFA concentrations. Consistent with this, both total SCFA and % butyrate increased significantly with decreasing pH across five studies for which faecal pH measurements were available. Colonic pH influences butyrate production through altering the stoichiometry of butyrate formation by butyrate-producing species, resulting in increased acetate uptake and butyrate formation, and facilitating increased relative abundance of butyrate-producing species (notably Roseburia and Eubacterium rectale).

Keywords

Short chain fatty acids
butyrate
branched chain fatty acids
human gut microbiota
faecal pH
Rural and Environment Science and Analytical Services Division http://dx.doi.org/10.13039/100011310 World Cancer Research Fund http://dx.doi.org/10.13039/501100000321
==== Body
pmcIntroduction

Short-chain fatty acids (SCFA) are the major products of microbial fermentation of non-digested dietary substrates that reach the human large intestine. They are the main metabolites of microbial fermentation of dietary carbohydrates and fibre, while branched-chain fatty acids (BCFA) are additional products of protein metabolism (Smith and Macfarlane, 1998; Scott et al., 2013; Yao et al., 2016). It has long been recognised that these acids have major impacts on the gut environment, on mucosal absorption and host physiology (Cummings et al., 1978). In addition to supplying energy from dietary fibre via SCFA absorption, these acids have multiple effects on host gene expression and cellular development as inhibitors of histone de-acetylation and through signaling via G-protein coupled receptors (see recent review by Blaak et al., 2020). The effects have been shown to include anti-inflammatory action via the maturation of regulatory T-cells, and the production of hormones that influence satiety (Morrison and Preston, 2016).

The three predominant SCFA products (acetate, propionate and butyrate) differ in their distribution through the host, modes of action and consequences for health (Chambers et al., 2018). Butyrate is the preferred energy source for colonic epithelial cells and has been particularly associated with the maintenance of gut health because of its role in the prevention of colitis and colorectal cancer (Pryde et al., 2002; Hamer et al., 2008; Louis et al., 2014). It is therefore important to determine to what extent, and why, the relative production of these major SCFA in the gut varies within human populations. Since most SCFA produced are rapidly absorbed in the colon, faecal concentrations represent a balance between production and absorption and represent approximately 5–10 percent of the total production (Boets et al., 2017). The higher concentrations in the proximal compared to the distal colon (Cummings et al., 1987) reflect both greater bacterial fermentation (production) at that site and less time for absorption. Transit rate through the colon also affects the amount of absorption and correlates with faecal SCFA concentrations (Lewis and Heaton, 1997; Müller et al., 2020). Nevertheless, many studies have demonstrated diet-associated changes in faecal SCFA concentrations, indicating that they can be a proxy for monitoring overall changes in the balance between production and absorption (Duncan et al., 2007; Boets et al., 2015).

Although some of the earliest literature represents carbohydrate fermentation by anaerobic gut communities as a single balanced equation (Wolin, 1960), it is now clear that fermentation stoichiometry can vary with the species composition of the gut microbiota, with the types of substrate fermented, and with the general gut environment (Flint and Juge, 2015; Louis and Flint, 2017; Reichardt et al., 2018). Thus many factors can potentially influence the relative production rates of acetate, propionate and butyrate in vivo. Added to this is the influence of metabolite cross-feeding, as many butyrate-producing bacteria in the colon are also net consumers of acetate (Duncan et al., 2002; Louis and Flint, 2017). A stable isotope study illustrating the systemic availability of colonically administered labelled SCFA showed that 24 per cent of acetate was converted to butyrate (Boets et al., 2017).

We report here an analysis of multiple studies conducted with human volunteers in Aberdeen, UK, in which data are available for SCFA concentrations in human stool samples. This is a valuable dataset with comparable information for a large number of volunteers that can provide significant insights into the function of the human gut microbiota. The results reveal some highly significant relationships, in particular between % butyrate and total SCFA concentrations that build on classic dietary intervention studies involving human volunteers (Cummings et al., 1978; Stephen et al., 1987; Lewis and Heaton, 1997).

Methods

Human studies

All the human studies included (Table 1) recruited volunteers locally who did not take antibiotics in the 3 months preceding study participation, nor during the studies, and were conducted according to guidelines laid down in the Declaration of Helsinki, with approval from the appropriate local ethical committee (North of Scotland Research Ethics committee or internal Rowett Institute Human Studies committee). Written consent was obtained from all participants. Samples within studies were collected at the timepoints stated in the respective study protocol. Here, we mainly compare the baseline samples from the volunteers entering the different studies, which represent the free-living population. Any volunteers who participated in more than one study were only considered once by excluding them from datasets in later studies. Details of the dietary regimes followed are detailed in publications relating to individual studies, and are described in the text when appropriate (see Table 1).Table 1. Volunteer studies included in this analysis.

