You are here

Article Text

Article menu
Inflammatory bowel disease
Original article
Abnormal fibre usage in UC in remission
  1. Sally L James1,
  2. Claus T Christophersen2,
  3. Anthony R Bird2,
  4. Michael A Conlon2,
  5. Ourania Rosella1,3,
  6. Peter R Gibson1,3,
  7. Jane G Muir1,3
  1. 1Eastern Health Clinical School, Monash University, Box Hill Hospital, Box Hill, Victoria, Australia
  2. 2Commonwealth Scientific & Industrial Research Organisation (CSIRO) Food Futures Flagship and CSIRO Animal, Food and Health Sciences, Adelaide, Australia
  3. 3Department of Gastroenterology, Central Clinical School, Monash University, Melbourne, Victoria, Australia
  1. Correspondence to Professor Peter Gibson, Department of Gastroenterology, Alfred Hospital, 99 Commercial Road, Melbourne, VIC 3004, Australia; peter.gibson{at}monash.edu

Abstract

Objective Colonic fermentation in patients with UC in remission was compared with that in matched healthy subjects on habitual diets and when dietary fibre was increased.

Design Fibre intake, faecal output of fibre (measured as non-starch polysaccharide (NSP)), starch, microbiota and fermentation products, and whole gut transit time (WGTT) were assessed in association with habitual diet and when dietary intake of wheat bran (WB)-associated fibre and high amylose-associated resistant starch (RS) was increased in an 8-week, randomised, single-blind, cross-over study.

Results Despite a tendency to lower habitual fibre intake in UC patients, faecal NSP and starch concentrations were threefold higher than in controls, whereas concentrations of phenols and short-chain fatty acids, pH and WGTT were similar. Increasing RS/WB intake was well tolerated. In controls (n=10), it more than doubled faecal NSP and starch excretion (p=0.002 for both), had no effect on NSP usage and reduced WGTT (p=0.024). In UC patients (n=19), high intake of RS/WB tended to normalise gut transit, but did not increase the proportion of NSP fermented. Increasing intake of RS/WB had little effect on faecal fermentation patterns or the structure of the microbiota. However, faeces from the UC cohort had lower proportions of Akkermansia muciniphila and increased diversity within Clostridium cluster XIVa compared to controls.

Conclusions Gut fermentation of NSP and starch is diminished in patients with UC. This cannot be explained by abnormal gut transit and was not corrected by increasing RS/WB intake, and may be due to abnormal functioning of the gut microbiota.

Trial registration number Australian New Zealand Clinical Trials Registry: ACTRN12614000271606.

  • Intestinal Bacteria
  • Chronic Ulcerative Colitis
  • Colonic Microflora
  • Diet
  • Dietary Fibre

https://doi.org/10.1136/gutjnl-2014-307198

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Significance of this study

    What is already known on this subject?

    • Fibre intake goals not achieved by patients with UC.

    • Gut microbiota differ in UC.

    • Dietary management of UC is not well established.

    What are the new findings?

    • Patients with quiescent UC often have reduced ability to ferment indigestible carbohydrates.

    • Increasing the dietary intake of a combination of resistant starch and wheat bran-associated non-starch polysaccharide is well tolerated in patients with UC in remission.

    • Increasing the dietary intake of this combination tends to normalise gut transit, but does not increase the proportion of carbohydrate fermented, nor increase the concentration or daily output of faecal short-chain fatty acids.

    • Diminished fermentation is not related to rapid gut transit, but more likely resides in the abnormal microbiota.

    How might it impact on clinical practice in the foreseeable future?

    • The effects of dietary manipulation directed at changing colonic physiology and/or gut microbiota in healthy subjects cannot be extrapolated to patients with UC.

    • Increasing the dietary intake of a combination of resistant starch and wheat bran in patients with UC may have some benefits to colonic physiology and health.

    Introduction

    The role of diet in the clinical management of patients with IBD generally remains uncertain, short of the generic desire for nutritional balance and adequacy. There is, however, no shortage of dietary hypotheses and beliefs, but translating them into a therapy with beneficial clinical outcomes has been infrequent and often unsuccessful. Yet diet can influence the biology of the gut considerably. One dietary target that has received some attention in patients with UC is fibre, most commonly defined as the fraction of the edible parts of plants or their extracts, or synthetic analogues, with the promotion of beneficial physiological effects. This definition includes polysaccharides (starch and non-starch), oligosaccharides and lignins, but excludes monosaccharides and disaccharides.1 Non-starch polysaccharides (NSP) are the major form of ‘dietary fibre’ found in the human diet, while the other major indigestible polysaccharides are starches that resist digestion (ie, resistant starch, RS). Fermentation of fibre by luminal bacteria delivers to the mucosa butyrate and other short-chain fatty acids (SCFA), which have beneficial properties.2 Fibre also may reduce less favourable bacterial fermentation of protein3 ,4 by the provision of adequate metabolic substrate for bacteria reducing the need to ferment protein, and by promotion of bacterial proliferation ensuring nitrogen is diverted away from metabolic to important structural synthetic pathways.

