Oral Supplementation of Butyrate Reduces Mucositis and Intestinal Permeability Associated with 5-Fluorouracil Administration
Mucositis affects about 40 % of patients undergoing chemotherapy. Short chain fatty acids (SCFA), mainly butyrate, are claimed to improve mucosal integrity, reduce intestinal permeability and act as anti-inflammatory agents for the colon mucosa. We evaluated the effects of oral administration of SCFA or butyrate in the 5FU-induced mucositis. Mice received water, SCFA or butyrate during all experiment (10 days) and a single dose of 5FU (200 mg/kg) 3 days before euthanasia. We evaluated inflammatory and histological score by morphometry, and by activity of enzymes specific to neutrophil, eosinophil and macrophage and TLR-4, TNF-alpha and IL6 expressions. Intestinal permeability and tight junction protein ZO-1 expression were evaluated. Mice from the 5FU (5-Fluorouracil) group presented weight loss, ulcerations and inflammatory infiltration of neutrophils and eosinophils, increased expression of IL6 and TNF-alpha and increased intestinal permeability. SCFA minimized intestinal damage, reduced ulcerations without affecting intestinal permeability. Butyrate alone was more efficient at improving those parameters than in SCFA solution and also reduced intestinal permeability. The expression of pro-inflammatory cytokines and ZO-1 tended to be higher in the SCFA supplemented but not in the butyrate supplemented group. We showed the beneficial effects of butyrate on intestinal mucositis and its promising function as an adjuvant in the treatment of diseases not only of the colon, but also of the small intestine.
KeywordsMucositis Inflammation Short-chain fatty acids Butyrate 5-Fluorouracil Lipids
Monocyte chemotactic protein-1
Short-chain fatty acid
Short-chain fatty acids (SCFA) mainly acetate, propionate and butyrate are produced by bacterial fermentation, particularly of dietary fiber and carbohydrate in the large intestine. They are readily absorbed and are metabolized in the liver, producing energy. Butyrate, the 4-carbon fatty acid, has been tested as one of the therapeutic options for colon inflammatory diseases but was not previously tested for small intestine conditions, such as mucositis. Several in-vitro and in-vivo studies have shown that butyrate stimulates cell proliferation, inhibits inflammatory mediator production, reduces intestinal permeability and induces apoptosis in colon cancer cells [11, 19, 22, 28]. Moreover, SCFA and butyrate enemas have been considered in the therapy of ulcerative colitis [10, 15, 28].
Mucositis secondary to 5FU chemotherapy is related to alterations in intestinal permeability, causing bacterial translocation and changes in the intestinal immune status. Gut barrier disruption could be related to reduction of cell proliferation or an increase in apoptosis, both influenced by the presence of butyrate in the intracellular milieu. Intestinal paracellular permeability is regulated by the tight junctions  which consist of junctional protein complexes located in the apical portion of enterocytes and formed by transmembrane as well as peripheral membrane proteins such as occludin, claudin-1 and zonula occludens (ZO)-1.
Our group has previously shown the beneficial effect of a solution of SCFA (acetate, propionate and butyrate) in the clinical manifestations of ARA-C induced mucositis in mice . In that study, the mechanism of such an action and the specific fatty acid responsible for this effect was not investigated. Due to its metabolic relevance on cell metabolism butyrate is the more promising candidate for such an effect.
Considering the significant clinical impact of mucositis and the beneficial effects of the SCFA solution on the small intestinal mucosa, we aimed to characterize the effects of oral administration of either SCFA or butyrate solutions on the amelioration of intestinal mucosa after 5-fluorouracil (5FU) administration.