Study ID	n a	Females	Males	Fully controlled	Habitual diet data	Bacterial quantification method	pH y/n	SCFA y/n	Energy Intake y/n	BMI range	Reference	
778	18	0	18	Yes	No	FISH	No	Yes	Yes	29.6–41.5	Duncan et al. (2007)	
779	17	0	17	Yes	No	FISH, qPCR	Yes	Yes	Yes	27.9–48.5	Russell et al. (2011) and Duncan et al. (2008)	
780	14	0	14	Yes	No	qPCR	No	Yes	Yes	27.9–51.3	Walker et al. (2011), Holtrop et al. (2012) and Salonen et al. (2014)	
782	18	0	18	Yes	No	No	Yes	Yes	Yes	27.8–52.6	Lobley et al. (2015) and Gratz et al. (2019)	
783	20	0	20	Yes	No	qPCR	No	Yes	Yes	26.5–43	Neascu et al. (2014)	
Plantain	17	12	5	No	Yes	FISH	No	Yes	No	18.3–35.8	Scott et al. (this study)	
Inulin	12	9	3	No	Yes	qPCR	Yes	Yes	No	19.8–31.5	Fuller et al. (2007) and Ramirez-Farias et al. (2009)	
FruitVeg	38	24	14	No	Yes	qPCR	No	Yes	Yes	17.8–36.6	Duthie et al. (2018) and Louis et al. (this study)	
Oatibix	5	4	1	No	Yes	FISH	Yes	Yes	No	17.6–26.3	Scott et al. (this study)	
Timebugs	4	2	2	No	Yes	FISH	Yes	Yes	Yes	20.9–26.3	Duncan et al. (this study)	
Total n =	163	51	112									
a n = Number of volunteers on each study, followed by number of Females/Males.

Where recorded, energy intake was the average sum of energy (kcal) from carbohydrate, protein, fat and fiber per day based on standardised baseline diets (studies 778 – 783) or by use of a 4-day self-reported food diary (FruitVeg) or weekly averages of daily intake records for Timebugs study.

Abbreviations: BMI, body mass index; FISH, fluorescent in situ hybridisation; qPCR, quantitative polymerase chain reaction.

Sample processing

Faecal samples were collected and the fresh samples processed within 12 h of collection following standard protocols in place at the Rowett Institute (Duncan et al., 2007; Walker et al., 2011). The number of samples collected varied in the different studies, and the number used in each comparison is indicated at the appropriate point in the results section. Specific methods used for sample processing and analysis are shown in Table S1, illustrating the consistency across all studies.

SCFA analysis

SCFA analysis was carried out using gas chromatography using derivatised samples as previously described in Richardson et al. (1989). In brief, following derivatisation of the samples using N-tert-butyldimethylsilyl-N-methyltrifluoroacetamide with 2-ethyl butyrate as the internal standard, the samples were analysed using a Hewlett Packard gas chromatograph (GC) fitted with a silica capillary column and using helium as the carrier gas.

Microbiota analysis

Fluorescent in situ hybridisation (FISH) and quantitative polymerase chain reaction (qPCR) analyses were carried out as detailed in Walker et al. (2005) and Ramirez-Farias et al. (2009), respectively. Any variations on standard protocols are detailed in publications relating to specific individual studies.

Statistical analysis

Associations between individual SCFA ratios and total SCFA were examined via Pearson’s correlation. All statistical tests were performed using R 3.6.0 (R Foundation for Statistical Computing, Vienna), with significance limits set at p < 0.05. Significance of correlations based on multiple observations per volunteer was assessed using a linear mixed model with volunteer as a random effect. Figures were prepared using R version 3.6.3 (R foundation for Statistical Computing, Vienna).

Results

Data on the concentrations of SCFA in faecal samples were compiled from 10 studies conducted in Aberdeen, UK, between 2006 and 2012. Five of these studies involved obese or overweight male volunteers who were enrolled onto carefully controlled dietary intervention trials looking at the impact of diet upon weight loss. The remaining five studies involved healthy volunteers from the general population who received different supplements to their habitual diets (four studies), while Timebugs was a longitudinal study with no dietary intervention (see Table 1 and previous publications, references provided in Table 1, for details).

We first examined data from baseline samples provided by each volunteer before any major dietary intervention. These corresponded to the subjects’ habitual diets in the five ‘normal weight’ studies, or following a balanced weight maintenance diet for 1–3 days in the case of the five obesity studies. Results from these single samples from 158 different volunteers are shown in Figure 1, in which the percentage of each fermentation acid within the total SCFA pool was plotted against absolute total faecal SCFA concentrations. Three relationships were highly significant (p < 0.001). The % butyrate increased with increasing total SCFA concentration (correlation r = 0.627, p < 0.001) while % iso-butyrate and % iso-valerate decreased with increasing total SCFA concentration (r = −0.492 and r = −0.373 respectively, p < 0.001). There was also a significant negative correlation between the % acetate and total SCFA concentration (r = −0.247, p = 0.002). The same relationships were also observed when all datapoints (n = 502) across the time courses of the various dietary interventions were included (Figure S1), again with the strongest positive correlation between % butyrate and total SCFA (r = 0.597, p < 0.001). Across the entire dataset, as the total SCFA concentration increased, percentages of acetate, valerate, iso-valerate and iso-butyrate all decreased significantly (p < 0.001; Figure S1).Figure 1 Relationship between total faecal short-chain fatty acid (SCFA) concentrations and proportional abundance (%) of individual fermentation acids for single baseline samples from 158 of the 163 volunteers. Excludes five volunteers with no samples for SCFA analysis. The sum of all SCFA concentrations (Total SCFA, mM) is on the x-axis while the percentage of each respective SCFA (individual mM divided by total SCFA concentration in mM) is on the y-axis. Pearson’s correlation and linear mixed model analysis gave the following values. Acetate: r = −0.247, p = 0.002; Propionate: r = 0.042, p = 0.602; Iso-butyrate: r = −0.492, p < 0.001; Butyrate: r = 0.627, p < 0.001; Valerate: r = −0.126, p = 0.116; Iso-valerate: r = −0.373, p < 0.001.