    However, the study of the effects of fibre on the bowel in patients with UC is limited. Epidemiological studies have suggested a weak or no association of dietary fibre intake prior to the development of UC.5 ,6 In patients with UC, fermentable fibres have been reported to increase faecal butyrate concentrations3 ,7–9 and some evidence of efficacy with fermentable substrates as recently reviewed.10 These limited data have not been translated into clinical practice guidelines, which generally advise against fibre in active disease and make little in the way of dietary recommendations in maintenance.11

    The delivery of butyrate to the colonic epithelium and mucosa, particularly in the distal colon, is generally regarded as a key mechanism for any putative positive effect of dietary fibre in UC, and RS is an attractive option as it preferentially delivers butyrate when fermented. There are two main types: RS1 that is physically trapped and so difficult for digestive enzymes to access (eg, coarsely ground grains), and RS2 in which nature of the starch granules themselves resists digestion (eg, high amylose maize starch granules). While RS is readily fermented, this occurs predominantly in the proximal colon with relatively poor delivery of fermentation products to the distal colon. However, studies in rats, pigs and humans have shown that combining RS with wheat bran (WB), of which its fibre content is slowly fermentable, shifts the fermentation distally with enhanced delivery of butyrate to the distal colon and rectum, and with hastening of gut transit.3 ,12 ,13 These are likely to be functionally important since, in a rat model of colorectal carcinogenesis, the combination of WB and RS was more protective of tumorigenesis in the distal colon than RS alone.12

    We hypothesised that the effect of a combination of WB and RS would have similar effects in patients with UC as healthy controls and, hence, such a combination of WB and RS might have efficacy due to the improved delivery of fermentation products to the distal colon. However, this assumes that the physiology of the large bowel and the functional capabilities of the microbiota in UC are normal. Such an extrapolation is hazardous given the reduced diversity of microbiota and variable intestinal transit that have been documented in patients with UC, particularly as subjects with IBD have lower abundance of clostridial groups and Faecalibacterium prausniztii specifically,14 which are strongly butyrate-producing.15 Thus, the current study aimed to determine how dietary NSP and RS is used in patients with UC in whom usual fibre intake, whole gut transit and faecal indices were first determined, and to examine the influence on those indices by increasing the dietary intake of WB and RS in a randomised, cross-over single-blinded controlled study. A secondary aim was to assess the tolerability of such a dietary change.

    Methods

    Subjects

    The study was conducted in two parts. In the first, a pilot study of five patients with UC in remission was conducted over four dietary periods with additional two arms examining supplementary effects of a probiotic preparation. Since the probiotic had no apparent influence on any index measured, those two arms were omitted from the subsequent study of a different cohort of patients with UC in remission and healthy subjects. The studies otherwise only differed in that a ‘baseline’ period of observation on habitual diet was added to the second part. Data from the two studies were combined.

    Subjects over the age of 18 years with UC in remission were recruited from IBD clinics from June 2002 until October 2008. UC was confirmed using standard criteria. Remission was defined as a colitis activity index (CAI) ≤4.16 Patients with Crohn's disease, indeterminate colitis or colectomy were excluded. Other exclusion criteria were the use of sulfasalazine, antibiotics, prebiotics, probiotics, laxatives or topical rectal therapies within the preceding 3 months, or other medications known to affect bowel function within the preceding 4 weeks. The only exception was mesalazine, which was not excluded despite its effects on gut microbiota.17 Control subjects over 18 years and without gastrointestinal symptoms, disease or gut surgery were recruited via advertising through the Monash University and Eastern Health websites. All participants gave written informed consent prior to study commencement. The study protocols were approved by the Eastern Health and Monash University Human Research and Ethics Committees.

    Study protocol

    Baseline medical history, physical examination and measurement of weight, height and Body Mass Index (BMI) were obtained for each participant. Participants in the second cohort had a 1-week period without dietary supplements during which baseline characteristics and dietary intake were assessed. Subjects continued their habitual diet throughout the study. They were randomly allocated (via computer-generated random numbers) to consume for 17 days either ‘low RS/WB’ foods containing 2–5 g RS and 2–5 g WB fibre per day, or ‘high RS/WB’ foods containing 15 g RS and 12 g WB fibre per day. These doses were chosen due to previous experience in healthy subjects, although the RS content was reduced to improve tolerance.2 ,3 Since the foods were replacing the equivalent food types in the subjects’ habitual diets, small amounts of WB were added to those in the low RS/WB group. The source of RS was coarsely ground high amylose maize (Himaize, Goodman Fielder, North Ryde, New South Wales, Australia), which has about 30% RS content, including RS types 1 and 2. The WB (about 45% fibre) was added to achieve the desired fibre content with a combination of commercially produced processed twig cereal (Goodman Fielder) and unprocessed WB (purchased from the local supermarket). The RS and WB were added to bread, cereal and muffins, produced in a commercial kitchen, in order to optimise blinding of the patient to the nature of the dietary arm. They were allocated on a daily basis to achieve the goal intakes and introduced at 25% of total final amount and increased incrementally to the full dose over 3 days to improve tolerance. The full dose of RS/WB was continued for 14 days. A wash-out period of 14 days was used between treatment periods to prevent a potential carry-over effect of the previous diet.

