Secondary Logo

Journal Logo

Butyrate and Type 1 Diabetes Mellitus: Can We Fix the Intestinal Leak?

Li, Nan*; Hatch, Marguerite; Wasserfall, Clive H; Douglas-Escobar, Martha*; Atkinson, Mark A; Schatz, Desmond A*; Neu, Josef*

Journal of Pediatric Gastroenterology and Nutrition: October 2010 - Volume 51 - Issue 4 - p 414–417
doi: 10.1097/MPG.0b013e3181dd913a
Original Articles: Gastroenterology

Objectives: An intestinal permeability defect precedes type 1 diabetes mellitus and may be a permissive factor in its pathogenesis. Butyrate strengthens the intestinal tight junctions. We hypothesized that enteral administration of sodium butyrate (NaB) in preweaned rats would result in differences in the development of diabetes associated with decreased inflammation and pancreatic β-cell destruction.

Materials and Methods: Using biobreeding diabetes-prone rat pups, oral NaB or saline was administered twice per day via micropipette from postnatal days 10 to 23. Rat pups were randomly assigned to 1 of 4 groups for the first experiment (control group, n = 7) and 3 different doses of butyrate groups (n = 8 for each group) and 2 groups for the second and third experiments (control n = 23; NaB at 400 mg · kg−1 · day−1, n = 20). Animals were studied into adulthood (up to day 140) for development of diabetes.

Results: The results showed that the survival rates were 28% versus 20% (butyrate vs control). No significant differences in survival were seen; however, there was a trend of delaying of onset of diabetes in the butyrate group. There were no differences of pancreatic histology score of islet inflammation between the 2 groups. Cytokine-induced neutrophil chemoattractant-1 was lower in the butyrate group at a dose of 400 mg · kg−1 · day−1 in the distal small intestine (P = 0.008) and in the liver (P = 0.01). There were no significant differences in the tracer flux measurements across the distal ileum and colon between the 2 animal groups.

Conclusions: Oral NaB given during the preweaning period did not significantly decrease the subsequent development of death from diabetes in biobreeding diabetes-prone rats.

*Department of Pediatrics, USA

Department of Pathology, University of Florida, Gainesville, USA.

Received 23 October, 2009

Accepted 16 February, 2010

Address correspondence and reprint requests to Josef Neu, MD, University of Florida, College of Medicine, Department of Pediatrics, PO Box 100296, Gainesville, FL 32610-0296 (e-mail:

The authors report no conflicts of interest.

The pathogenesis of type 1 diabetes mellitus (T1D) is characterized by the autoimmune destruction of insulin-secreting β cells, resulting in insulin deficiency and hyperglycemia in genetically susceptible individuals. Previous studies have suggested that an underlying defect in the intestinal interepithelial junctions, which results in increased permeability, may contribute to the pathogenesis of this disease (1–5). In healthy individuals, these intestinal interepithelial complexes provide a selective barrier between the external environment and the internal milieu of the body, allowing the absorption of nutrients and preventing the passage of potential antigens. Altering the permeability of this barrier by nutritional or pharmacological means may prevent the development of T1D.

Biobreeding diabetes-prone (BBDP) rats (Biomedical Research Models, Worcester, MA) are a well-studied model of T1D because at a predictable point in their life span (70–120 days), a large majority (>90%) develop T1D manifested as overt glycosuria, hyperglycemia, and ketosis (6,7). The majority of these animals die several days to weeks after onset of these symptoms. In the BBDP rat as well as the nonobese diabetic mouse, another rodent model of T1D, manipulation of the intestinal microbes with antibiotics (8,9) and probiotics (10) have resulted in differences in the development of diabetes. It has been suggested that the short-chain fatty acid (SCFA) milieu that results from microbial fermentation in the distal small intestine and large intestine may play a role in the pathogenesis of T1D (11). Most of the SCFAs have significant physiological properties, but 1 that appears to predominate is the 4-carbon fatty acid butyrate, which plays a role in proliferation, differentiation, apoptosis, energy metabolism, and maintenance of interepithelial tight junctions (12).

