SEARCH

SEARCH BY CITATION

Keywords:

  • ulcerative colitis;
  • pouchitis;
  • butyrate oxidation

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Background: Healthy colonic mucosa uses butyrate as the major energy source. In ulcerative colitis (UC) butyrate oxidation has been shown to be disturbed, but it remains unclear whether this is a primary defect. The aim of this study was to measure mucosal butyrate oxidation in UC (involved and noninvolved colon) and in pouchitis and to study the relationship with endoscopic as well as histological disease activity.

Methods: Butyrate oxidation was measured in 73 UC patients, 22 pouchitis patients, and 112 controls (95 colon, 17 ileum) by incubating biopsies with 1 mM 14C-labeled Na-butyrate and measuring the released 14CO2.

Results: Compared with that in normal colon, butyrate oxidation was significantly impaired in endoscopically active but not in quiescent disease or uninvolved colon segments. The severity of the metabolic defect was related to histological disease activity and decreased epithelial cell height. In active pouchitis, butyrate oxidation was significantly decreased compared with that in normal ileum and excluded pouches without inflammation. The histological pouchitis score correlated significantly with butyrate oxidation.

Conclusions: Active UC and pouchitis show the same inflammation-related metabolic defect. Our data suggest that the defect is a consequence of inflammation and that pouchitis is metabolically similar to active UC.

(Inflamm Bowel Dis 2008)

Short-chain fatty acids (SCFAs), mainly acetate, propionate, and butyrate, are produced in the colon by anaerobic bacterial fermentation of undigested dietary carbohydrates and have been attributed an important role in the maintenance of colonic homeostasis.1 Butyrate is considered the major energy source for colonic mucosa, providing up to 70% of the energetic needs.2 This energy is used for important intestinal processes like synthesis of mucus, absorption and secretion of electrolytes, and maturation and differentiation of the epithelium.1, 3 Moreover, butyrate seems to play a key role in the regulation of immunological mechanisms in the mucosa, including apoptosis4 and NF-κB activation,5 and also in the maintenance of the gut barrier.6 Butyrate has been shown to decrease proinflammatory cytokine expression via inhibition of NF-κB activation.7

Impaired metabolism of mucosal SCFAs, especially butyrate, may cause a state of energy deficiency and has been suggested to be implicated in the pathogenesis of ulcerative colitis (UC). UC is a mucosal disease of the large bowel predominantly affecting the left colon. Roediger was the first to describe defective colonic mucosal oxidation of butyrate in quiescent and active UC.8 Since then many studies in vitro (in isolated colonocytes and in biopsies)8, 9 and in vivo (animal models and human studies)10–13 have confirmed that, indeed, the utilization of butyrate by inflamed mucosa in UC is decreased, but it remains unclear whether this is a primary defect or an aspect of inflammation. However, unlike with active UC, impaired butyrate oxidation has not always been found in quiescent disease, arguing against a primary metabolic effect.9, 11, 13 Differences in results may be a result of methodological differences. In most studies the number of patients studied was rather limited, and the criteria for classification of disease activity were not standardized.

Furthermore, it is not clear whether the defect arises only in colonic mucosa or concerns an oxidation defect also involving the small bowel. Chapman et al measured butyrate oxidation in the terminal ileums of patients with quiescent UC and also found defective butyrate metabolism.14 This would explain the development of pouchitis in patients after restorative proctocolectomy for UC. However, Finnie et al did not find this disturbed butyrate metabolism in the ileum,9 and measurement of butyrate oxidation in pouch biopsies led to conflicting results.15, 16

The aim of the present study was to investigate the oxidation of butyrate in the colonic mucosa of a large cohort of patients with UC. The oxidation rates were carefully corrected for the protein content of the mucosal biopsies. The relationship between butyrate oxidation and the endoscopic and histological activity of the disease was studied as well as the differences in oxidation between involved and uninvolved colon. Finally, ileal pouches were studied before and after restoration of gut continuity.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Subjects

Butyrate oxidation of the colon was studied in a total of 73 UC patients (37 men, 36 women) with a median age of 36 years (range: 15–79 years) and in 95 controls (44 men, 51 women) with a median age of 59 years (range 14–91 years) who underwent colonoscopy for the investigation of irritable bowel syndrome or screening for adenomas. In 7 UC patients with active but limited disease, biopsies were taken in both the affected and unaffected parts of the colon. Most patients received drug treatment at the time of colonoscopy and biopsy: 21 patients received aminosalicylates or sulfasalazine (grouped together as ASAs) as the sole therapy, whereas others received steroids (n = 14), immunomodulators (n = 5), or both (n = 5) along with ASAs. Immunomodulators or steroids were received as the sole therapy by 5 patients, whereas 11 patients received a combination of both. One patient received infliximab combined with immunomodulators. Six patients were not receiving any gastrointestinal-related medication.

