Article Text


Original article
Butyricicoccus pullicaecorum in inflammatory bowel disease
  1. Venessa Eeckhaut1,
  2. Kathleen Machiels2,
  3. Clémentine Perrier2,
  4. Carlos Romero3,
  5. Sofie Maes1,
  6. Bram Flahou1,
  7. Marjan Steppe1,
  8. Freddy Haesebrouck1,
  9. Benedikt Sas4,
  10. Richard Ducatelle1,
  11. Severine Vermeire2,
  12. Filip Van Immerseel1
  1. 1Department of Pathology, Bacteriology and Avian Diseases, Ghent University, Faculty of Veterinary Medicine, Merelbeke, Belgium
  2. 2Department of Gastroenterology, Translational Research Center for Gastrointestinal Disorders (TARGID), University Hospital Gasthuisberg, Leuven, Belgium
  3. 3Department of Animal Science, School of Agricultural Engineering, Technical University of Madrid, Madrid, Spain
  4. 4Centre of Excellence Food2Know, Ghent University, Ghent, Belgium
  1. Correspondence to Professor Filip Van Immerseel, Department of Pathology, Bacteriology and Avian Diseases, Ghent University, Faculty of Veterinary Medicine, Salisburylaan 133, Merelbeke 9820, Belgium; filip.vanimmerseel{at}


Objective Many species within the phylum Firmicutes are thought to exert anti-inflammatory effects. We quantified bacteria belonging to the genus Butyricicoccus in stools of patients with ulcerative colitis (UC) and Crohn's disease (CD). We evaluated the effect of Butyricicoccus pullicaecorum in a rat colitis model and analysed the ability to prevent cytokine-induced increases in epithelial permeability.

Design A genus-specific quantitative PCR was used for quantification of Butyricicoccus in stools from patients with UC or CD and healthy subjects. The effect of B pullicaecorum on trinitrobenzenesulfonic (TNBS)-induced colitis was assessed and the effect of B pullicaecorum culture supernatant on epithelial barrier function was investigated in vitro.

Results The average number of Butyricicoccus in stools from patients with UC and CD in active (UC: 8.61 log10/g stool; CD: 6.58 log10/g stool) and remission phase (UC: 8.69 log10/g stool; CD: 8.38 log10/g stool) was significantly lower compared with healthy subjects (9.32 log10/g stool) and correlated with disease activity in CD. Oral administration of B pullicaecorum resulted in a significant protective effect based on macroscopic and histological criteria and decreased intestinal myeloperoxidase (MPO), tumour necrosis factor α (TNFα) and interleukin (IL)-12 levels. Supernatant of B pullicaecorum prevented the loss of transepithelial resistance (TER) and the increase in IL-8 secretion induced by TNFα and interferon γ (IFN gamma) in a Caco-2 cell model.

Conclusions Patients with inflammatory bowel disease have lower numbers of Butyricicoccus bacteria in their stools. Administration of B pullicaecorum attenuates TNBS-induced colitis in rats and supernatant of B pullicaecorum cultures strengthens the epithelial barrier function by increasing the TER.

  • IBD
  • Probiotics
  • Butyrate
  • Epithelial Barrier
  • Inflammation

Statistics from

Significance of this study

What is already known on this subject?

  • The gut microbiota composition of patients with inflammatory bowel disease (IBD) is significantly altered compared with healthy individuals.

  • Multiple studies using metagenomic sequencing tools have identified bacterial genera that are significantly reduced or upregulated in stools and gut biopsies of patients with IBD. The most well known is Faecalibacterium prausnitzii, which is an important anti-inflammatory bacterium, and is reduced in abundance in stool samples of patients with IBD.

  • The bacterial genus Butyricicoccus was recently found to be reduced in intestinal biopsies and stool samples from patients with IBD. However, no functional studies using a member of this genus have been carried out.

What are the new findings?

  • Quantitative PCR data confirm the decreased abundance of the genus Butyricicoccus in stool samples of patients with IBD.

  • Butyricicoccus pullicaecorum is able to decrease lesion sizes and inflammation in a rat colitis model.