    Indices assessed

    Food intake

    Subjects completed a food diary for the last 7 days of each study period. Subjects were instructed to record all food consumed recorded as standardised serves and drinks by volumes. The food diary data were entered into Foodworks (Professional Edition V.7, Xyris Software (Australia), Brisbane, Queensland, Australia). Adherence to the intake of RS/WB-supplemented foods was assessed using a combination of returned foods, check lists and diary entries.

    Bowel symptoms and tolerability

    Disease activity was assessed using the CAI at the end of each study period. Symptoms were recorded in the diary for the baseline period (in second cohort) and the last 7 days of each dietary period. Tolerability of the interventions was recorded using a patient-reported 4-point Likert scale as detailed in online supplementary table S3.

    Faecal indices

    Total faecal output was collected by all subjects during the last 72 h of each dietary period. Faecal samples were placed in sealed plastic containers and frozen to −20°C in portable freezers immediately after their passage and transported frozen to minimise effect on downstream microbial analysis.18 On arrival at the laboratory, the faecal collection was weighed, thawed briefly (to minimise aerobic exposure), and homogenised. The pH was determined at 25°C according to established procedures,19 and aliquots frozen to −40°C or −80°C for further analysis to avoid sample degradation.20 Faecal aliquots of approximately 100 g were freeze-dried, enabling calculation of moisture content as an average of triplicate samples and measurement of residual fibre content. Faecal starch was determined enzymatically via a commercially available kit (Megazyme Australia, Warriewood, Sydney, Australia). The major form of dietary fibre is NSP and is free from associated starches. In this study, faecal NSP was quantified by the method described by Englyst,21 as previously applied.4 Internal standards confirmed the accuracy of the assays. Faecal NSP and starch excretion were reported as g/day, adjusted for total daily output and moisture content of the stool. An estimate of ‘NSP usage’ was calculated in each interventional period as the percentage of the difference between total dietary NSP intake (sum of the estimated NSP intake from known food composition data plus the actual NSP content as directly measured in the supplements) and faecal NSP output divided by the NSP intake. Usage of RS could not be calculated as there are no tables of RS content of food.

    Faecal SCFA were measured using high-performance gas chromatography (HPLC) as previously described.22 Total SCFA concentration was calculated as the sum of acetic, propionic, butyric, isobutyric, caproic, isovaleric and valeric acid concentrations. Faecal SCFA output (mmol/day) was calculated by multiplying SCFA concentration (mmol/L) by wet weight (g/day). Faecal phenols and cresols were also measured using HPLC based on published methods.23

    Faecal microbial abundances were assessed in a randomly selected subgroup of patients in the second cohort using denaturing gradient gel electrophoresis (DGGE)24 to assess diversity and quantitative PCR (QPCR)25 for quantification of specific bacterial groups, as described in detail in supplementary information. Bacterial targets and specific amplification details can be found in online supplementary table S1.

    Whole gut transit time

    A gelatine capsule containing 24 radio-opaque markers (Sitzmarks, Konsyl Pharmaceuticals, Easton Maryland, USA) was consumed the day prior to the 3-day stool collection. The times of consumption and of each stool collection were recorded, the number of markers identified on X-ray of the stool counted, and the whole gut transit time (WGTT) was calculated as previously described.26

    Statistical analysis

    Analyses were performed on subjects who completed both treatment arms. Statistical analysis for clinical data was performed using GraphPad Prism V.5.02. To examine the differences between controls and subjects with UC, unpaired t tests were used and, for non-parametric data, the Kruskal–Wallis test was used. Bonferroni corrections were made to determine statistical significance. Biochemical and microbiological data were analysed using the PRIMER 6+Permanova package (PRIMER-E Limited, Plymouth, UK). Differences between volunteers and diets during diet supplementation were assessed using a Euclidian distance matrix followed by a two-way and pair-wise Permanova analysis. DGGE banding patterns were analysed using a Bray–Curtis similarity matrix on log-transformed data followed by Permanova analysis. The Shannon index values, calculated based on log-transformed DGGE banding patterns, were analysed using a Euclidian distance matrix as described above. For analysis with less than 500 unique permutations Monte Carlo p values are stated.

    Results

    Characteristics of the subjects

    Thirty-seven subjects (12 controls and 25 UC) were recruited. Six UC and two control subjects did not complete one or both of the treatment arms, due to unplanned pregnancy, work commitments and medical procedures. Thus, data were analysed for 10 healthy controls (3 men) with a mean age of 41 (range 26–66) years, mean BMI 22.4 (SD 0.7) kg/m2, and 19 patients (9 men) aged 38 (18–72) years, BMI 25.8 (1.1) kg/m2. The two groups were similar except for BMI (p=0.047; t test). Two in the UC group and none in the healthy controls were current smokers. None of the controls, and all but two of the patients with UC, were taking medications—17 oral aminosalicylates, two oral corticosteroids and 7 thiopurines. Disease extent was proctitis in 4 (21%), distal colitis in 6 (32%) and extensive colitis in 9 (47%). All patients were in remission during the baseline period with a median (range) CAI of 0 (0–3).