Materials and Methods
Experimental groups of Swiss mice receiving water or experimental solutions and chow diet during 10 days
Treatment (single dose)
C2 (35 mM) + C3(15 mM) + C4 (9 mM)b
Butyrate (9 mM)
5FU (200 mg/kg)
5FU + SCFA
C2 (35 mM) + C3(15 mM) + C4 (9 mM)
5FU (200 mg/kg)
Butyrate (9 mM)
5FU (200 mg/kg)
The animals were subdivided into 6 groups: 1, control group (n = 12): no mucositis induction and receiving water; 2, 5FU group: with mucositis and receiving water; 3, SCFA group (n = 4): no mucositis induction and receiving SCFA solution; 4, 5FU + SCFA group (n = 18): with mucositis and receiving SCFA solution, 5, Butyrate group(n = 18): no mucositis induction and receiving butyrate solution and 6, 5FU + butyrate group (n = 18): with mucositis and receiving butyrate solution.
On the 10th experimental day, all animals were anesthetized and euthanized for blood and organ collection. Body weight, liquid intake and food intake were measured on the 1st and 10th experimental days. The energy intake was calculated, taking into consideration calories from food and solutions. The protocol was approved by the Animal Care Committee of Universidade Federal de Minas Gerais (UFMG), CETEA # 46/2008.
The small intestine and colon were removed from the pylorus to the ileocecal valve and from the cecum to the rectum, respectively. The organs were washed, gently perfused with PBS and measured with an inextensible millimeter ruler. The intestine was divided into the duodenum, jejunum and ileum and then fixed in paraformaldehyde (4 %) for 15 min. The segments were opened along their longitudinal axes, fixed in Bouin's solution for 6 h, embedded in paraffin and cut into 5 µm thick sections before being stained with hematoxylin-eosin (H&E). Images were obtained using a JVC TK-1270/RGB microcamera and the KS300 software built into a Kontron Eletronick/Carl Zeiss image analyzer. Ten fields from H&E sections were randomly chosen for villi height measurement.
Alterations of the mucosal architecture (general structure, cell distribution, mucosa and submucosa aspect), ulcerations, inflammation, villus height and inflammatory infiltration were used to determine the histological score. The samples were coded and then scored by a trained pathologist. The score ranged from zero (no alteration) to 3 (severe alteration) according to Soares et al. . The results are presented as the sum of the score obtained for each parameter.
Neutrophil, Macrophage and Eosinophil infiltrations were evaluated by analyzing the enzyme activity of myeloperoxidase (MPO), N-acetylglucosaminidase (NAG) and eosinophil peroxidase (EPO), respectively. Samples were homogenized and centrifuged, and precipitates were used for quantification of enzyme activities as previously described . Briefly, precipitates were dissolved in HETAB 0.5 % (Sigma-Aldrich®, USA) in phosphate buffer and centrifuged.
For EPO quantification, 75 µL of supernatant was added to 75 µL of OPD (Sigma-Aldrich®, USA), diluted in Tris–HCl and H2O2 and incubated at 37 °C for 30 min. The reaction was stopped by adding 50 µL H2SO4 before being read at wavelength 492 nm in a microplate spectrophotometer (TermoPlate, Brazil).
For MPO quantification, 25 µL of supernatant was added to 25 µL of TMB in DMSO (Sigma-Aldrich®, USA). After addition of 100 µL H2O2, the solution was incubated at 37 °C for 5 min. The reaction was stopped by adding H2SO4 before being read at 450 nm in a microplate spectrophotometer (TermoPlate, Brazil).
For NAG quantification, precipitates were dissolved in 0.1 % Triton X-100 (Sigma-Aldrich®, USA) and centrifuged before the supernatant was added to p-nitrophenyl-N-acetyl-β-d-glucosamine solution in citrate/phosphate. After incubation, the reaction was stopped by the addition of glycine buffer and read at 400 nm in a microplate spectrophotometer (TermoPlate, Brazil). Results were expressed in arbitrary units (based on absorbance) by 100 mg of tissue.