Faecal pH measurements were available for samples from five of the studies listed in Table 1. Figure 2 shows the relationship between SCFA concentrations and pH in the baseline samples (n = 51). This reveals a significant inverse correlation between total SCFA concentration and faecal pH (r = −0.425, p =0.002, and specifically between high % butyrate and low faecal pH (r = −0.588, p < 0.001). In contrast, there was a significant positive correlation between % iso-butyrate (r = 0.415, p = 0.002) and % iso-valerate (r = 0.377, p = 0.006) and faecal pH. There was no significant correlation between the % acetate, or % propionate and faecal pH (Figure 2). The strong correlations between decreasing total SCFA concentrations and decreasing % butyrate with increasing pH were also observed when data for all available timepoints (n = 198; including dietary intervention periods) were plotted against pH (Figure S2, r = −0.503, p < 0.001; r = −0.602, p < 0.001, respectively). Across the full dataset, weak positive correlations were observed between pH and the % propionate (r = 0.142, p = 0.049) and % acetate (r = 0.163, p = 0.022), while the strong positive correlation for % iso-butyrate (r = 0.245, p < 0.001) and pH was still present.Figure 2 Correlation between faecal pH, short-chain fatty acid (SCFA) and SCFA proportions in baseline samples collected from five volunteer studies (n = 51). The pH is on the x-axis while total SCFA concentration (mM) or the individual SCFA percentage is on the y-axis. Pearson’s correlation and linear mixed model analysis gave the following values. Total SCFA: r = −0.425, p = 0.002; Propionate: r = 0.058, p = 0.686; Iso-butyrate: r = 0.415, p = 0.002; Butyrate: r = −0.588, p < 0.001; Acetate: r = 0.198, p = 0.164; Iso-valerate: r = 0.377, p = 0.006.

Populations of butyrate-producing bacteria related to Roseburia spp. (including Eubacterium rectale) were previously shown to decrease in parallel with % butyrate in faecal samples in dietary interventions (Duncan et al., 2007). Here, data from all timepoints for four studies in which pH and microbial composition was measured (n = 130; Table 1) showed a significant inverse relationship (r = −0.239, p = 0.017) between pH and the relative abundance of the Roseburia group within the faecal microbiota (Figure 3). No such relationship was found between pH and Faecalibacterium prausnitzii, which is another commonly abundant butyrate producing taxon in the human gut, after combining data from three studies (n = 109; Figure 3).Figure 3 Relationship between faecal pH and proportional abundance (%) of butyrate-producing bacterial groups (Roseburia/E. rectale and F. prausnitzii) within faecal microbiota. Data are based on FISH microscopy counts (Oatibix, Timebugs and 779 study) or quantitative polymerase chain reaction (qPCR) analysis (Inulin study) of all samples from volunteers in four studies (n = 130) in the case of Roseburia, and three studies (n = 109) in the case of F. prausntizii (F. prausnitzii data were not obtained from the Oatibix study). Pearson’s correlation and linear mixed model analysis gave values of: Roseburia % r = −0.239, p = 0.017; F. prausnitzii % r = −0.010, p = 0.829.

In order to specifically investigate links between dietary carbohydrate consumption and butyrate-producing bacteria, for one of the longitudinal studies (study 779), we followed changes in the population of the Roseburia group (as measured by qPCR) over time following the switch from maintenance to a low carbohydrate diet in eight volunteers. These data demonstrate a rapid response to the dietary intervention, with numbers of the Roseburia group declining markedly (p = 0.008) from 5 days of the shift to the low carbohydrate diet in all volunteers (Figure 4).Figure 4 Change in the proportion of Roseburia/E. rectale 16S rRNA gene amplicons in faecal samples from overweight volunteers following switch from a maintenance diet to a decreased carbohydrate, weight loss diet. Estimates are from quantitative polymerase chain reaction (qPCR) data from study 779, analysed as described in Hamer et al. (2008) and Ramirez-Farias et al. (2009). The number of days after the diet shift is on the x-axis while the % of Roseburia/E. rectale group 16S rRNA gene calculated as a % of total bacterial 16S rRNA genes is on the y-axis. The proportional representation of the Roseburia group declined significantly (p = 0.008) from five days after the diet shift (non-parametric test of means before and after day 5).

Across volunteers from all 10 studies, total faecal SCFA concentrations in baseline samples increased significantly with body mass index (BMI) (p < 0.001, Figure 5). We should note that the 5 studies, 778–783, involved only male volunteers who were overweight or obese, while the other studies involved volunteers of both sexes, of mixed BMI (n = 145; Table 1). Increased faecal SCFA have been reported previously in obese subjects of both sexes (Schwiertz et al., 2010). Plots of individual SCFAs against BMI showed that concentrations of each of the three major SCFAs (butyrate, acetate and propionate) also all increased significantly with increasing BMI (p < 0.001). These significant increases in SCFA concentrations converted into significant positive correlations between increased proportional abundance of butyrate and propionate (p = 0.001 and p = 0.023 respectively, Figure 5) with increased BMI. Although the mean concentrations of butyrate and total SCFA were significantly lower in females than males (p < 0.001; Figure S3A), this appears to reflect the higher mean body mass and energy intakes of male compared to female volunteers involved in these studies (Figure S3B). Total food intake and intake of dietary fibre were highly correlated in these studies (data not shown), making it impossible to distinguish separate relationships between fibre intake and SCFAs.Figure 5 Relationship between total faecal short-chain fatty acid (SCFA) or concentrations of individual SCFA in baseline samples versus body mass index in nine studies where body mass index (BMI) was specifically recorded (n = 145). Pearson’s correlation and linear mixed model analysis gave the following values. Total SCFA: r = 0.369, p < 0.001; Butyrate: r = 0.308, p < 0.001; % Butyrate: r = 0.266, p = 0.001; Propionate: r = 0.393, p < 0.001; % Propionate: r = 0.188, p = 0.023; Acetate: r = 0.369, p < 0.001; % Acetate: r = −0.112, p = 0.179.