Because of the known effect of butyrate on intestinal junctions, we hypothesized that enteral administration of sodium butyrate (NaB) via gavage in preweaned rats would result in differences in the development of diabetes associated with decreased inflammation and pancreatic β-cell destruction.

Back to Top | Article Outline



All of the animal studies were approved by the Institutional Animal Care and Use Committee at the University of Florida. Pregnant BBDP rats were obtained from Biomedical Research Models and housed in the Animal Care Services facility, where they delivered litters of 6 to 10 pups. Three separate experiments were performed. The first was a dosage experiment that compared the effects of 3 doses of NaB—400, 600, and 800 mg · kg−1 · day−1—to a saline-administered control (n = 7, 8, 8, and 8). After results were obtained on this first experiment, we focused on 400 mg · kg−1 · day−1 versus control (n = 9, 12) in a second experiment and a third experiment (n = 11 and 11). The results from these 3 experiments done at separate times were initially analyzed separately and then pooled. The pups were fed by their mothers and received butyrate twice per day via micropipette gavage from days 10 to 23. The control animals received normal saline at the same volume. After day 23 they were weaned to a conventional rat-chow diet. Animals were followed into adulthood (until at least day 140) for clinical development of overt diabetes (decreased weight, polyuria, polydipsia, lethargy, and glucosuria) and survival. Additional animals from a third experiment (8 rats in control and 7 rats in the butyrate 400 mg · kg−1 · day−1 group) were sacrificed at age 66 to 80 days for intestinal permeability assay.

Back to Top | Article Outline

Pancreatic Histology

Staining was done using hematoxylin and eosin. Scoring of inflammation was performed in a blinded fashion by 3 different scorers with a score of 1 to 4, with 4 being the worst.

Back to Top | Article Outline

Intestinal Ex Vivo Permeability

As described previously (13), flat sheets of intestinal tissue were mounted in modified Ussing chambers with an exposed tissue area of 0.64 cm2. The sheets of distal ileum and distal colon were bathed on both sides by 10 mL of standard saline solution at 37°C that was vigorously circulated by bubbling with a gas mixture of 95% O2/5% CO2. The standard saline contained the following solutes (mmol/L): Na+ 139.4, K+ 5.4, Ca2+ 1.2, Mg2+ 1.2, Cl 123.2, HCO3 21.0, H2PO4 0.6, HPO2− 2.4, and glucose 10. Simultaneous measurements of the mucosal to serosal flux of mannitol and polyethylene glycol (PEG)-4000, nmol · cm−2 · h−1, were measured using D-[1–14C] mannitol (specific activity of 58 mCi/mmol or 2.15 GBq/mmol) purchased from Amersham Pharmacia Biotech (Buckinghamshire, UK) and 3H-PEG-4000 (specific activity of 1.5 mCi/g or 55.5MBq/g) purchased from Perkin Elmer Life and Analytical Sciences (Boston, MA) as paracellular permeability markers. The isolated tissue preparations were spiked with 1 μCurie of each tracer and the concentration of mannitol and PEG-4000 in the solution bathing the tissue was 0.5 mmol/L and 25 μmol/L, respectively. The magnitude of the mucosal to serosal unidirectional flux (Jms) was measured under short-circuit conditions at 15-minute intervals for a period of 45 minutes. In 1 series using proximal and distal ileum segments removed from BBDP rats, fluxes were similarly measured both before (period I) and after acute mucosal addition of 2 mmol/L butyrate (period II). The electrical parameters of the tissues were also recorded at 15-minute intervals throughout the entire experiment. Tissue conductance (GT, mS/cm2) was calculated as the ratio of the open-circuit potential (mV) to the short-circuit current (Isc, μA/cm2).