  • Supernatant of B pullicaecorum cultures prevents cytokine-induced epithelial integrity losses in an in vitro cell culture model.

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

  • Butyricicoccus bacteria are conceptually attractive as probiotics because they normally reside in the gut and could potentially result in permanent alterations of the gastrointestinal microbiota.

  • Preventive use of B pullicaecorum in patients with IBD may have potential to prevent disease relapse.


The pathogenesis of the inflammatory bowel disease (IBD) phenotypes, Crohn's disease (CD) and ulcerative colitis (UC), involves a complex interplay of environmental triggers, the immune system, the genotype and the intestinal microbiome.1 The prevalence of IBD is highest in North America, northern Europe and the UK and affects 100–200 patients per 100 000 people.2 Patients with IBD experience diverse symptoms and lesions ranging from weight loss, abdominal pain, diarrhoea and intestinal obstruction to ulceration and perforation of the gastrointestinal tract.3 The conventional treatment, which is generally considered supportive rather than curative, consists of aminosalicylates, corticosteroids, immunomodulators and antitumor necrosis factor α (TNFα) to counteract inflammation.4 While these therapies have shown efficacy in treating the symptoms of IBD, they can induce serious side effects.5 In addition, these therapies do not directly affect the perturbations in the gut microbiome, termed dysbiosis, which is considered an important trigger of disease induction.6 Patients with IBD exhibit a reduced bacterial biodiversity in the intestine compared with healthy individuals. Specifically, an increase in numbers of Escherichia coli7 ,8 and lower counts of specific bacterial species belonging to the Firmicutes and Bacteroidetes phyla have been documented.1 ,7 ,8 At the functional level, the butyrate producers belonging to clostridial clusters IV and XIVa of the Firmicutes phylum are significantly less abundant in the gastrointestinal microbiome of patients with CD, explaining the decrease in short chain fatty acid (SCFA) concentrations in faecal extracts of patients with IBD.9 Among the SCFAs, butyrate, the key energy source for colonocytes, exerts important effects on intestinal function, including the repression of pro-inflammatory cytokine production.10 The effect of butyrate has been studied in patients with colonic inflammation using rectal enemas. While some studies revealed an improvement in clinical and inflammatory parameters,11–13 others did not find beneficial effects.14 Because of the effects of butyrate on gut function and because butyrate-producing clostridial cluster IV and XIVa bacteria seem to be depleted in the gut of patients with IBD, oral administration of such strains may have therapeutic benefits.

Frank et al7 showed a correlation between shifts in abundance of the clostridial cluster IV genera Faecalibacterium and Butyricicoccus, and the IBD phenotype in resected tissues from patients with IBD. While the lower abundance of Faecalibacterium prausnitzii in IBD has been confirmed in stool samples15 and the beneficial effects of administration of this bacterial species in a colitis model16 has been documented, no data are available on the abundance of strains of the Butyricicoccus genus in IBD versus healthy control subjects and no data are published yet on its anti-colitic effect. Recently, we succeeded for the first time in isolating and culturing a strain of the Butyricicoccus genus and named it Butyricicoccus pullicaecorum.17 The aim of the present study was to analyse the abundance of Butyricicoccus bacteria in stool samples from patients with IBD versus healthy subjects and to study the preventive potential of B pullicaecorum to suppress inflammation and concomitantly ameliorate the lesions in a 2,4,6-trinitrobenzenesulfonic acid (TNBS)-induced colitis model in rats. The effect of B pullicaecorum culture supernatant on transepithelial resistance (TER) and interleukin (IL)-8 secretion was investigated using a Caco-2 cell model.