In 22 patients (9 men, 13 women), biopsies were obtained from pouches constructed for intractable UC. In 5 patients, biopsies were taken from excluded pouches before restoration of gut continuity. The median time since construction of the pouches was 6 years (range 0.6–14 years) for the pouches in transit and 9 weeks (range 4–9 weeks) for the excluded pouches. At the time of the pouchoscopy, 8 patients received antibiotics, 2 received steroids (combined with immunomodulators [n = 1]), and 2 patients received ASAs (combined with immunomodulators [n = 1]). Six patients received other gastrointestinal medication, whereas 5 patients were not receiving any medication. Biopsies of normal ileum were taken in 17 control patients (6 men, 11 women) without diagnosis of inflammatory bowel disease. Median age was 39 years (range 25–48 years) for the pouch group and 49 years (range 23–70 years) for the controls.

Written consent was obtained from all participating subjects, and the study was approved by the University Hospital Ethics Committee.

Measurement of Butyrate Oxidation

Our method to measure butyrate metabolism in biopsies was adapted from the technique described by Finnie et al.9 Biopsies (2 for each test condition) were taken during a routine endoscopic examination and were transported immediately to the laboratory in ice-cold, pregassed (95% O2/5% CO2) Krebs–Henseleit buffer (Sigma-Aldrich, St. Louis, Mo.) containing 11 mM glucose with a pH of 7.4. Each specimen was placed in a 10-mL conical incubation flask together with 1 mL of sterile Krebs–Henseleit buffer (pH 7.4) containing 11 mM glucose and 1 mM Na-butyrate. Two flasks without biopsies also underwent the procedure as blanks. After the addition of 1 μCi Na-[1-14C]butyrate (Sigma-Aldrich, St. Louis, Mo.), the flasks were gassed with 95% O2/5% CO2 for 30 seconds and sealed immediately with a gas-tight rubber stopper equipped with a polypropylene center well. The flasks were suspended in a shaking water bath (37°C, 120 oscillations/min) for a 2-hour incubation. The reaction was stopped by injecting 0.25 mL of 10% perchloric acid through the stopper into the incubation medium. To trap the released 14CO2, 0.3 mL of hyaminehydroxide was injected into the center wells. After an equilibration period of 90 minutes at 4°C, the center wells were transferred into scintillation counting vials together with 10 mL of scintillation liquid (Hionic Fluor, Packard, Groningen, The Netherlands). Radioactivity was assessed by a β liquid scintillation counter (Packard, model 2100, Downers Grove, Ill.).

Biopsies were blotted dry and frozen at −20°C. After lyophilization and weighing, the nitrogen content of the biopsies was determined by a continuous flow elemental analyzer (ANCA-2020, Europa Scientific, Crewe, UK). It was assumed that 0.16 g of nitrogen equaled 1 g of protein. Results are expressed as nanomoles of Na-butyrate metabolized per milligram of protein per hour.

Endoscopic Evaluation of Activity

The activity of ulcerative colitis was assessed using the Mayo Endoscopic Score proposed by Schroeder et al.17 A score of 0 or 1 was considered quiescent disease, a score of 2 mild disease, and a score of 3 active disease. Pouchitis was assessed using the endoscopic subscale of the Pouchitis Disease Activity Index.18 An endoscopic subscore of at least 4 on a table of 6 was diagnosed as active pouchitis.

Histological Assessment

Histological disease activity was assessed on biopsies taken at the same endoscopic examination. In UC, the grading scale for histological assessment of inflammation developed by Geboes et al was used.19 First, hematoxylin-eosin stained sections from biopsies were evaluated for architectural change (grade 0), chronic inflammatory infiltrate (grade 1), lamina propria neutrophils and eosinophils (grade 2), neutrophils in the epithelium (grade 3), crypt destruction (grade 4), and erosion or ulceration (grade 5). Patients with grades 0 to 2 were classified as having quiescent disease, with grade 3 as mild disease, and with grade 4 or 5 as active disease.

Second, the height of colonic surface absorptive epithelial cells was measured by conventional light microscopy using a calibrated lens. For each biopsy sample (4 to 5 biopsies per patients), cell height was measured in 4 or 5 representative areas, avoiding zones where cutting artifact was present, that is, where the surface epithelium was not sectioned perpendicularly. Mean values were calculated for each sample.

Pouch biopsies were scored from 0 to 6 using the histological criteria of the Pouchitis Disease Activity Index.20 This score takes into account polymorphonuclear leukocyte infiltration (absent = 0, mild = 1, moderate + crypt abscess <25% = 1, 25%–50% = 2, >50% = 3).