    Baseline comparison of patients with UC and healthy subjects

    Baseline intake of nutrients in the patients with UC and healthy subjects are described in table 1. The UC group consumed significantly less iron and zinc than the control group. The mean intake of dietary fibre was about 1.5 times greater in controls than in those with UC, but this did not reach statistical significance.

    View this table:
    Table 1

    Dietary differences between UC and healthy subjects at baseline according to 7-day food diaries (from the second part of the study)

    Faecal daily weights, pH, phenolic compounds and SCFA (whether assessed in concentrations or daily excretion) showed no significant differences between the UC and control groups, as shown in table 2, except for caproate (p=0.002). By contrast, the output of faecal starch (p=0.003) and NSP (p=0.009) was around threefold higher in UC than controls (table 2). The mean (95% CIs) WGTT was similar in UC (45 (35 to 54) h) and controls (47 (40 to 54 h)).

    View this table:
    Table 2

    Baseline faecal weight and pH, and daily output of short-chain fatty acids (SCFA) and phenols in patients with UC (n=14) and control subjects (n=10) from the second part of the study (see matched dietary intake in online supplementary table S2)

    Effects of dietary supplementation with RS and WB

    Dietary intake

    Analysis of food intake during the interventional periods is shown in online supplementary table S2. The major differences were the intake of fibre and fat, which were higher in the high RS/WB arm for both subject groups.

    Tolerability and adherence

    Two UC subjects had a CAI over 4 (5 and 6) in the high RS/WB period, where the median CAI overall was 1 (0–6) compared to 1 (0–4) in the low RS/WB arm. Adherence was excellent (88–100%) in both groups in both dietary arms. The UC group reported numerically more symptoms at baseline and during the two dietary periods compared with controls. However, there was no difference in symptoms reported between the two treatment arms (see online supplementary table S3).

    Effects on faecal mass and metabolites

    As shown in figure 1, starch (p=0.002) and NSP (p=0.002) output more than doubled during the high RS/WB dietary arm in the healthy controls, but no significant changes were observed in the patients with UC. Increasing RS/WB intake had little effect on faecal pH, phenols and p-cresols or SCFA concentrations and daily excretion in either cohort, except valerate (p=0.035) and caproate (p=0.009), which were both higher in healthy volunteers (table 3). Molar ratios of propionate were slightly increased in UC patients (p=0.019) (data not shown).

    View this table:
    Table 3

    Effects of increasing the dietary intake of resistant starch and wheat bran on daily output of faecal short-chain fatty acids (SCFA) and phenols

    Figure 1
    Figure 1

    Faecal indices in healthy subjects and patients with UC supplemented with low or high amounts of resistant starch and wheat bran (RS/WB). (A) Faecal wet weight: Faecal output was significantly higher in the healthy controls with the high RS/WB supplementation (p=0.049) but not the UC group; (B) Faecal pH: There were no significant changes in pH in either group (p=0.68, healthy, p=0.18 UC subjects); (C) Faecal starch output: Faecal starch output in the healthy controls more than doubled with the high RS/WB supplements (p=0.002), but no difference was observed in the UC group (p=0.40); and (D) Faecal output of non-starch polysaccharide (NSP): NSP increased significantly in the healthy (p=0.015) and the UC (p=0.008) subjects. The bars represent mean values.

    Effects on NSP usage

    Increasing the intake of high RS/WB did not change NSP usage in patients with UC (8.2 (−10.0 to 26.3) % change) or healthy controls (−1.4 (−7.0 to 4.1) % change; p=0.41, unpaired t test), as shown in figure 2. NSP usage in patients with UC was significantly reduced compared with controls on either diet (p<0.001 for both).

    Figure 2
    Figure 2

    Percentage of non-starch polysaccharide usage in UC and control subjects with resistant starch and wheat bran (RS/WB) supplementation. The bars represent the median. No statistical significant change was observed for either subject group. Non-starch polysaccharide usage was significantly different between corresponding subject groups (p<0.001 for both).

    Effects on WGTT

    The effects of differing amounts of dietary fibre on WGTT are shown in figure 3. No relationship was identified between WGTT and faecal NSP, faecal starch or NSP usage for either the UC or control groups by linear regression analyses in the low RS/WB group representing habitual fibre intake (data not shown). The WGTT was also not associated with any demographic or clinical indices (data not shown). In the healthy subjects, increasing the intake of RS/WB reduced WGTT from a mean of 51 (40 to 62, 95% CI) h to 42 (37 to 47) h (p=0.024), while patients with UC showed no significant overall change (40.0 (31.6 to 48.4) to 39.9 (32.5 to 47.3) h).

    Figure 3
    Figure 3

    Effect of resistant starch and wheat bran (RS/WB) supplementation on whole gut transit time in: (A) healthy control subjects, where high RS/WB was associated with significantly shorter transit time (p=0.024); and (B) patients with UC, where no overall difference was observed. Patients with proctitis are represented by triangles, distal colitis by circles and extensive colitis by squares.