Quantification of Butyricicoccus in human faecal samples

DNA extraction

Stool samples were collected from 88 healthy subjects (39 men and 49 women, median age 41 years), 51 patients with CD (23 men and 28 women, median age 39 years) and 91 patients with UC (54 men and 37 women, median age 44 years). Of the patients with CD, 24 were in the active phase (Harvey-Bradshaw Index (HBI)≥518) and 27 were in the inactive phase (HBI<5). Of the patients with UC, 25 were in the active phase (Partial Mayo score (= Mayo score without endoscopy) 2 or 319) and 66 patients were in the inactive phase (score 0 or 1). None of the patients were following a specific diet, treated with antibiotics or using probiotics within 4 weeks before the faecal sampling. Upon collection, the faecal samples were immediately frozen at −80°C. Total bacterial DNA was extracted from 500 mg faeces according to a modified version of the protocol of Pitcher and colleagues.20

Primer design

The 16S rRNA gene sequences of 27 isolates belonging to the genus Butyricicoccus and 31 closely related genera belonging to the family Ruminococcaceae were obtained from NCBI GenBank and used for primer design. After multiple sequence alignment, two genus-specific primer sets spanning a target region of 69 and 895 bp in the 16S rRNA gene of Butyricicoccus were designed using KODON (Applied Maths BVBA, Sint-Martens-Latem, Belgium) (table 1). The selected primer target sites were tested for specificity by submitting the sequences to the Probe Match program provided by Ribosomal Database Project II.21

Table 1

Butyricicoccus-specific 16S rRNA-targeted primers

Real-time quantitative PCR

Members of the genus Butyricicoccus were quantified using the Applied Biosystems 7500 Fast Real-Time PCR System (Applied Biosystems, Foster City, California, USA). Amplification and detection were carried out in 96-well plates using SYBR-green low ROX 2x master mix (Applied Biosystems). Each reaction was done in triplicate in a 12 µl total reaction mixture using 2 µl of appropriate dilutions of the DNA sample and 0.3 µM final quantitative PCR primer concentration (table 1). The amplification cycle used was one cycle of 95°C for 10 min, followed by 40 cycles of 95°C for 30 s, 60°C for 1 min. A stepwise increase of the temperature from 50°C to 95°C (at 10 s/0.5°C) was added and melting curve data were analysed to confirm the specificity of the reaction. For construction of the standard curve, the 895 bp long PCR product was generated using the standard PCR primers listed in table 1 and DNA from B pullicaecorum. After purification and determination of the DNA concentration, the volume of the linear double-stranded DNA standard was adjusted to 6.04×109 copies µl−1 assuming an average molecular weight of 660 per nucleotide pair. This stock solution was 10-fold serially diluted to obtain a standard series from 6.04×107 to 6.04×101 copies µl−1. The copy numbers of samples were determined by reading off the standard series with the Ct values of the samples. Gene copy numbers were expressed as log10 values per gram wet weight of faeces. To exclude a dilution effect in diarrhoea samples, a precisely weighed portion of 0.5 g of each sample was taken and lyophilised. The weight difference before and after lyophilisation represented the water content and was used to express the percentage of dry faeces. Using this percentage, the gene copy numbers as log10 values per gram dry weight of faeces was calculated.

In vivo trial with B pullicaecorum

Strains, culture conditions and inoculum preparation

B pullicaecorum (CCUG 55265) and F prausnitzii (M21/2) were grown overnight at 37°C in an anaerobic (84% N2, 8% CO2 and 8% H2) workstation (Ruskinn Technology, Bridgend, UK) in M2GSC broth medium22 at pH 6. The bacterial cells were collected by centrifugation (10 min, 5000 g, 37°C) and washed in Hank's balanced salt solution (HBSS) without Ca2+/Mg2+ pH 6 (Life Technologies, Bleiswijk, The Netherlands), supplemented with 1 mg/ml cysteine-HCl (Sigma-Aldrich, St Louis, Missouri, USA). The resulting pellet was suspended in the above-mentioned HBSS to reach a final concentration of 109 colony forming units (CFU)/ml. Rats were daily administered 109 CFU of freshly cultivated bacteria via intragastric gavage, under anaesthesia by inhalation of isoflurane (IsoFlo, EcuPhar Animal Health, Oostkamp, Belgium).