Study of Intestinal Permeability
We studied intestinal permeability using 99mTcdiethyleneaminopentacetic acid (DTPA). Because DTPA is not absorbed by the healthy small intestine, its presence in the blood is proportional to the increase in the damage of intestinal paracellular permeability. Animals were gavaged with 0.1 mL of DTPA labeled with 3.7 MBq 99mtechnetium in the form of sodium pertechnetate (Na99mTcO4), obtained by a 99molybdenum/99mtechnetium generator (IPEN/CNEMA, São Paulo, Brazil) . Four hours after gavage, animals were anesthetized and exsanguinated by the axillary plexus. The radioactivity of the standard dose and blood samples was determined in an automatic pit scintillator (ANSR, Abbott®, USA), and the percentage of recovered dose in each animal was calculated as follows: radioactivity of blood/radioactivity of standard dose × 100. The average of the control group values was used as a reference. The results were presented as the increase seen in experimental groups over the control group (expressed as a percent).
The total RNA from the ileum was extracted using the TRIzol® reagent according to the manufacturer’s protocol. The reverse transcription was performed using 2 µg of the total RNA, 200 U of the reverse transcriptase, 2.5 µL of the 5× RT buffer, 1.8 µL of the 10 mM dNTPs, 0.2 µL of the 10,000 U/mL RNasin, and 1.0 µL of the 50 µM oligo dT. The temperature settings for this reaction were 70 °C for 5 min, on ice for 2 min, 42 °C for 60 min, 70 °C for 15 min and 4 °C for the final step. The resulting cDNA was used for real-time PCR as described below. The specific primers were designed using the Primer Express software and synthesized by IDT. Real time PCR was carried out on a StepOne sequence detection system (Applied Biosystems) using the Power SYBR Green PCR Master Mix (Applied Biosystems). The dissociation curve indicated that only one product was obtained in each reaction. The relative levels of gene expression were determined using the ΔΔCycle threshold method as described by the manufacturer, in which data for each sample is normalized to the β-actin expression. The PCR results were analyzed with the SDS 2.1 software (Applied Biosystems), and the amount of mRNA of each gene of interest was normalized to the amount of the murine β-actin gene. mRNA expression levels were calculated as the fold difference relative to the housekeeping gene: relative expression = 2−(CT [target gene] − CT [β-actin-1]).
The sequences of the primers used are as follows:
TNFα: 5′CGTCGTAGCAAACCACCAAG3′ and 5′GAGATAGCAAATCGGCTGACG3′
IL6: 5′ACAACCACGGCCTTCCCTACTT-3′ and 5′CACGATTTCCCAGAGAACATGTG3′
TLR4: 5′TGACAGGAAACCCTATCCAGAGTT3′ and 5′TCTCCACAGCCACCAGATTCT3′
ZO-1: 5′CCAGCTTATGAAAGGGTTGTTC3′ and 5′TCCTCTCTTGCCAACTTTTCTC3′
β-actin: 5′CTGCCTGACCAAGTC3′ and 5′CAAGAAGGAAGGCTGGAAAGG A3′
Statistical analysis was performed using the Graph Pad Prism 7.0® software (San Diego CA). The results were tested for outliers (Grubbs’ test) and normality using the Kolmogorov–Smirnov test. The one-way ANOVA and the Newman–Keuls multiple comparison post-test were used for all parameters except for architecture alteration and villus height (non-parametric distribution), which were instead analyzed by the Kruskal–Wallis and Dunn’s post-test. To compare gene expression of a specific group with the control group, an unpaired t test or a Mann–Whitney test was used. A significant difference was defined as p ≤ 0.05.
Initially we compared the effect of the three control groups (control, SCFA and butyrate groups) on weight gain and intestinal morphology. The results showed that the three groups presented the same weight evolution and intestinal mucosa aspect (data not shown), demonstrating that SCFA and butyrate administrations do not interfere with intestinal mucosa integrity. For this reason, we omitted data from SCFA and butyrate control groups, presenting only the results of control mice (without SCFA or butyrate supplementation).