Discussion

We reported previously that % butyrate decreased and % iso-valerate and iso-butyrate increased in faecal SCFA in response to reduced carbohydrate intake in small groups of overweight volunteers (Duncan et al., 2007; Russell et al., 2011). The wider analysis presented here of samples donated by 163 individuals from 10 human volunteer studies reveals significant changes in the ratios of different SCFA with increasing total faecal SCFA concentration. Specifically, as the total SCFA concentration increases, the proportion (%) of butyrate increases and % iso-valerate and % iso-butyrate decreases. These relationships were significant both across the baseline samples, and when complete longitudinal data from the full dietary intervention periods were included. The simplest interpretation is that inter-individual variation in the volunteers’ habitual dietary intake results in variation in the delivery of fermentable fibre to the large intestine, which is reflected in the bacterial fermentation products detected in faecal samples.

There are two distinct mechanisms involving colonic pH that could explain the increase in % butyrate with increasing total faecal SCFA concentration. The first is that the stoichiometry of butyrate formation by dominant butyrate-producing bacteria, including Roseburia spp. and Faecalibacterium prausnitzii, changes with gut luminal pH. These species rely on the butyryl-CoA:acetate CoA-transferase reaction for the final step in butyrate formation that can involve the net uptake of acetate (Duncan et al., 2002). More butyrate is formed, and more acetate taken up, per mol of hexose fermented by pure cultures when the pH is slightly acidic than when it is closer to neutrality (Kettle et al., 2015; Louis and Flint, 2017). This effect appears to account for a 25–50 per cent higher proportion of butyrate among the SCFA products of in vitro batch incubations with mixed faecal bacteria observed when the initial pH was 5.5 compared to 6.5 (Reichardt et al., 2018). This effect occurred across a wide range of microbial community composition and was not observed for propionate formation in the same incubations (Reichardt et al., 2018). Since there is evidence that higher SCFA concentrations result in lower pH values in the colon in vivo (Bown et al., 1974; Cummings et al., 1978) this stoichiometric shift would increase the % butyrate produced as total SCFA concentrations increase.

pH may also play a crucial role in determining the competition for carbohydrates in the colon. Chemostat studies with human colonic microbiota have shown that major butyrate-producing species compete better for soluble carbohydrates when the pH is slightly acidic than when it is close to neutrality. This is explained by decreased competition between these butyrate-producers and Bacteroides species that are more sensitive to acidic pH (Walker et al., 2005; Duncan et al., 2009; Chung et al., 2016). Theoretical modelling was used to demonstrate that the experimental changes in community composition and butyrate formation were consistent and predictable from the behaviour of cultured human colonic bacteria (Duncan et al., 2009; Kettle et al., 2015).

Another possible mechanism is that an increased supply of dietary fibre to the proximal colon promotes the growth and activity of butyrate-producing species, thereby increasing their representation within the microbial community. Evidence for this in vivo comes from carefully controlled dietary intervention studies that show a significantly higher proportional representation (and absolute numbers) of butyrate-producing bacteria related to Roseburia (including Eubacterium rectale) in faecal microbiota from individuals when consuming diets high in fibre compared with diets low in total carbohydrate and fibre (Duncan et al., 2007; Russell et al., 2011). Indeed, a fourfold decrease in the Roseburia population was associated with fourfold lower % butyrate (and butyrate concentration) among faecal SCFAs (Duncan et al., 2007). The link between high fibre consumption and higher abundance of the Roseburia group has also been reported in other studies (Adamberg et al., 2020), and there is evidence that some of these bacteria specialise in the utilisation of insoluble fibres (Duncan et al., 2016). Fermentable carbohydrates also tend to favour lactate-producing Bifidobacterium species. Since several species of colonic bacteria are known to produce butyrate from lactate (Louis and Flint, 2017) this could stimulate butyrate formation indirectly via metabolite cross-feeding.

Iso-butyrate and iso-valerate are products of the fermentation of branched chain amino acids, mainly by Bacteroides and Clostridia species (Aguirre et al., 2016; Rios-Covian et al., 2020). The % iso-butyrate and iso-valerate among total SCFA is reported to increase with high protein, low carbohydrate diets (Russell et al., 2011; Gratz et al., 2019), while BCFA levels are inversely correlated with consumption of dietary fibre (Rios-Covian et al., 2020). The decrease in % BCFA with increasing total SCFA therefore suggests that high faecal SCFA concentrations are largely attributable to increased carbohydrate fermentation that does not yield BCFA. If we assume that dietary protein is largely digestible (Cummings and Macfarlane, 1997; Van der Wielen et al., 2017), then the delivery of ‘resistant’ dietary protein to the colon is not likely to increase greatly, if at all, on high-fibre diets, while endogenous sources of protein should be largely independent of fibre intake. Therefore, increased fibre intake is expected to lead to fermentation of more non-digestible carbohydrate relative to protein in the large intestine.