Rats and induction of TNBS colitis

Male Wistar rats (450–500 g) were supplied by Janvier (St Berthevin, France), housed five per polycarbonate cage and maintained on a 12 h light–dark cycle. The animals were fed standard laboratory chow (Ssniff R/M-H, Bioservices, Uden, The Netherlands) and allowed access to water ad libitum. All rats were acclimatised to the study conditions before entering experiments, at 12–14 weeks of age.

Experimental colitis was induced by a single intrarectal instillation of TNBS (Sigma-Aldrich) in 50% ethanol (vol/vol) following a 24 h fasting period. Enemas were gently instilled through a soft endotracheal tube (Vygon Corporation, Montgomeryville, Pennsylvania, USA) inserted 8 cm into the anus under inhalation anaesthesia with 3% isoflurane. The same procedure was used with rats from the non-colitic group but they were administered phosphate-buffered saline (PBS). After instillation, the rats were kept upside down for 1 min.

All procedures involving animals were approved by the ethical committee of the Faculty of Veterinary Medicine of Ghent University.

Experimental design

Rats were randomly allocated to four groups of 10 animals. No bacterial strains were administered to the animals of two groups, that is, the colitis and non-colitis control groups. Instead, for 8 consecutive days, they were inoculated intragastrically once a day with the vehicle (HBSS+1 mg/ml cysteine-HCl pH 6). The other groups were administered 109 CFU B pullicaecorum or F prausnitzii once a day from day 7 preceding to day 1 following induction of colitis. Forty-eight hours after instillation of 400 µl of 25 mg/ml TNBS in 50% ethanol, all rats were sacrificed.

Necropsy and colonic damage

The body weight was recorded daily. After euthanasia using intracardial embutramid (T61, MSD, Brussels, Belgium) injection under deep isoflurane anaesthesia, the colon was removed and opened longitudinally. After gently rinsing with cold sterile saline, the weight and length of the colon and the length of the lesion was recorded. The colon weight/colon length (g/cm) and colon weight/body weight (mg/g) ratios were calculated as parameters to assess the degree of colon oedema. Macroscopic assessment of the disease grade was scored using the criteria described by Wallace et al23 (table 2). Segments of the colon were taken, fixed in 10% buffered formalin and embedded in paraffin for histological examination. Other colon segments were frozen in liquid nitrogen and stored at −80 °C for quantitative analysis of MPO activity, TNFα and IL-12 concentration.

Table 2

Macroscopic scoring of colitis23

Histological scoring

Paraffin-embedded samples were sliced into 5 µm sections and stained with hematoxylin eosin for light microscopic examination. Samples were blindly analysed for evidence of inflammation by an experienced pathologist. Intestinal injury was scored on a scale of 1–20 as described by Lahat et al24 (table 3). For each parameter a score was given as indicated in table 3 and then added together with a maximum value of 20, providing a global assessment of colitis.

Table 3

Histological scoring of colitis24

MPO activity

Frozen colonic segments were assayed for MPO activity (units per mg of protein), as described previously.25

Cytokine measurements

Cytokine levels were determined after homogenisation of colon tissue in cold PBS containing a protease-inhibitor cocktail (Sigma-Aldrich). After 20 min centrifugation at 10 000 g, the supernatants were used at various dilutions for quantitative analysis of IL-12 and TNFα concentrations using commercially available ELISA kits (eBioscience, San Diego, California, USA). Results were expressed as picograms of cytokine per milligram of protein.

Evaluation of B pullicaecorum's supernatant on the intestinal permeability

TER measurements

Caco-2 cells were grown on Costar Transwell polycarbonate filters of 65 mm diameter, and 0.4 μm pore size (Corning Incorporated, Corning, New York, USA) for 2 weeks until TER reached at least 1000 Ω*cm2. Cells were then exposed to a filtered supernatant of B pullicaecorum or medium complemented with 10 and 20 mM butyrate in the apical compartment. Simultaneously cells were stimulated with or without 100 ng/ml TNFα (R&D Systems, Minneapolis, Massachusetts, USA) and 100 ng/ml interferon γ (IFNγ) in the basolateral compartment to mimic inflammation. Supernatants were diluted in Caco-2 medium in a ratio of 4–6. TER was measured in duplicate before and 24 h after stimulation and each condition was performed at least in triplicate. At the end of the experiment, the basolateral medium was harvested and stored at −20°C. IL-8 was measured in that medium using an ELISA (BD Pharmingen, San Diego, California, USA).