    When the patients were divided into subgroups according to whether they responded normally to the high RS/WB arm by hastening (n=7) or paradoxically by slowing (n=12) the WGTT, those with the abnormal response had a shorter WGTT (p=0.0036) and lower NSP usage (p=0.0043) in the low RS/WB arm. These results are shown in table 4.

    View this table:
    Table 4

    Subgroup analysis of patients with UC according to whether they demonstrated slower or faster whole gut transit times (WGTT) when consuming high-resistant starch and wheat bran (RS/WB) compared with low RS/WB

    Effects on faecal microbiota

    The absolute abundance of bacteria (per g faeces) was similar (data not shown), but microbiota composition was different between the two cohorts. Faeces of patients with UC had more diverse microbiota in their Clostridium cluster XIVa compared to healthy controls, according to the Shannon index, irrespective of which dietary arm was examined (table 5). QPCR disclosed a lower proportion of Akkermansia muciniphila (p=0.031) in patients with UC (table 5). For both cohorts, increasing the intake of RS/WB gave no indication of changes in relative (table 5) or absolute abundance (data not shown).

    Discussion

    There are several factors, including types of fibre being consumed, gut transit time, and the functional capabilities of gut microbiota, that influence the fermentation of carbohydrates and subsequent regional delivery of anti-inflammatory molecules, such as butyrate, to the colonic epithelium in the large bowel. The findings of the current study reflect this complexity. The present study has investigated these influencing factors on fermentative activity with regard to dietary long-chain carbohydrates in patients with quiescent UC in two ways—in association with habitual dietary intake and following an increase in the intake of RS and NSP in a double-blind, randomised study. When compared with a healthy cohort, those with UC in remission demonstrated considerable abnormalities characterised first by reduced proportion of ingested dietary NSP and starch being fermented in the large bowel and, second, by different and heterogeneous responses to RS/WB supplementation.

    View this table:
    Table 5

    Effects of increasing dietary intake of resistant starch and wheat bran on relative abundance of bacteria by quantitative PCR analysis, shown as percentage of total bacteria, and bacterial diversity by denaturing gradient gel electrophoresis expressed by the Shannon index

    Despite the tendency of the current cohort of patients with UC to consume less dietary fibre (as defined by the dietary software used) than did the healthy controls on their habitual diet, the faecal content of starch and NSP was near to three times higher than those in controls. The elevated faecal starch levels might have related to reduced intake of digestible starch, since the relative intake of RS was not quantified due to limitation of current databases of RS content of food. No differences in any the measured soluble faecal content, including pH, phenols and SCFA, were observed compared with those from controls. Previous studies in patients with UC have reported varying faecal concentrations of SCFA from lower to higher than healthy populations. 27–29 In the current cohort, the paradoxical situation of reduced fermentation but similar levels of fermentation products in the faeces is unlikely to be mediated by transit rate of digesta because WGTT, which is likely to predominantly represent colonic transit, was overall similar in the two groups and did not correlate with fermentation capacity, as estimated by NSP usage. However, transit specifically across distal segments was not assessed. The mismatch may reflect reduced uptake of SCFA by the colonic epithelium, although such uptake when measured in vivo using dialysis bags appeared to be normal.30 Therefore, altered microbial fermentative activity must have contributed to the reduced usage of NSP in the gut.

    The effect of increasing RS/WB intake—a combination more effective than RS alone in increasing the delivery of fermentation products to the entire large bowel, not just to the proximal colon3 ,13 ,14—on fermentation, microbiota and gut transit, were subsequently addressed. This dietary intervention also altered fat and possibly micronutrient intake, but the effects were the same in each subject group. Increasing RS/WB intake in the healthy controls mimicked those previously reported.4 The exception was that SCFA concentrations and output did not change. This may have related to the lower amount of RS used in the present study in order to minimise symptom induction. However, delivery of SCFA to the colonic mucosa presumably increased, as shown by similar output but greater total NSP and starch fermentation. The patients with UC showed heterogeneous responses, with some similar to those in the healthy controls and others with almost paradoxical effects. Of importance, high intake of RS/WB was not associated with consistent change in the activity of the colitis although two patients in that group had a mild increase in symptoms. There were no significant changes overall in the proportion of NSP used in either subject group. Thus, similar NSP usage in association with a greater substrate load indicates that there was greater delivery of SCFAs to the colonic epithelium.

    More striking was the effect that increasing intake of RS/WB had on WGTT. In healthy controls, transit was faster, as previously reported,4 as it was in a minority of patients with UC. In these groups, NSP usage did not change. In the majority of patients with UC, however, whole gut transit slowed with increasing RS/WB intake; these patients had significantly shorter WGTT and, not surprisingly, lower NSP usage on the low RS/WB diet. This, together with the longer transit that was associated with a tendency to increasing NSP usage, might suggest that at least some of the inefficient fermentation may have been related to abnormally fast transit. However, NSP usage did not normalise in either subgroup, suggesting an underlying problem with the functional capabilities of the microbiota.