Another potential contributor to the observed correlations between higher % butyrate, higher total faecal SCFAs and lower faecal pH is gut transit time. Several studies have shown that gut transit rate can play a major role in determining metabolite concentrations and microbiota profiles. In early studies that used drug or fibre intake to vary gut transit, generally in small groups of volunteers, more rapid whole gut transit was associated with increased faecal SCFA concentrations, increased % butyrate and decreased colonic pH (Cummings et al., 1978; Stephen et al., 1987; El Oufir et al., 1996; Lewis and Heaton, 1997; Abell et al., 2006). More recently, longer gut transit times correlated with higher microbial diversity and species richness (Roager et al., 2016; Vandeputte et al., 2016). Constipated patients across different age-groups appear to have lower total SCFA concentrations, much lower concentrations of butyrate and lower numbers of butyrate-producing bacteria than age-matched healthy controls (Zhuang et al., 2019).

Increased gut transit rate is considered to lead to increased faecal SCFA concentrations for two main reasons (Stephen et al., 1987; Flint, 2011). First, there is evidence that an increase in transit rate through the upper gut may result in less complete digestion of food components, thus increasing the amount of digestive residue arriving in the large intestine and increasing the rate of microbial fermentation (Holgate and Read, 1983; Chapman et al., 1985). Second, more rapid transit is expected to decrease the proportion of SCFA produced by microbial fermentation that is subsequently absorbed across the colonic mucosa. Individuals with slow colonic transit tend to show lower faecal concentrations of SCFA (Müller et al., 2020), and longer average colonic transit times are reported in healthy females than in healthy males (Rao et al., 2009; Wang et al., 2015). The present study was not designed to examine the influence of gender, but we can note that gender-associated differences in the microbiota composition have been linked to BMI (Haro et al., 2016).

In conclusion, inter-individual variation in habitual dietary intakes, in particular fibre intake, are likely to be the main factors accounting for the systematic variation in microbial fermentation and SCFA concentrations observed among human volunteers. The stimulation of butyrate, the main energy source for colonocytes, by decreased colonic pH associated with high microbial SCFA production, noted here, is likely to be a significant factor in the protective effect of fibre consumption against colorectal cancer and certain other bowel diseases.

Acknowledgements

We would like to thank all the volunteers who participated in the 10 human studies included here. We would also like to thank the Rowett Institute Human Nutrition Unit, and the many other staff and students who were involved in each of these studies in recruitment, sample collection and analysis.

Supplementary material

For supplementary material accompanying this paper visit https://doi.org/10.1017/gmb.2022.1.

click here to view supplementary material

Funding

The Rowett Institute (University of Aberdeen) receives financial support from the Scottish Government Rural and Environmental Sciences and Analytical Services (RESAS). Studies 779 and 780 were supported by a grant from the World Cancer Research Fund.

Disclosure statement

The authors declare no conflicts of interests.

Notes on contributors

Maria LaBouyer , Master of Science University of Aberdeen, Scotland. Main Research interests: Gut microbial metabolism.

Grietje Holtrop, PhD 1998, Biomathematics and Statistics Scotland (BioSS), Aberdeen, Scotland. Main research Interests: Mathematical modelling of gut bacteria.

Graham Horgan, PhD 1990. Biomathematics and Statistics Scotland (BioSS), Aberdeen, Scotland. Main research Interests: data integration, linear programming and high dimensional data.

Silvia W. Gratz, PhD 2007, Senior Research Fellow, Rowett Institute, University of Aberdeen. Main Research interests: food toxicology, gut microbial metabolism, intestinal function and toxicity.

Alvaro Belenguer, PhD 2004, Tenured Scientist, Institute of Aquaculture Torre de la Sal (IATS - CSIC). Main Research interests: Gut microbial metabolism, intestinal health and feed efficiency, fish nutrition, genomics and epigenomics.

Nicola Smith, PhD 2021, Rowett Institute, University of Aberdeen. Currently trainee at European Food Safety Authority, Parma, Italy. Main Research interests: gut health, microbiome and food safety.

Alan W. Walker, PhD 2006, Senior Lecturer, Rowett Institute, University of Aberdeen. Main Research interests: Anaerobic gut microbiology, impact of diet on microbial composition, colonisation resistance.

Sylvia H. Duncan, PhD 2001, Senior Research Fellow, Rowett Institute, University of Aberdeen. Main Research interests: human intestinal microbiology, impact of dietary fibre on health, microbial metabolism and cross feeding.

Alexandra M. Johnstone, PhD 2001, Research Professor, Rowett Institute, University of Aberdeen. Main Research interests: Obesity, appetite and energy balance across the lifecourse.

Petra Louis, PhD 1996, Senior Research Fellow, Rowett Institute, University of Aberdeen. Main Research interests: human gut microbiota, microbial ecology, microbial metabolism.

Harry J. Flint, PhD 1977, Emeritus Professor, Rowett Institute, University of Aberdeen. Main Research interests: Anaerobic microbiology, microbial fermentation of carbohydrates, gut health and nutrition.