Statistical analysis

A non-parametric Mann–Whitney U test was performed to compare the mean number of Butyricicoccus bacteria in the different groups. All other values in the figures and tables are presented as least-square means±SEM. Means of all variables, except scores, were compared using a Fisher's protected t test. The non-parametric Kruskal–Wallis test was used to compare differences among groups for Wallace score and histological measurements. Differences were considered statistically significant at p<0.05.


Quantification of Butyricicoccus bacteria in human stools

The average number of bacteria belonging to the genus Butyricicoccus was 8.69 log10/g wet faeces and 8.38 log10/g wet faeces respectively in patients with UC and CD in remission, 8.61 log10/g wet faeces and 6.58 log10/g wet faeces respectively in patients with UC and CD in the active disease phase and 9.32 log10/g wet faeces in healthy subjects. The average number of Butyricicoccus bacteria was significantly (p<0.0001) lower in the stools of IBD patient groups compared with the healthy controls. In addition, a significantly (p<0.0188) lower level of Butyricicoccus species was observed in the faecal microbiota of patients with active CD compared with CD in remission (figure 1). Lyophilisation of the stool samples showed a significantly higher water content in stools of patients with IBD compared with stools of the healthy subjects. Nevertheless the same significant differences were observed when comparing the average number of Butyricicoccus bacteria expressed as log10 per g dry weight of faeces in stools of patients with IBD and healthy individuals (see online supplementary table S1).

Figure 1

Quantification of Butyricicoccus bacteria in stool samples from 142 patients with inflammatory bowel disease and 88 healthy subjects. Bacterial density is expressed as log10 copy number of 16S rRNA genes for Butyricicoccus per g of stool. ***p<0.0001; *p<0.0188. C, Crohn's disease; UC, ulcerative colitis.

Preventive effect of B pullicaecorum in TNBS-induced colitis

Administration of B pullicaecorum and F prausnitzii did not induce any symptoms of diarrhoea and did not affect weight gain (data not shown). Once colitis was induced using TNBS, the B pullicaecorum-treated rats showed an overall lower impact of the TNBS treatment compared with the F prausnitzii-treated and the colitis control group. Administration of B pullicaecorum prevented loss of body weight (figure 2). At necropsy, the non-colitis control group and the group treated with B pullicaecorum and F prausnitzii had an average body weight that was significantly (p<0.001) higher than the colitis control group. The anti-inflammatory effect of B pullicaecorum was evidenced macroscopically by a significantly (p<0.001) lower macroscopic score and a significant (p<0.001) reduction in the colon/body weight ratio and the colon weight/length ratio compared with the colitis control rats (table 4). The surface area with mucosal ulceration was significantly smaller in the colon of rats in the B pullicaecorum-treated group compared with the F prausnitzii-treated and colitis control group (data not shown). Histological assessment of the colonic samples from rats treated with the probiotics revealed a more pronounced recovery in the intestinal architecture compared with the colitis control group (table 4). The improvement in microscopic colonic damage was accompanied by a significant (p<0.001) reduction in colonic MPO activity (figure 3A). The colonic inflammation induced by TNBS was characterised by significantly (p<0.001) increased levels of colonic TNFα and IL-12. Treatment with B pullicaecorum or F prausnitzii resulted in a significant (p<0.001) reduction of colonic TNFα and IL-12 levels, to a level comparable with non-colitis normal rats (figure 3B, C).