    Indeed, a snapshot of the microbiota of the current cohort of patients with UC indicated abnormalities were present, as has been well documented previously31 ,32; for example, fewer Akkermansia were observed.33 Whether such difference provides a ready explanation for the reduced functional capacity of the microbiota in patients with UC is debateable since examination of 16S ribosomal RNA patterns does not provide direct information about the function of the microbiota. The faecal microbiota from UC patients had increased diversity of a cluster of bacteria including many butyrate producers and starch degraders. The small number of observations on which this finding is based must be acknowledged. Nevertheless, reduced diversity is a common finding in diseased colons, particular in association with IBD, where diversity is inversely proportional to the degree of intestinal inflammation.34 However, the focus was specifically on the Clostridium cluster XIVa, and may represent an alteration in dynamics of this cluster of bacteria between the two cohorts, from a cluster with fewer and more dominant species to one with more but less dominant species in UC patients. Therefore, such change can influence the degradation of starch and the production of SCFA. The main aim of the microbial studies, however, was to explore the effect of increasing fibre intake on relative bacterial abundance. Variability across subjects, and the relatively small number of subjects in whom faecal microbiota were analysed may have impaired the ability to demonstrate statistically significant changes and to define subgroups. Nevertheless, the lack of major change in composition of the microbiota was consistent with the lack of major change in fermentative capacity.

    The reduced capacity of the microbiota to ferment dietary carbohydrates could readily be examined using methodologies ex vivo by, for example, adapting faeces of patients with UC to models of gut microbiota.35 Examination of individual or collective bacteria populations has not been validated as a method of predicting fermentative capacity. Other factors such as regional motility, potentially modulated by local SCFA production,36 may influence the sum total capacity, and tools such as pH-sensing telemetry capsules37 may give important information about regional fermentation. Despite this reduced fermentation, faecal SCFA concentration and output was not significantly lower in the patients with UC. However, faecal output of SCFA do not necessarily reflect their production more proximally; intake of highly fermentable RS only had impact on faecal SCFA concentration and output when the fermentation was pushed distally by coingestion of WB.4 ,13 ,14 This indicates that the faecal concentrations of SCFA might reflect mainly the amount of fermentation occurring in the very distal colon. Indeed, the reason why the patients with UC in the current study had concentrations similar to those of the healthy controls despite a reduced fermentative capacity may have been due to the higher fermentable substrate load in the distal large bowel. If such substrate loads were achieved in the healthy controls, SCFA concentrations would most likely have been markedly increased in that cohort. Thus, the similar SCFA levels in the faeces were consistent with the other observations of reduced overall fermentative ability.

    If confirmed, reasons for the reduced fermentative capacity of the microbiota in patients with UC will require elucidation. One candidate would be the use of mesalazine, as this appears to alter the structure of the microbiota in a cohort with irritable bowel syndrome,17 and has effects on its metabolic activity, specifically hydrogen sulfide production.38 Only two patients in the present cohort were not taking mesalazine and there was no apparent signal regarding higher NSP usage in them. The effects of mesalazine on carbohydrate fermentation by faecal microbiota warrants further examination.

    The results of the present study may have implications for clinical practice. First, they reinforce the concept that findings in healthy subjects cannot be extrapolated to patients with UC. Second, they highlighted a considerable heterogeneity of effects of supplementing RS/WB in patients with UC, with two clear groups emerging—those with normal responses (the minority) and those with considerably diminished fermentative capacity and gut transit (the majority). Third, supplementation with RS/WB can improve the physiology of the bowel through provision of butyrogenic substrates as suggested by the greater absolute amount of NSP fermented and a tendency to normalisation of whole gut transit. These effects would potentially be beneficial to epithelial health and bowel function, and delivery of drugs, such as mesalazine, to the colonic mucosa. Fourth, the diminished capacity for fermentation that is indicated in many patients may have implications to pH changes in the proximal colon. These are obviously important for the release of drugs coated with pH-dependent resins as failure to reduce the pH sufficiently via carbohydrate fermentation might lead in some patients to failure of release of mesalazine when pH-dependent delivery mechanisms are used. Finally, despite improvement of the absolute amount of NSP and starch being fermented, the functional abnormality of the fermentative ability of colonic microbiota is not corrected by changing RS/WB intake in this way. At least hypothetically, such supplementation may have been clinically beneficial, but ways of effectively regulating the function of the microbiota in such patients need to be developed.

    In conclusion, the colonic fermentative ability is heterogeneous in patients with quiescent UC. Wide variations in gut transit are strongly associated with this heterogeneity. Dietary supplementation of a combination of RS and WB seemed to normalise transit and improve delivery of potentially beneficial fermentation products, such as butyrate, to the colonic mucosa, but did not correct the fermentative deficiency. While differences in the composition of the microbiota were evident between patients with UC and healthy controls, no change was evident by increasing the intake of RS/WB. Further studies are needed to determine why fermentation by colonic microbiota appears compromised in patients with UC, and whether the underlying problem can be corrected by dietary or other approaches.

    Acknowledgments

    The authors thank Kelly Liels and Corinna Bennett for technical assistance, and Goodman Fielder for the wheat bran twig cereal and Himaize.