Karen P. Scott, PhD 1993, Senior Research Fellow, Rowett Institute, University of Aberdeen. Main Research interests: Anaerobic gut microbiology, impact of diet on microbial composition and health, antimicrobial resistance, bacterial gene transfer.

Author Contributions

Conceptualisation: S.H.D., S.W.G., H.J.F., A.M.J., P.L., K.P.S., and A.W.W.

Methodology: A.B., M.L.B., S.H.D., P.L., K.P.S., N.S. and A.W.W.

Formal analysis: A.B., M.L.B., S.H.D., H.J.F., S.W.G., G.Hol., G.Hor., P.L., A.M.J., K.P.S. and N.S.

Data curation: G.Hol. and G. Hor.

Supervision: H.J.F. and K.P.S.

Funding acquisition: S.H.D., S.W.G., H.J.F., A.M.J., P.L., K.P.S. and A.W.W.

Writing – original draft: K.P.S. and H.J.F.

Writing – review & editing: all authors.

Research transparency and reproducibility

The data used to prepare this manuscript were generated from 10 previous studies, many published as primary papers. The combined dataset is available in the Supplementary file entitled ‘faecal SCFA and butyrate study data tables’.
==== Refs
References

Abell GCJ, Conlon MA and McOrist AL (2006) Methanogenic archaea in adult human faecal samples are inversely related to butyrate concentration. Microbial Ecology in Health and Disease 18 , 154–160.
Adamberg K, Jaagura M, Aaspõllu A, Nurk E and Adamberg S. (2020) The composition of faecal microbiota is related to the amount and variety of dietary fibres. International Journal of Food Sciences and Nutrition 71 , 845–855.32083496
Aguirre M, Eck A, Koenen ME, Marjorie E, Savelkoul PHM, Budding AE and Venema K (2016) Diet drives quick changes in the metabolic activity and composition of human gut microbiota in a validated in vitro gut model. Research in Microbiology 167 , 114–125.26499094
Blaak EE, Canfora EE, Theis S, Frost G, Groen AK, Mithieux G, Nauta A, Scott KP, Stahl B, van Harsselaar J, van Tol R, Vaughan EE and Verbeke K (2020) Short chain fatty acids in human gut and metabolic health. Beneficial Microbes 11 , 411–455.32865024
Boets E, Deroover L, Houben E, Vermeulen K, Gomand SV, Delcour JA and Verbeke K (2015) Quantification of in vivo colonic short chain fatty acid production from inulin. Nutrients 7 , 8916–8929.26516911
Boets E, Gomand SV, Deroover L, Preston T, Vermeulen K, de Preter V, Hamer HM, Van den Mooter G, de Yuyst L, Courtin CM, Annaert P, Delcour JA and Verbeke KA (2017) Systemic availability and metabolism of colonic-derived short-chain fatty acids in healthy subjects: a stable isotope study. The Journal of Physiology 595 , 541–555.27510655
Bown RL, Gibson JA, Sladen GE, Hicks B and Dawson AM (1974) Effects of lactulose and other laxatives on ileal and colonic pH as measured by a radiometry device. Gut 15 , 999–1004.4448417
Chambers ES, Preston T, Frost G and Morrison, DJ (2018) Role of gut microbiota-generated short-chain fatty acids in metabolic and cardiovascular health. Current Nutrition Reports 7 , 198–206.30264354
Chapman RW, Sillery JK, Graham MM and Saunders DR (1985) Absorption of starch by healthy ileostomates: effect of transit time and of carbohydrate load. The American Journal of Clinical Nutrition 41 , 1244–1249.4003331
Chung WCF, Walker AW, Louis P, Parkhill J, Vermeiren J, Bosscher D, Duncan SH and Flint HJ (2016) Modulation of the human gut microbiota by dietary fibres occurs at the species level. BMC Biology 14 , 3.26754945
Cummings JH and Macfarlane GT (1997) Role of intestinal bacteria in nutrient metabolism. Clinical Nutrition 16 , 3–11.16844615
Cummings JH, Pomare EW, Branch HWJ, Naylor CPE and MacFarlane, GT (1987) Short chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 28 , 1221–1227.3678950
Cummings JH, Southgate DAT, Branch W, Houston H, Jenkins DJA and James WPT (1978) The colonic response to dietary fibre from carrot, cabbage, apple, bran and guar gum. Lancet 1 , 5–9.74533
Duncan SH, Barcenilla A, Stewart CS, Pryde SE and Flint, HJ (2002) Acetate utilization and butyryl CoA:acetate CoA transferase in human colonic bacteria. Applied and Environmental Microbiology 68 , 5186–5190.12324374
Duncan SH, Belenguer A, Holtrop G, Johnstone AM, Flint HJ and Lobley GE (2007) Reduced dietary intake of carbohydrates by obese subjects results in decreased concentrations of butyrate and butyrate-producing bacteria in feces. Applied and Environmental Microbiology 73 , 1073–1078.17189447
Duncan SH, Lobley GE, Holtrop G, Ince J, Johnstone AM, Louis P & Flint HJ (2008). Human colonic microbiota associated with diet, obesity, and weight loss. International Journal of Obesity 32 , 1720–1724.18779823
Duncan SH, Louis P, Thomson JM and Flint HJ (2009) The role of pH in determining the species composition of the human colonic microbiota. Environmental Microbiology 11 , 2112–2122.19397676