Table 4

The effect of oral administration of Faecalibacterium prausnitzii and Butyricicoccus pullicaecorum on macroscopic and microscopic manifestations of TNBS-induced colitis in rats

Figure 2

Effect of administration of Faecalibacterium prausnitzii and Butyricicoccus pullicaecorum on body weight loss in rats with trinitrobenzenesulfonic (TNBS)-induced colitis. Percent body weight gain or loss was calculated based on the body weight recorded on the day of TNBS instillation (day 7). At day 9, average body weights (g) for each group were 511b, 508b, 489b and 463ª for the non-colitis control, B pullicaecorum, F prausnitzii and colitis control groups, respectively. The body weight loss was significantly lower in the B pullicaecorum and F prausnitzii treated groups compared with the colitis control group 48 h after TNBS administration. A different superscript indicates significant differences between the groups (p<0.05).

Figure 3

Bar graph showing the effect of Butyricicoccus pullicaecorum and Faecalibacterium prausnitzii on trinitrobenzenesulfonic (TNBS)-induced colitis, as reflected by (A) myeloperoxidase (MPO) activity, (B) tumour necrosis factor α (TNFα) and (C) interleukin (IL)-12 concentrations. Administration of TNBS increased the levels of MPO, TNFα and IL-12 significantly compared with the non-colitis control. Treatment with either B pullicaecorum or F prausnitzii had a significant inhibitory effect on the upregulation of cytokine production and the MPO activity in the colonic mucosa of colitic rats. Results are the mean±SEM from 10 rats per group. Means with no common superscript differ significantly (p<0.05).

Supernatant of B pullicaecorum protects epithelial cells from increase of intestinal permeability mediated by inflammation

Supernatant of B pullicaecorum culture was able to protect Caco-2 cells from the increase of intestinal permeability mediated by inflammatory stimulus, as TER was preserved 24 h after stimulation with TNFα and IFNγ (figure 4A). In non-inflamed conditions, supernatant of B pullicaecorum slightly (p=0.0282) increased the TER. Interestingly, Caco-2 cells exposed to B pullicaecorum as lyophilised bacteria diluted in Caco-2 medium were not protected. Note that the bacteria did not grow in aerobic conditions in Caco-2 medium, suggesting that the bacteria must be metabolically active to exert their beneficial effect. Since butyrate is found in supernatants of B pullicaecorum (10 mM), we hypothesised that butyrate might be the protective compound. Medium complemented with increasing concentrations of butyrate was able to confer similar protective effect, suggesting that butyrate produced by B pullicaecorum confers the beneficial effect. Under non-inflamed conditions, IL-8 was produced in low amounts and was slightly decreased by the supernatant or medium complemented with increasing concentrations of butyrate (figure 4B). When Caco-2 cells were stimulated with TNFα and IFNγ, cells produced large amounts of IL-8. This production was reduced (p=0.0106) when cells were coincubated with the supernatant of B pullicaecorum. Medium complemented with butyrate induced similar protection, though less efficiently. Bacteria alone induced a non-significant production of IL-8 in both conditions.

Figure 4

Caco-2 cells were grown on transwell filters and subsequently exposed on the apical side to Butyricicoccus pullicaecorum supernatant, B pullicaecorum cells and medium complemented with various butyrate concentrations. Simultaneously, in the basolateral side, cells were either left untreated (no stimulation) or stimulated with 100 ng/ml tumour necrosis factor α (TNFα) and 100 ng/ml interferon γ (IFNγ) to mimic colonic inflammation. (A) Changes in transepithelial resistance (TER) are expressed as percentage of original TER measured before stimulation. (B) Interleukin (IL)-8 production by Caco-2 cells expressed in pg/ml. ***p<0.0004; *p<0.02. NS, non-significant.


It is generally accepted that changes in the intestinal bacterial microbiota are associated with IBD. Evidence from several studies has highlighted alterations in abundance26 ,27 and diversity28 ,29 of the intestinal microbiota in patients with IBD compared with healthy individuals. Theoretically, introduction of probiotic bacteria, modulating the dysbiosis and aberrant immune response observed in patients with IBD, could result in clinical improvement. Therefore, the gut microbiota has emerged as a potential therapeutic target and a repository for novel drug discovery.30 Commercially available probiotics used to attenuate inflammatory activity and prevent relapses in UC are mostly Lactobacillus and Bifidobacterium species,31 selected for their low faecal levels in IBD and their ease of culture, storage, transport and stability within food products. In contrast, the probiotic potential of anaerobes which constitute the mass of the indigenous flora of the large intestine has been less explored.