Energy and Hydric Intake and Ponderal Evolution
Mucosa improvement was confirmed by the worse histological score for 5FU-treated animals compared to the control animals (Fig. 2b–d). When each intestinal segment was analyzed separately, the duodenum was moderately affected by 5FU treatment, while the inflammation score was intense for both the jejunum and ileum of mice from the 5FU group. Nonetheless, after SCFA and butyrate treatment, the histological score was reduced in all intestinal segments (Fig. 2b, c).
Leukocyte Infiltration and Cytokine Expression
Amplification of mRNA of Pro-Inflammatory Molecules
Gene expression of TLR-4, IL-6 and TNF-α was analyzed by RT-PCR. The TLR4 expression presented a strong tendency (p = 0.06) to be higher in 5FU mice compared to controls but it was not different between supplemented groups (Fig. 5a). Regarding TNF-α and IL-6 expressions, a strong tendency (p = 0.057) of increased values was seen in 5FU and 5FU + SCFA, but not in 5FU + butyrate (Fig. 5b).
Since paracellular permeability is linked to tight junction protein interactions, ZO-1 protein expression was also investigated, also showing a strong tendency to increased expression in 5FU (p = 0.056) and 5FU + SCFA (p = 0.09) groups. Once again, 5FU + butyrate mice presented ZO-1 expression closer to control ones.
In the current study, we have shown that SCFA and, in particular, butyrate are effective in improving mucosa integrity and reducing inflammation in an experimental model of mucositis. Moreover, these effects were obtained by the oral supplementation of fatty acids rather than rectal via as is usually done. In a previous study, we showed that SCFA solution improves intestinal manifestation of mucositis . However, we did not evaluate the effects of butyrate used alone and limited our analyses to the histological aspects of the intestinal mucosa. As far as we know, the current study is the first one to address the effect of butyrate supplementation on the evolution of chemotherapy-induced intestinal mucositis.
Role of SCFA and Butyrate on the Intestinal Integrity
Although the SCFA and butyrate solutions had beneficial effects on mucositis, the butyrate solution was the most effective. While it contained the same concentration of butyrate as the butyrate-alone solution, the SCFA solution only partially reduced weight loss and did not prevent permeability alteration or reduced the pro-inflammatory cytokine expression in the intestine. We believe that, rather than presenting deleterious effects, the presence of acetate and propionate in the SCFA solution reduced the intestinal absorption of butyrate via the border brush transporter. The transport of butyrate through monocarboxylate transporters (MCT) is saturable, coupled with H+ and inhibited by several monocarboxylates, such as acetate, propionate, pyruvate, l-lactate and α-ketobutyrate . We hypothesize that the higher concentration of acetate (35 mM) and propionate (15 mM) in contrast to the lower concentration of butyrate (9 mM) in the SCFA solution could compete with the butyrate transport by MCT and/or another intestinal transporter, resulting in a smaller absorption and, consequently, a smaller post absorptive effect of butyrate in the SCFA solution compared to the butyrate alone.
The mucosa analyses showed that neutrophil infiltration, as measured by MPO activity had increased in 5FU-treated animals and that SCFA or butyrate solutions did not interfere in this aspect. Moreover, macrophage infiltration, as measured by NAG activity, was not different in any 5FU-treated groups compared to the control group. This latter result could be explained by the pattern of cell migration towards the inflammatory site. Neutrophils are the first cells arriving at the inflammatory site causing an increase in MPO activity in the first hours after administration of 5FU as seen in the current study [3, 24]. In contrast, macrophages are effectors cells, most frequently seen in later inflammatory stages, mainly after 3 days of inflammatory stimulus . Our mice were euthanized 3 days after the administration of 5FU, which may explain the lack of differences in macrophage concentration between the control and 5FU groups. Regarding eosinophils, we found an intense infiltration related to mucositis that was prevented by SCFA and butyrate supplementation. Although the role of eosinophils in mucositis is seldom studied, the detection of eosinophils in the intestinal mucosa of patients with inflammatory intestinal diseases, even in small quantities, has been associated with adverse clinical consequences, such as weight loss, malabsorption and shortening of large intestine crypts [21, 31]. Thus, a reduction of these cells in the groups treated with SCFA and butyrate is considered a sign of a better prognosis.