Duncan SH, Russell WR, Quartieri A, Rossi M, Parkhill J, Walker AW and Flint HJ (2016). Wheat bran promotes enrichment within the human colonic microbiota of butyrate-producing bacteria that release ferulic acid. Environmental Microbiology 18 , 2214–2225.26636660
Duthie SJ, Duthie GG, Russell WR, Kyle JA, Macdiarmid JI, Rungapamestry V, Stephen S, Megias-Baeza C, Kaniewska JJ, Shaw L, Milne L, Bremner D, Ross K, Morrice P, Pirie LP, Horgan G and Bestwick CS (2018) Effect of increasing fruit and vegetable intake by dietary intervention on nutritional biomarkers and attitudes to dietary change: a randomised trial. European Journal of Nutrition 57 , 1855–1872.28560503
El Oufir L, Flourié B, Bruley des Varannes S, Barry JL, Cloarec D, Bornet F and Galmiche JP (1996) Relations between transit time, fermentation products, and hydrogen consuming flora in healthy humans. Gut 38 , 870–877.8984026
Flint HJ (2011) Obesity and the gut microbiota. Journal of Clinical Gastroenterology 45 , S128–S132.21992951
Flint HJ and Juge N (2015) Role of microbes in carbohydrate digestion. Food Science and Technology 29 , 24–26.
Fuller Z, Louis P, Mihajlovski A, Rungapamestry V, Ratcliffe B and Duncan AJ (2007) Influence of cabbage processing methods and prebiotic manipulation of colonic microflora on glucosinolate breakdown in man. The British Journal of Nutrition 98 , 364–372.17403273
Gratz SW, Hazim S, Richardson AJ, Scobbie L, Johnstone AM, Fyfe C, Holtrop G, Lobley GE and Russell WR (2019). Dietary carbohydrate rather than protein intake drives colonic microbial fermentation in weight loss. European Journal of Nutrition 58 , 1147–1158.29464347
Hamer HM, Jonkers D, Venema K, Vanhoutvin S, Troost FJ and Brummer RJ (2008) Review article: the role of butyrate on colonic function. Alimentary Pharmacology & Therapeutics 27 , 104–119.17973645
Haro C, Rangel-Zúñiga OA, Alcalá-Díaz JF, Gómez-Delgado F, Pérez-Martínez P, Delgado-Lista J, Quintana-Navarro GM, Landa BB, Navas-Cortés JA, Tena-Sempere M, Clemente JC, López-Miranda J, Pérez-Jiménez F and Camargo A (2016) Intestinal microbiota is influenced by gender and body mass index. PLoS One 11 , e0154090.27228093
Holgate AM and Read NW (1983) Relationship between small bowel transit time and absorption of a solid meal. Digestive Diseases and Sciences 28 , 812–819.6884167
Holtrop G, Johnstone AM, Fyfe C and Gratz S (2012) Diet composition is associated with endogenous formation of N-nitroso compounds in obese men. The Journal of Nutrition 142 , 1652–1658.22833653
Kettle H, Louis P, Holtrop G, Duncan SH and Flint HJ (2015) Modelling the emergent dynamics and major metabolites of the human colonic microbiota. Environmental Microbiology 17 , 1615–1630.25142831
Lewis SJ and Heaton KW (1997) Increasing butyrate concentration in the distal colon by accelerating intestinal transit. Gut 41 , 245–251.9301506
Lobley GE, Holtrop G, Horgan GW, Bremner D, Fyfe C and Johnstone AM (2015) Responses in gut hormones and hunger to diets with either high protein or a mixture of protein plus free amino acids supplied under weight-loss conditions. The British Journal of Nutrition 113 , 1254–1270.25809236
Louis P and Flint HJ (2017) Formation of propionate and butyrate by the human colonic microbiota. Environmental Microbiology 19 , 29–41.27928878
Louis P, Hold GL and Flint HJ (2014) The gut microbiota, bacterial metabolites and colorectal cancer. Nature Reviews. Microbiology 12 , 661–672.25198138
Morrison DJ and Preston T (2016) Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism. Gut Microbes 7 , 189–200.26963409
Müller M, Hermes GDA, Canfora EE, Smidt H, Masclee AAM, Zoetendal EG and Blaak EE. (2020) Distal colonic transit is linked to gut microbiota diversity and microbial fermentation in humans with slow colonic transit. American Journal of Physiology. Gastrointestinal and Liver Physiology 318 , G361–G369.31869241
Neascu M, Fyfe C, Horgan G and Johnstone AM (2014) 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. The American Journal of Clinical Nutrition 100 , 548–558.24944057
Pryde SE, Duncan SH, Stewart CS, Hold GL and Flint HJ (2002) The microbiology of butyrate formation in the human colon. FEMS Microbiol Letts 217 , 133–139.12480096
Ramirez-Farias C, Slezak K, Fuller Z, Duncan A, Holtrop G and Louis P (2009) Effect of inulin on the human gut microbiota: Stimulation of Bifidobacterium adolescentis and Faecalibacterium prausnitzii. British Journal of Nutrition 101 , 541–550.18590586