Statistics

Statistical analysis was performed with SPSS software (SPSS 15.0 for Windows; SPSS Inc., Chicago, Ill.). As the data were not normally distributed (Shapiro–Wilks), they are presented as medians and interquartile ranges. Data were compared by nonparametric tests (Kruskal–Wallis + Mann–Whitney U test with Bonferroni correction and Spearman correlation).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The control groups (colon and ileum) were significantly older than the UC and pouchitis groups (P < 0.001), respectively, but no relationship was found between age and butyrate oxidation.

Ulcerative Colitis

Butyrate Oxidation

Butyrate oxidation was statistically significantly decreased in UC patients compared with that in the control group (11.23 [6.41–20.30] versus 35.29 (26.34–44.94) nmol/mg protein · h; P < 0.0001). Subsequent division of the UC patients into disease activity subgroups using endoscopic criteria demonstrated a significantly decreased butyrate oxidation in mild (15.84 [8.67–26.87] nmol/mg protein · h) and active (7.40 [4.71–12.10] nmol/mg protein · h) disease compared with in the control group with normal colon (P < 0.0001 for both subgroups), as shown in Figure 1. For quiescent disease (25.94 [20.13–35.62] nmol/mg protein · h), the difference with the control colon group was not statistically significant. Furthermore, the group with active disease differed significantly from the groups with quiescent and mild disease (P < 0.0001 and P = 0.003, respectively), whereas the difference between quiescent and mild disease was not significant.

thumbnail image

Figure 1. Butyrate oxidation in control colon (n = 95) and in quiescent (n = 14), mild (n = 22), and active (n = 37) UC (horizontal bar, median).

Download figure to PowerPoint

In addition, UC patients were also categorized into disease activity subgroups based on histological criteria. Butyrate oxidation in 50 UC patients, in which disease activity score was based on histological criteria, was compared with that in the control colon group and is shown in Table 1. Again, butyrate oxidation was significantly decreased in active and mild disease compared with that in the control colon group (P < 0.0001 for both subgroups). In patients with quiescent disease, butyrate oxidation did not significantly differ from controls. Butyrate oxidation was also significantly decreased in the active disease group compared with that in the quiescent (P < 0.0001) and mild (P = 0.007) disease groups, and mild disease significantly differed from quiescent disease (P = 0.006).

Table 1. Butyrate Oxidation in UC Compared with Control Colon Groups
GroupButyrate Oxidation (nmol/mg protein · h)
  • Disease activity score based on histological criteria (median and IQR, α = 0.05).

  • Significantly different from

  • a

    control colon,

  • b

    quiescent UC, and

  • c

    mild disease.

Control colon (n = 95)35.29 (26.34–44.94)
Quiescent UC (n = 11)30.93 (25.94–37.19)
Mild UC (n = 9)13.90 (10.44–20.20)ab
Active UC (n = 30)6.60 (3.44–10.79)abc

Furthermore, a significant inverse relationship was found between butyrate oxidation and endoscopic evaluation and between butyrate oxidation and histological assessment of disease activity (Spearman r = −0.56, P < 0.0001; and r = −0.77, P < 0.001, respectively). A positive correlation was found between the endoscopic evaluation and the histological assessment of disease activity (Spearman r = 0.70, P < 0.0001).

Involved Versus Noninvolved Colon in Ulcerative Colitis

In 7 UC patients with mild and active disease, biopsies from the involved colon showed a significantly decreased butyrate oxidation compared with noninvolved (10.04 [8.90–13.51] versus 29.9 [25.92–54.42] nmol/mg protein · h; P = 0.009) and control (P < 0.0001) colon. There was no difference between noninvolved UC colon and control colon (Fig. 2).

thumbnail image

Figure 2. Intraindividual comparison of butyrate oxidation between noninvolved and involved UC colon (n = 7) and with control colon (n = 95); horizontal bar, median.

Download figure to PowerPoint

Epithelial Cell Height

Epithelial cell heights for the different patient groups based on endoscopic classification are listed in Table 2. Significantly decreased cell height was found in active UC patients compared with that in controls (P < 0.0001) and in those with mild (P = 0.003) and quiescent (P = 0.005) UC. Cell height in mild or quiescent UC did not differ from controls. When combining the patient groups, cell height significantly correlated with butyrate oxidation (Spearman r = 0.42, P < 0.0001; Fig. 3).

Table 2. Epithelial Cell Height in UC Compared with Control Colon Groups
GroupCell height (μm)
  • Disease activity score based on endoscopic criteria (median and IQR, α = 0.05).

  • a

    Significantly different from control colon (P < 0.0001), quiescent (P = 0.005), and mild (P = 0.003) disease.