Changes in the proportions of intestinal anaerobes belonging to clostridial clusters IV and XIVa have been suggested to play a role in the development of IBD.32 The relative loss of bacteria belonging to the clostridial clusters IV and XIVa is suggested to deplete gut fluids of key anti-inflammatory metabolites and other cell-associated immunomodulatory ligands.33 Many members of the microbiota metabolise fibre and resistant starch to butyrate which is the key energy source for the colonocytes,34 and possesses potent anti-inflammatory properties.35 A reduction in the number of these butyrate-producing bacteria can thus result in a degree of focal metabolic stress and vulnerability to inflammatory disease.36 The most abundant butyrate producers appear to be F prausnitzii, belonging to clostridial cluster IV, and Eubacterium rectale/Roseburia spp, which belong to clostridial cluster XIVa.37 F prausnitzii has been shown to be significantly underrepresented in stool samples15 and in surgically resected intestinal tissue7 from patients with IBD. The IBD phenotype could also be correlated with a lower level of the clostridial cluster IV genus Butyricicoccus.7 In the present study we showed that Butyricicoccus species were significantly less represented in the faecal microbiota from patients with UC and CD compared with healthy subjects. Moreover the Butyricicoccus genus was shown to be significantly decreased in patients with CD with clinically active disease compared with patients with CD with inactive disease. This finding is in agreement with the recently published study by Papa et al38 who generated a list of taxa specifically associated with each disease state (active IBD, remission samples, CD and UC) and showed Butyricicoccus as one of the taxa associated with remission. Butyricicoccus species belong to the same phylogenetic family (Ruminococcaceae) as F prausnitzii and are considered autochthonous microbes predominantly colonising the mucosa-associated mucus layer of the colon of mice and humans.39

Based on the correlation between the decreased intestinal abundance of the genus Butyricicoccus and the IBD phenotype, we suggest that a targeted increase of these potentially beneficial bacteria could reduce the incidence and severity of inflammation. We tested this hypothesis in a TNBS rat model using a strain from the Butyricicoccus genus B pullicaecorum, which had demonstrated desirable properties in vitro such as good adhesion capacity and tolerance to low pH (2) and bile (3.7 mM).40 Oral administration of B pullicaecorum had a preventive efficacy in controlling colitis in a TNBS model and was at least equivalent to the effect of F prausnitzii. Treatment with B pullicaecorum was shown ineffective in protecting against dextran sulphate sodium (DSS)-induced acute colitis (see online supplementary figure S1 and table S2). The reason for the lack of protection against DSS-induced inflammation is unclear. However, the mechanism by which B pullicaecorum exerts its protection in TNBS-induced colitis may reside in the effect of the bacterium on immune regulation. Indeed, dosing with B pullicaecorum decreased the production of pro-inflammatory cytokines in the TNBS model. B pullicaecorum is able to produce high concentrations of butyrate.41 Butyrate exerts a wide array of effects on intestinal function that can potentially contribute to disease therapy in IBD. Butyrate is known to modulate the transcription of numerous genes through its ability to inhibit histone deacetylase activity, resulting in hyperacetylation of histones and as a consequence suppression of nuclear factor (NF)-κB.10 NF-κB is a key transcription factor that controls the expression of genes encoding pro-inflammatory cytokines, chemokines, acute phase proteins, adhesion molecules and immune receptors.42 The increased mucosal levels of activated NF-κB in patients with UC are reduced by butyrate, an effect which correlates well with disease activity.11 Recently butyrate has been shown to antagonise the effect of metabolic stress in the colonocytes and to significantly reduce bacterial translocation.43 This inhibition of bacterial transcytosis across metabolically stressed epithelia is associated with reduced I-κB phosphorylation and hence NF-κB activation.43 In addition, NF-κB is inhibited by butyrate in mononuclear cells isolated from biopsies, and in vivo in the TNBS colitis models of IBD, resulting in decreased expression of TNFα.44 In IBD an initial upregulation of TNFα expression is observed during flair-ups, which increases the release of reactive oxygen species, resulting in DNA damage and as a consequence more activation of NF-κB and further TNFα, creating a perpetuating cycle. The decrease in TNFα expression in the colon of animals to which B pullicaecorum was administered could thus be a consequence of increased colonic butyrate levels. In addition to the effects on NF-κB, butyrate has been demonstrated to upregulate the anti-inflammatory peroxisome proliferator-activated receptor γ (PPARγ).45 In the colonic mucosa of healthy controls PPARγ protein expression is much higher compared with the non-inflamed mucosa of patients with UC.46 Butyrate thus inhibits inflammatory pathways and stimulates anti-inflammatory pathways.