Rao SSC, Kuo B, McCallum RW, Chey WD, DiBaise JK, Hasler WL, Koch KL, Lackner JM, Miller C, Saad R, Semler JR, Sitrin MD, Wilding GE and Parkman HP (2009). Investigation of colonic and whole-gut transit with wireless motility capsule and radiopaque markers in constipation. Clinical Gastroenterology and Hepatology 7 , 537–544.19418602
Reichardt N, Vollmer M, Holtrop G, Farquarson FM, Wefers D, Bunzel M, Duncan SH, Drew JE, Williams LM, Millligan G, Preston T, Morrison D, Flint HJ and Louis P (2018) Specific substrate-driven changes in human faecal microbiota compostion contrasts with functional redundancy in short chain fatty acid production. The ISME Journal 12 , 610–622.29192904
Richardson AJ, Calder AG, Stewart CS and Smith, A (1989) Simultaneous determination of volatile and non‐volatile acidic fermentation products of anaerobes by capillary gas chromatography. Letters in Applied Microbiology 9 , 5–8.
Rios-Covian D, González S, Nogacka AM, Arboleya S, Salazar N, Gueimonde M and de los Reyes-Gavilán CG. (2020) An overview on fecal branched short-chain fatty acids along human life and as related with body mass index: associated dietary and anthropometric factors. Frontiers in Microbiology 11 , 973.32547507
Roager HM, Hansen LBS, Bahl MI, Frandsen HL, Carvalho V, Gøbel RJ, Dalgaard MD, Plichta DR, Sparholt MH, Vestergaard H, Hansen T, Sicheritz-Pontén T, Nielsen HB, Pedersen O, Lauritzen L, Kristensen M, Gupta R and Licht TR. (2016) Colonic transit time is related to bacterial metabolism and mucosal turnover in the gut. Nature Microbiology 1 , 16093 10.1038/nmicrobiol.2016.93
Russell WR, Gratz S, Duncan SH, Holtrop G, Ince J, Scobbie L, Duncan G, Johnstone AM, Lobley GE, Wallace RJ, Duthie GG and Flint HJ (2011) High protein, reduced carbohydrate diets promote metabolite profiles likely to be detrimental to colonic health. The American Journal of Clinical Nutrition 93 , 1062–1072.21389180
Salonen A, Lahti L, Salojärvi J, Holtrop G, Korpela K, Duncan SH, Date P, Farquharson F, Johnstone AM, Lobley GE, Louis P, Flint HJ and de Vos W (2014) Impact of diet and individual variation on intestinal microbiota composition and fermentation products in obese men. The ISME Journal 8 , 2218–2230.24763370
Schwiertz A, Taras D, Schäfer K, Beijer S, Bos NA, Donus C and Hardt P (2010) Microbiota and SCFA in lean and overweight healthy subjects. Obesity 18 , 190–195.19498350
Scott KP, Gratz SW, Sheridan PO, Flint HJ and Duncan SH (2013) The influence of diet on the gut microbiota. Pharmacological Research 69 , 52–60.23147033
Smith EA and Macfarlane GT (1998) Enumeration of amino acid fermenting bacteria in the human large intestine: effects of pH and starch on peptide metabolism and dissimilation of amino acids. FEMS Microbiology Ecology 25 , 355–368.
Stephen AM, Wiggins HS and Cummings JH (1987) Effect of changing transit time on colonic microbial metabolism in man. Gut 28 , 601–609.3596341
Van der Wielen N, Moughan PJ and Mensink M (2017) Amino acid absorption in the large intestine of humans and porcine models. The Journal of Nutrition 148 , 1493–1498.
Vandeputte D, Falony G, Vieira-Silva S, Tito RY, Joossens M and Raes J (2016) Stool consistency is strongly associated with gut microbiota richness and composition, enterotypes and bacterial growth rates. Gut 65 , 57–62.26069274
Walker AW, Duncan SH, Leitch ECM, Child MW and Flint HJ (2005) pH and peptide supply can radically alter bacterial populations and short-chain fatty acid ratios within microbial communities from the human colon. Applied and Environmental Microbiology 71 , 3692–3700.16000778
Walker AW, Ince J, Duncan SH, Webster LM, Holtrop G, Ze X, Brown D, Stares MD, Scott P, Bergerat A, Louis P, McIntosh F, Johnstone AM Lobley GE, Parkhill J and Flint HJ (2011) Dominant and diet-responsive groups of bacteria within the human colonic microbiota. The ISME Journal 5 , 220–230.20686513
Wang YT, Mohammed SD, Farmer AD, Wang D, Zarate N, Hobson AR, Hellström PM, Semler JR, Kuo B, Rao SS, Hasler WL, Camilleri M and Scott SM (2015) Regional gastrointestinal transit and pH studied in 215 healthy volunteers using the wireless motility capsule: influence of age, gender, study country and testing protocol. Alimentary Pharmacology & Therapeutics 42 , 761–772.26223837
Wolin MJ (1960) A theoretical rumen fermentation balance. Journal of Dairy Science 43 , 1452–1459.
Yao CK, Muir JG and Gibson PR (2016) Review article: insights into colonic protein fermentation, its modulation and potential health implications. Alimentary Pharmacology & Therapeutics 43 , 181–196.26527169
Zhuang M, Shang W, Ma Q, Strappe P and Zhou Z (2019) Abundance of probiotics and butyrate‐production microbiome manages constipation via short‐chain fatty acids production and hormones secretion. Molecular Nutrition & Food Research 63 , 1801187.