The expression of the pro-inflammatory molecules TLR4, IL6 and TNFα tended to be higher in the 5FU group. This increase was possibly due to the inflammation caused by 5FU itself and due to the rupture of the intestinal barrier permitting bacterial translocation and LPS-induced TLR4 activation. In concordance to the permeability data and MPO activity, 5FU + butyrate group kept cytokine expression closer to the control group. These data suggest that the better intestinal trophism seen in this group attenuates the inflammatory stimulus secondary to LPS and bacterial translocations, reducing activation of mucosal immune cells.
The absence of differences between the 5FU and butyrate groups for NAG and MPO activities as well as inflammatory score suggest that a trophic rather than immunologic effect is the main mechanism for both butyrate and SCFA protection. Nonetheless, butyrate demonstrated a more intense effect on the intestinal barrier that could be due to its metabolic effect as an energetic source associated to its action on gene expression. The results of butyrate on cytokines and TLR4 expressions compared to SCFA are in agreement with the improvement of intestinal permeability which will reduce bacterial translocation and LPS induced immune response activation.
Cell infiltration and intestinal permeability are important markers of tissue damage and mucosal inflammation [2, 17]. We have observed that butyrate reduced the alteration in intestinal permeability that is typically seen in 5FU mice. Our results are in agreement with other in-vitro studies showing the improvement of intestinal permeability with butyrate solutions [14, 19, 27]. The mechanism of this effect can be linked to gene expression of tight junction proteins, since butyrate and trichostatin A, an inhibitor of histone acetylase decreased tight junction permeability in Caco-2 cells via lipoxynase activation .
ZO-1 is a TJ protein that interacts with occludin, ZO-2, ZO-3 and actin, reducing intestinal permeability and inducing cell differentiation. Since 5FU treatment induced the increase of intestinal permeability, one would expect a reduction in ZO-1 protein in tight junctions [13, 16, 23]. However, our results showed that ZO-1 expression was increased in groups which presented an increase in permeability (5FU and 5FU + SCFA) compared to the control. We believe that this higher expression of ZO-1 reflects the more intense mucosal repair after 5FU-induced damage. ZO-1 protein and expression are generally tested under conditions of continuous inflammatory stimuli such as cell incubation with pro-inflammatory agents  or animal models with chronic inflammatory diseases [13, 16]. This continuous inflammatory stimulus maintains ZO-1 expression down-regulated in those models, avoiding the repair of the intestinal barrier. However, in our mucositis model, 5FU that has a short half-life  was given as a single dose 3 days before euthanasia of the mice. Since intestinal mucosa is renewed every 3–4 days, we believe that new intestinal cells formed after 5FU injection overexpressed ZO-1 in order to compensate the important intestinal barrier disruption. In the 5FU + butyrate group, mucosal damage was lower (as seen by the reduction in intestinal permeability and mucosa recovery) compared to the 5FU non-supplemented group, reducing the necessity of a compensatory ZO-1 expression. Probably, ZO-1 expression, intestinal permeability in the 5FU + butyrate group was closer to the control group, reinforcing our hypothesis. Moreover, ZO-1 altered expression in the 5-FU and 5FU + SCFA groups can be related to the ZO-1 translocation from the cell boundary (tight junction location) to the cytoplasm as previously described 
Possible Mechanisms of Action
The mechanisms of action of SCFA and butyrate on intestinal cells are not totally understood. Oral administration of SCFA exposes the stomach and small intestine mucosa to these fatty acids before reaching the colon [25, 29] and are transported to the liver . There, they can be metabolized to glutamate, glutamine and acetoacetate  important fuels for enterocytes [1, 5, 6]. Butyrate also increases the pancreatic secretion and the activity of jejunal brush-border enzymes , increasing availability of nutrients for enterocyte regeneration, stimulates of Glucagon-like peptide-2 (GLP-2) , a pleiotropic intestinotrophic hormone that enhances digestive and absorptive capacity . All these components propitiate the mucosa integrity, protecting cells from 5FU damage, including the increase in intestinal permeability. As a consequence, the bacterial translocation is reduced, minimizing the inflammatory response.