Besides anti-inflammatory effects, butyrate has also been proposed to reinforce the colonic defence barrier, by increasing production of mucins and antimicrobial peptides.47 Indeed, the otherwise repressed expression of these goblet cell-secreted proteins was increased by butyric acid in a TNBS-induced rat model of colitis.48 More so, transglutaminase activity, involved in intestinal mucosal healing, correlates well with the severity of inflammation in colitis and was restored to normal levels in a rat model of colitis using butyrate.49 Butyric acid also decreases intestinal epithelial permeability, by increasing the expression of tight junction proteins50 and enhances proliferation of normal intestinal epithelial cells, thus leading to longer villi and an increased absorption capacity. Altered tight junction structures51 and abnormalities in mucus barrier functions leading to excessive bacterial translocation52 are characteristics of IBD that can thus be reverted by butyric acid. We showed that the supernatant including secreted metabolic components and butyrate from B pullicaecorum increased the TER, suggesting that metabolically active B pullicaecorum influence the epithelial barrier function. Although actions of butyrate are the most logical way to explain the effects observed in our study, there are multiple other possibilities, such as the production of other metabolites and the induction of a global shift in the microbiota composition.

Reconditioning of the gut microbiota through direct supplementation with beneficial bacteria or by indirect stimulation of colonisation and proliferation of beneficial bacteria could play a protective role in the inflammatory process. This is particularly the case for diseases in which the indigenous microbiota is altered due to the pathological condition, such as is the case in IBD. Although probiotic bacteria do not seem to be able to cure active disease, butyrate producers may be given to patients with IBD in times of remission to prevent disease relapse.

In conclusion, B pullicaecorum is a clostridial cluster IV species which seems to be a good candidate for use as a probiotic with anti-inflammatory potential. If a probiotic effect was confirmed in humans, future studies would be essential to unravel its mechanism of action and targets.


The authors would like to thank Professor H. Flint who kindly provided Faecalibacterium prausnitzii M21/2. We also thank Karolien Claes for running quantitative PCRs and Christian Puttevils for paraffin embedding and histological staining. Professor R Titball is acknowledged for critical review of the manuscript and improvement of language style.


View Abstract

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:


  • Contributors VE initiated the study. VE, SM, BF and MS were responsible for the animal study. KM and VE carried out quantitative PCR analysis. CP performed in vitro cell culture assays. CR performed the statistical analyses. BS, FH, RD, SV and FVI participated in the study design. Overall supervision of the research project was done by RD, SV and FVI. All authors were involved in the interpretation of results and the critical revision of the manuscript. All authors approved the final manuscript.

  • Funding This work was supported by the Institute of Science and Technology, Flanders (IWT), under contract no SBO-100016.

  • Competing interests V Eeckhaut, B Sas, R Ducatelle, F Haesebrouck and F Van Immerseel are listed as coinventors on a patent application for use of butyrate-producing bacterial strains related to Butyricicoccus pullicaecorum in the prevention and/or treatment of intestinal health problems (International Application Number PCT/EP2010/052184 and International Publication Number WO2010/094789 A1).

  • Ethics approval The Ethics Committee of the University of Leuven.

  • Patient consent Obtained.

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

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.

Linked Articles