    References

    1. ↵
      Anonymous. Food Standards Australia and New Zealand, Application 3, Appendices 4 & 5. Canberra. 2003. http://www.foodstandards.gov.au
    2. ↵
      1. Hamer HM,
      2. Jonkers D,
      3. Venema K,
      4. et al
      . Review article: the role of butyrate on colonic function. Aliment Pharmacol Ther 2008;27:104–19.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Birkett AM,
      2. Phillips J,
      3. Jones G,
      4. et al
      . Resistant starch lowers fecal concentrations of ammonia and phenols in humans. Am J Clin Nutr 1996;63:766–72.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Muir JG,
      2. Yeow EGW,
      3. Keogh J,
      4. et al
      . Combining wheat bran with resistant starch has more beneficial effects on fecal indices than wheat bran alone. Am J Clin Nutr 2004;79:1020–8.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Spooren CE,
      2. Pierik MJ,
      3. Zeegers MP,
      4. et al
      . Review article: the association of diet with onset and relapse in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2013;38:1172–87.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Ananthakrishnan AN,
      2. Khalili H,
      3. Konijeti GG,
      4. et al
      . A prospective study of long-term intake of dietary fiber and risk of Crohn's disease and ulcerative colitis. Gastroenterology 2013;145:970–7.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Ahmed R,
      2. Segal I,
      3. Hassan H
      . Fermentation of dietary starch in humans. Am J Gastroenterol 2000;95:1017–20.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Hallert C,
      2. Bjorck I,
      3. Nyman M,
      4. et al
      . Increasing fecal butyrate in ulcerative colitis patients by diet: controlled pilot study. Inflamm Bowel Dis 2003;9:116–21.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Jenkins DJA,
      2. Vuksan V,
      3. Kendall CWC,
      4. et al
      . Physiological effects of resistant starches on fecal bulk, short chain fatty acids, blood lipids and glycemic index. J Am Coll Nutr 1998;17:609–16.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Wedlake L,
      2. Slack N,
      3. Andreyev HJ,
      4. et al
      . Fiber in the treatment and maintenance of inflammatory bowel disease: a systematic review of randomized controlled trials. Inflamm Bowel Dis 2014;20:576–86.
      OpenUrlCrossRefPubMed
    11. ↵
      1. Travis SPL,
      2. Stange EF,
      3. Lemann M,
      4. et al
      . European evidence-based Consensus on the management of ulcerative colitis: current management. J Crohn's Colitis 2008;2:24–62.
      OpenUrlFREE Full Text
    12. ↵
      1. Young GP,
      2. Folino M,
      3. McIntyre A,
      4. et al
      . Wheat bran suppresses potato-starch potentiated tumorigenesis at the aberrant crypt stage in a rat model of colorectal cancer. Gastroenterology 1996;100:508–14.
      OpenUrl
    13. ↵
      1. Govers M,
      2. Gannon N,
      3. Dunshea F,
      4. et al
      . Wheat bran affects the site of fermentation of resistant starch and luminal indexes related to colon cancer risk: a study in pigs. Gut 1999;45:840–7.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Sokol H,
      2. Seksik P,
      3. Furet JP,
      4. et al
      . Low counts of Faecalibacterium prausnitzii in colitis microbiota. Inflamm Bowel Dis 2009;15:1183–9.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Benus RF,
      2. van der Werf TS,
      3. Welling GW,
      4. et al
      . Association between Faecalibacterium prausnitzii and dietary fibre in colonic fermentation in healthy human subjects. Br J Nutr 2010;104:693–700.
      OpenUrlCrossRefPubMed
    16. ↵
      1. D'Haens G,
      2. Sandborn WJ,
      3. Feagan BG,
      4. et al
      . A review of activity indices and efficacy end points for clinical trials of medical therapy in adults with ulcerative colitis. Gastroenterology 2007;132:763–86.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Andrews CN,
      2. Griffiths TA,
      3. Kaufman J,
      4. et al
      . Mesalazine (5-aminosalicylic acid) alters faecal bacterial profiles, but not mucosal proteolytic activity in diarrhoea-predominant irritable bowel syndrome. Aliment Pharmacol Ther 2011;34:374–83.
      OpenUrlCrossRefPubMed
    18. ↵
      1. Cardona S,
      2. Eck A,
      3. Cassellas M,
      4. et al
      . Storage conditions of intestinal microbiota matter in metagenomic analysis. BMC Microbiol 2012;12:158.
      OpenUrlCrossRefPubMed
    19. ↵
      1. Muir JG,
      2. Walker KZ,
      3. Kaimakamis MA,
      4. et al
      . Modulation of fecal markers relevant to colon cancer risk: a high- starch Chinese diet did not generate expected beneficial changes relative to a Western-type diet. Am J Clin Nutr 1998;68:372–9.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Bahl MI,
      2. Bergström A,
      3. Licht TR
      . Freezing fecal samples prior to DNA extraction affects the Firmicutes to Bacteroidetes ratio determined by downstream quantitative PCR analysis. FEMS Microbiol Lett 2012;329:193–7.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Englyst HN,
      2. Quigley ME,
      3. Hudson GJ