Back to Top | Article Outline

Cytokine Analysis

Samples of liver and small distal intestine were retrieved, and homogenates were used for protein assay and cytokine multiplex assay. Samples were homogenized on ice in a lysis buffer containing protease inhibitors. Homogenates were centrifuged twice for 30 minutes at 15,000 rpm and 4°C to remove tissue debris. Protein concentrations were determined using a modified Bradford-Lowry assay (BioRad Inc, Hercules, CA). Tissue extracts were placed into a commercially available TiterZyme EIA Rat GFO/CINC-1 kit from Assay Designs Inc (Ann Arbor, MI). Cytokine-induced neutrophil chemoattractant (CINC-1) and tumor necrosis factor-α levels were evaluated according to the manufacturer's instruction.

Back to Top | Article Outline

Statistical Analysis

Survival of the animals was analyzed using Kaplan-Meier analysis. Values are given as mean ± SD of duplicate measurements. t tests were performed using SigmaStat software (SPSS Inc, Chicago, IL), and differences among means were considered significant at P < 0.05. Statistical analysis of the flux data was performed using a paired or unpaired t test for the comparison of 2 means. Differences were considered significant if P ≤ 0.05.

Back to Top | Article Outline


Diabetes-free Survival

The first experiment involved treating the animals with NaB at 400 (low), 600 (medium), and 800 (high) mg · kg−1 · day−1 versus saline controls. The medium and high doses did not demonstrate trends toward differences in survival. However, the lowest dose, 400 mg · kg−1 · day−1, suggested a trend toward increased survival (Fig. 1). Based on the potentially encouraging results of this first group, we repeated the experiment twice using 400 mg · kg−1 · day−1.



As seen in Figure 2, this experiment was then repeated focusing on the 400 mg · kg−1 · day−1 dose. Amalgamating the results of the 3 experiments with 30 animals in the control group and 28 in the NaB mg · kg−1 · day−1 group, the survival rates were 28% versus 20% (butyrate vs control). No significant differences in survival were seen. However, there was a trend of delaying of onset of diabetes in the butyrate group. Blood glucose concentrations evaluated from the third experiment at 75 days of life were not different between the butyrate and control groups (data not shown).



Back to Top | Article Outline

Pancreatic Histology Score of Islet Inflammation

Eight butyrate-treated (400 mg · kg−1 · day−1) and 6 control animals' pancreata from the first experiment were evaluated. One third of the butyrate and control pancreata showed scores of 1, 3, and 4 with average scores at 2.5 ± 1.31 in the butyrate group and 2.67 ± 1.36 in the control group. There were no differences between the 2 groups.

Back to Top | Article Outline

CINC-1 in Distal Small Intestine

Rat CINC-1 was lower in the group treated with 400 mg · kg−1 · day−1 butyrate in the distal small intestine with P = 0.008 and in the liver with P = 0.01 (Fig. 3). There was no significant difference in levels of tumor necrosis factor-α in liver and distal small intestine (data not shown).



Back to Top | Article Outline

Flux Results

Using tracer fluxes of labeled PEG-4000 and mannitol, mucosal permeability was examined across the distal ileum and distal colon of butyrate-fed (400 mg · kg−1 · day−1) BBDP rats, compared with experimental BBDP controls at 66 to 80 days of life in the additional animal from the third experiment. As shown in Table 1 there were no significant differences in the tracer flux measurements across the distal ileum between the 2 animal groups. In addition, GT, which is the sum of cellular and paracellular conductances, was not altered by butyrate feeding. Similar flux results were observed in the distal colonic segment removed from both rat groups (Table 2). It is notable that compared with the small intestinal segment, the magnitude of the tracer fluxes across the distal colon was proportionately smaller because of the fact that this is a lower conductance tissue barrier. However, no alterations were observed in the fluxes of PEG and mannitol in the distal colon, which also suggests that the passive permeability of this tissue to these 2 markers is not affected in BBPD rats fed butyrate. Interestingly, in contrast to the distal ileum, butyrate treatment significantly reduced GT in the distal colon, indicating a change in barrier permeability. Additional studies are warranted to determine whether this is a consequence of a change in cellular versus paracellular conductance.