In conclusion, the results presented here highlight, for the first time, the potential use of butyrate in inflammatory diseases of the small intestine, such as mucositis. Oral administration of butyrate contributes to rebuilding the intestinal mucosa by quickly repairing ulcerated and inflamed tissue.
This study was supported by grants from the Conselho Nacional de Desenvolvimento Científico e Tecnologico (CNPq); Coordenação de Aperfeiçoamento de Nível Superior (CAPES), Fundação de Amparo a Pesquisa de Minas Gerais (FAPEMIG) e Pro-Reitoria de Pesquisa (PRPq) da UFMG. The authors are grateful to Maria Helena Alves, for taking care of the animals.
Conflict of interest
The authors declare that they have no conflict of interest.
- 4.Chu E, Sartorelli AC (2012) Cancer chemotherapy. In: Katzung BG, Masters SB, Trevor AJ (eds) Basic and clinical pharmacology, chap 54, 12th edn. McGraw-Hill, New YorkGoogle Scholar
- 7.Generoso SV, Viana M, Santos R, Martins FS, Machado JA, Arantes RM, Nicoli JR, Correia MI, Cardoso VN (2010) Saccharomyces cerevisiae strain UFMG 905 protects against bacterial translocation, preserves gut barrier integrity and stimulates the immune system in a murine intestinal obstruction model. Arch Microbiol 192:477–484PubMedCrossRefGoogle Scholar
- 13.Hudcovic T, Kolinska J, Klepetar J, Stepankova R, Rezanka T, Srutkova D, Schwarzer M, Erban V, Du Z, Wells JM, Hrncir T, Tlaskalova-Hogenova H, Kozakova H (2012) Protective effect of Clostridium tyrobutyricum in acute dextran sodium sulphate-induced colitis: differential regulation of tumour necrosis factor-alpha and interleukin-18 in BALB/c and severe combined immunodeficiency mice. Clin Exp Immunol 167:356–365PubMedCrossRefGoogle Scholar
- 16.Martinez C, Vicario M, Ramos L, Lobo B, Mosquera JL, Alonso C, Sanchez A, Guilarte M, Antolin M, de Torres I, Gonzalez-Castro AM, Pigrau M, Saperas E, Azpiroz F, Santos J (2012) The jejunum of diarrhea-predominant irritable bowel syndrome shows molecular alterations in the tight junction signaling pathway that are associated with mucosal pathobiology and clinical manifestations. Am J GastroenterolGoogle Scholar
- 17.Melo ML, Brito GA, Soares RC, Carvalho SB, Silva JV, Soares PM, Vale ML, Souza MH, Cunha FQ, Ribeiro RA (2008) Role of cytokines (TNF-alpha, IL-1beta and KC) in the pathogenesis of CPT-11-induced intestinal mucositis in mice: effect of pentoxifylline and thalidomide. Cancer Chemother Pharmacol 61(5):775–784PubMedCrossRefGoogle Scholar
- 28.Vieira EL, Leonel AJ, Sad AP, Beltrao NR, Costa TF, Ferreira TM, Gomes-Santos AC, Faria AM, Peluzio MC, Cara DC, Alvarez-Leite JI (2012) Oral administration of sodium butyrate attenuates inflammation and mucosal lesion in experimental acute ulcerative colitis. J Nutr Biochem (in press)Google Scholar