      . Determination of dietary fibre as non-starch polysaccharides with gas-liquid chromographic, high-performance liquid chromography or spectrophotometric measurement of constituent sugars. Analyst 1994;119:1497–509.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Brinkworth GD,
      2. Noakes M,
      3. Clifton PM,
      4. et al
      . Comparative effects of very low-carbohydrate, high-fat and high-carbohydrate, low-fat weight-loss diets on bowel habit and faecal short-chain fatty acids and bacterial populations. Br J Nutr 2009;11:1493–502.
      OpenUrl
    23. ↵
      1. Yoshikawa M,
      2. Taguchi Y,
      3. Arashidani K,
      4. et al
      . Determination of cresols in urine by high-performance liquid chromatography. J Chromatog 1986;362:425–9.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Abell GCJ,
      2. McOrist AL
      . Assessment of the diversity and stability of faecal bacteria from healthy adults using molecular methods. Microb Ecol 2007;19:229–40.
      OpenUrl
    25. ↵
      1. Walter J,
      2. Hertel C,
      3. Tannock GW,
      4. et al
      . Detection of Lactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by using group-specific PCR primers and denaturing gradient gel electrophoresis. Appl Environ Microbiol 2001;67:2578–85.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Cummings JH,
      2. Jenkins DJ,
      3. Wiggins HS
      . Measurement of the mean transit time of dietary residue through the human gut. Gut 1976;17:210–18.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Treem WR,
      2. Ahsan N,
      3. Shoup M,
      4. et al
      . Fecal short-chain fatty acids in children with inflammatory bowel disease. J Pediatr Gastroenterol Nutr 1994;18:159–64.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Vernia P,
      2. Gnaedinger A,
      3. Hauck W,
      4. et al
      . Organic anions and the diarrhea of inflammatory bowel disease. Dig Dis Sci 1988;33:1353–8.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Takaishi H,
      2. Matsuki T,
      3. Nakazawa A,
      4. et al
      . Imbalance in intestinal microflora constitution could be involved in the pathogenesis of inflammatory bowel disease. Int J Med Microbiol 2008;298:463–72.
      OpenUrlCrossRefPubMed
    30. ↵
      1. Hove H,
      2. Holtug K,
      3. Jeppesen PB,
      4. et al
      . Butyrate absorption and lactate secretion in ulcerative colitis. Dis Colon Rectum 1995;38:519–25.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Manichanh C,
      2. Rigottier Gois L,
      3. Bonnaud E,
      4. et al
      . Reduced diversity of faecal microbiota in Crohn's disease revealed by a metagenomic approach. Gut 2006;55:205–11.
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Manichanh C,
      2. Borruel N,
      3. Casellas F,
      4. et al
      . The gut microbiota in IBD. Nat Rev Gastroenterol Hepatol 2012;9:599–608.
      OpenUrlCrossRefPubMed
    33. ↵
      1. Png CW,
      2. Linden SK,
      3. Gilshenan KS,
      4. et al
      . Mucolytic bacteria with increased prevalence in ibd mucosa augment in vitro utilization of mucin by other bacteria. Am J Gastroenterol 2010;105:2420–8.
      OpenUrlCrossRefPubMed
    34. ↵
      1. Knights D,
      2. Lassen KG,
      3. Xavier RJ
      . Advances in inflammatory bowel disease pathogenesis: linking host genetics and the microbiome. Gut 2013;62:1505–10.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Hobden MR,
      2. Martin-Morales A,
      3. Guérin-Deremaux L,
      4. et al
      . In vitro fermentation of NUTRIOSE(®) FB06, a wheat dextrin soluble fibre, in a continuous culture human colonic model system. PLoS ONE 2013;8:e77128.
      OpenUrlCrossRefPubMed
    36. ↵
      1. Leonel AJ,
      2. Alvarez-Leite JI
      . Butyrate: implications for intestinal function. Curr Opin Clin Nutr Metab Care 2012;15:474–9.
      OpenUrlCrossRefPubMed
    37. ↵
      1. Nugent SG,
      2. Kumar D,
      3. Rampton DS,
      4. et al
      . Intestinal luminal pH in inflammatory bowel disease: possible determinants and implications for therapy with aminosalicylates and other drugs. Gut 2001;48:571–7.
      OpenUrlAbstract/FREE Full Text
    38. ↵
      1. Pitcher MC,
      2. Beatty ER,
      3. Cummings JH
      . The contribution of sulphate reducing bacteria and 5-aminosalicylic acid to faecal sulphide in patients with ulcerative colitis. Gut 2000;46:64–72.
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Contributors SLJ: design; data acquisition, analysis and interpretation; writing and editing; final approval; CTC, ARB, MAC: data acquisition, analysis and interpretation; editing; final approval; JGM: design; data analysis and interpretation; editing; final approval; OR: data acquisition, analysis and interpretation; editing; final approval; PRG: design; data analysis and interpretation; writing and editing; final approval.

    • Competing interests SLJ was in receipt of a scholarship from the Gastroenterological Society of Australia.

    • Ethics approval Eastern Health Research and Ethics Committee and Monash University Research Ethics Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.