In another experimental series using segments of proximal and distal ileum, we examined the effects of acute addition of butyrate, 2 mmol/L, on the paracellular permeability of PEG-4000 and mannitol. The results were clear and showed that mucosally applied butyrate did not alter GT or the flux of either permeability marker (results not shown).

Back to Top | Article Outline


Our previous studies using the permeability markers lactulose and mannitol have verified that BBDP rats, before the onset of T1D, exhibit a highly permeable intestine associated with low levels of intestinal claudin, a major intercellular tight junction protein (1). Intestinal myeloperoxidase activities and goblet cell density are also higher in the diabetes-prone rats than in the controls, supporting the notion of an early intestinal inflammatory response. Our studies (1), as well as those from Meddings et al (2) and Graham et al (3), show that not only is enteropathy a consistent feature in the BBDP rat but it also precedes the onset of insulitis and appears to be due to mechanisms distinct from those that cause diabetes. More important, this is not a phenomenon that only occurs in rodent models of diabetes because recent studies have noted that humans with a propensity to develop T1D as well as other autoimmune diseases possess an abnormal intestinal barrier, the so-called leaky gut (1,2,14,15). Similar to our studies in BBDP rats, recent reports using intestinal samples from individuals at risk for or with already diagnosed T1D demonstrated abnormalities seen with sugar permeability tests (4,5,16) associated with interepithelial junctions on electron microscopy (5). These finding are entirely consistent with the concept that a leaky intestine in the setting of T1D would allow for greater exposure of the intestinal immune system to antigens.

There is evidence that the SCFA milieu of the intestine of diabetic patients is altered with a higher level of butyrate in individuals who do not develop diabetes (11). This opens the possibility of using a “postbiotic” such as butyrate, which is produced by bacterial fermentation in the prevention of diabetes.

Our study is the first to evaluate the provision of butyrate during the preweaning period. These experiments evaluated the effects of NaB on survival and outcomes that relate to the hypothesis that NaB will improve survival from diabetes in the BBDP rats via altered intestinal permeability. In terms of survival, the first experiment was suggestive of better survival in the animals treated with 400 mg · kg−1 · day−1 of NaB. We therefore focused on this dose in 2 subsequent experiments. These showed no difference in survival in the group treated with NaB, and the 3 experiments taken together show no difference in survival between the control and the group treated with 400 mg · kg−1 · day−1 NaB, despite a suggestion of a delay in the onset of diabetes in the butyrate-treated group. Of interest is a lower level of the inflammatory mediator CINC-1 in both the distal small intestine and the liver in the butyrate-treated rats. The histology scores for insulitis were not significantly different. In addition, although we could not detect changes in the fluxes of the permeability markers across intestinal tissues removed from butyrate-fed rats, GT was significantly reduced in the distal colon, indicating some alteration in barrier permeability of this segment. It is possible that PEG-4000 and mannitol are not good markers for this high-resistance tissue and flux studies using a smaller-size marker may be required to reconcile the reduced GT with a reduction in the passive permeability of this tissue.

Although these results could be interpreted as being negative results, there are several caveats. One is that the first experiment did show a trend toward improved survival with the NaB in the 400 mg · kg−1 · day−1 group. Whether this may be a batch-specific effect is unknown, but should be considered a possibility because we have seen differences in onset of diabetes and survival in BBDP rats depending on the batch of animals obtained at any one time. The timing of the NaB administration may also be important. In these experiments, the NaB administration was started before and during weaning. Previous studies by our group have determined that the permeability defect in these animals is evident approximately 50 days well before the onset of diabetes. It is not known whether a critical window occurs at that time. Our assumption was that the critical window of permeability would occur before this time, hence the reason for beginning the NaB administration at day 10 after birth.

The differences in CINC-1 are of interest in that higher levels of CINC-1 were seen in the control animals compared with the NaB-administered animals. These samples were obtained from the rats right after developing diabetes or diabetes-free rats at 140 days of age. The higher concentrations of the CINC-1 in the control animals suggest a modulatatory effect on this proinflammatory neutrophil chemoattractant.

Whether early improvement in the integrity of interepithelial junctions is involved was not evaluated here, but the fact that CINC-1 was decreased in both ileum and liver in the butyrate-treated animals suggests modulation of the inflammatory response. The mechanisms of this remain to be elucidated.

In summary, butyrate treatment shortly after birth through the time of weaning in BBDP rats did not significantly decrease the subsequent development of death from diabetes. It is possible that treatment at a later time during which the permeability defect is most marked remains to be evaluated.

Back to Top | Article Outline


1. Neu J, Reverte CM, Mackey AD, et al. Changes in intestinal morphology and permeability in the biobreeding rat before the onset of type 1 diabetes. J Pediatr Gastroenterol Nutr 2005; 40:589–595.
2. Meddings JB, Jarand J, Urbanski SJ, et al. Increased gastrointestinal permeability is an early lesion in the spontaneously diabetic BB rat. Am J Physiol 1999; 276(4 Pt 1):G951–G957.
3. Graham S, Courtois P, Malaisse WJ, et al. Enteropathy precedes type 1 diabetes in the BB rat. Gut 2004; 53:1437–1444.
4. Kuitunen M, Saukkonen T, Ilonen J, et al. Intestinal permeability to mannitol and lactulose in children with type 1 diabetes with the HLA-DQB1*02 allele. Autoimmunity 2002; 35:365–368.
5. Secondulfo M, Iafusco D, Carratù R, et al. Ultrastructural mucosal alterations and increased intestinal permeability in non-celiac, type I diabetic patients. Dig Liver Dis 2004; 36:35–45.
6. Mordes JP, Bortell R, Blankenhorn EP, et al. Rat models of type 1 diabetes: genetics, environment, and autoimmunity. ILAR J 2004; 45:278–291.
7. Greiner DL, Rossini AA, Mordes JP. Translating data from animal models into methods for preventing human autoimmune diabetes mellitus: caveat emptor and primum non nocere. Clin Immunol 2001; 100:134–143.
8. Brugman S, Klatter FA, Visser JT, et al. Antibiotic treatment partially protects against type 1 diabetes in the bio-breeding diabetes-prone rat. Is the gut flora involved in the development of type 1 diabetes? Diabetologia 2006; 49:2105–2108.
9. Schwartz RF, Neu J, Schatz D, et al. Antibiotic treatment partially protects against type 1 diabetes in the bio-breeding diabetes-prone rat. Is the gut flora involved in the development of type 1 diabetes? Diabetologia 2007; 50:220–221.
10. Calcinaro F, Dionisi S, Marinaro M, et al. Oral probiotic administration induces interleukin-10 production and prevents spontaneous autoimmune diabetes in the non-obese diabetic mouse. Diabetologia 2005; 48:1565–1575.
11. Samuelsson U, Ludvigsson J. The concentrations of short-chain fatty acids and other microflora-associated characteristics in faeces from children with newly diagnosed type 1 diabetes and control children and their family members. Diabet Med 2004; 21:64–67.
12. Roy CC, Kien CL, Bouthillier L, et al. Short-chain fatty acids: ready for prime time? Nutr Clin Pract 2006; 21:351–366.
13. Hatch M, Freel RW, Vaziri ND. Intestinal excretion of oxalate in chronic renal failure. JASN 1994; 5:1339–1343.
14. Mazmanian SK, Liu CH, Tzianabos AO, et al. An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system. Cell 2005; 122:107–118.
15. Zhang L, Li N, Caicedo R, et al. Alive and dead Lactobacillus rhamnosus GG decrease tumor necrosis factor-{alpha}-induced interleukin-8 production in Caco-2 cells. J Nutr 2005; 135:1752–1756.
16. Sapone A, de Magistris L, Pietzak M, et al. Zonulin upregulation is associated with increased gut permeability in subjects with type 1 diabetes and their relatives. Diabetes 2006; 55:1443–1449.

butyrate; intestinal inflammation; intestinal permeability; type 1 diabetes

Copyright 2010 by ESPGHAN and NASPGHAN