Control colon (n = 47)26.7 (24.9–28.8)
Quiescent UC (n = 11)25.8 (23.7–26.7)
Mild UC (n = 12)26.9 (24.4–29.2)
Active UC (n = 27)21.0 (17.7–24.0)a
thumbnail image

Figure 3. Relation between butyrate oxidation and epithelial cell height (r, Spearman correlation coefficient).

Download figure to PowerPoint

Pouchitis

Butyrate oxidation in the mucosa of patients with endoscopically mild (12.80 [10.73–16.81] nmol/mg protein · h) or active (11.27 [6.56–16.31] nmol/mg protein · h) pouchitis was significantly decreased compared with that in normal ileum (37.17 [32.55–49.67] nmol/mg protein · h; P = 0.001 for both groups) and in excluded and normal pouches without inflammation (34.41 [32.79–45.11] nmol/mg protein · h; P = 0.005 for both groups; Fig. 4).

thumbnail image

Figure 4. Butyrate oxidation in control ileum (n = 17) and in excluded and normal pouches (n = 7) and mild (n = 9) and active (n = 6) pouchitis (horizontal bar, median).

Download figure to PowerPoint

A weak inverse relationship was found between histological pouchitis score and butyrate oxidation (Spearman r = −0.52, P = 0.013).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

There is evidence that substrate oxidation by colonic mucosa is impaired in ulcerative colitis. But it is not clear whether this abnormality is a primary defect or the result of inflammation. Previous studies measuring butyrate oxidation in UC have produced conflicting results. However, it is difficult to compare the results of different studies because of the use of different methodologies. In vitro experiments on biopsies,9, 21, 22 colonocytes,8, 23 and mucosal strips,24 as well as in vivo setups using a breath test after rectal instillation,11–13 have been used.

To elucidate if the decreased butyrate oxidation is a primary defect or an aspect of inflammation, it is interesting to focus on quiescent UC. Decreased oxidation in nonactive disease was only found by Roediger in isolated colonocytes8 and by Chapman et al in biopsy specimens.21 Attention needs to be paid to the criteria used to classify the disease as quiescent UC, especially if histological evaluation is included. Duffy et al separated UC patients into 2 groups based on histological assessment and found decreased butyrate oxidation in biopsies of patients with active UC but not in those with mild disease.22 As the same metabolic defect was found in the colons of patients with Crohn's colitis, it was concluded that this is only a reflection of the acute inflammatory response.

In the present study, butyrate oxidation in mucosal biopsies of a large patient group with ulcerative colitis was investigated to further elucidate whether the changes observed in the colonic mucosal metabolism of butyrate were secondary to the inflammatory process. The results have shown that the butyrate oxidation in quiescent UC biopsies was not statistically different from that in normal controls, independently of whether the patients were categorized using endoscopic or histological scoring. Furthermore, the observed reduction in butyrate oxidation seems to have been related to colonic disease activity. Although a significant positive correlation was found between the endoscopic evaluation and the histological assessment of disease activity, 39% of the UC patients were classified differently, indicating that histological changes may precede symptoms and/or that clinical and endoscopic remission may precede histological remission.

In a subgroup of UC patients, butyrate oxidation in diseased and noninvolved colon was compared intraindividually and with a control group. It was clearly shown that, contrary to in affected colons, butyrate oxidation in noninvolved UC colons did not differ from that in control colons.

The height of the colonic surface absorptive epithelial cells was also measured. In active disease, cell height was significantly decreased compared with that in control colon. This is an additional argument for decreased butyrate oxidation being a consequence of histological changes. Mariadason et al described that certain responses of Caco-2 cells to butyrate were dependent on the differentiation status of the colonic cells.25 In active UC, butyrate could be exposed to less mature cells because of changes in epithelial architecture.26–28

These observations, that is, decreased butyrate oxidation only in the presence of active disease, normal substrate utilization in unaffected segments, and the relationship with cell height all point toward a secondary origin of the defect.

Recently, Santhanam et al described a specific defect in the mitochondrial enzyme acetoacetyl CoA thiolase in patients with ulcerative colitis. This enzyme catalyses the important last step of butyrate oxidation and was shown to be significantly impaired in the colonic mucosal of UC patients.29 Furthermore, in another study, Thibault et al demonstrated that butyrate oxidation deficiency in intestinal inflammation is a consequence of reduced monocarboxylate transporter MCT1–mediated butyrate uptake by colonocytes.30 They hypothesized that butyrate oxidation may decline because of reduced intracellular butyrate availability. Both defects can explain the observed decrease in butyrate oxidation in ulcerative colitis and deserve further study.

Pouchitis is the inflammation of the ileal fecal reservoir after restorative proctocolectomy and seems to occur almost exclusively in patients operated on for ulcerative colitis.31 Although little is known about its pathogenesis, that pouchitis and extraintestinal manifestations of ulcerative colitis are strongly associated emphasizes the importance of genetic or phenotypic susceptibility, or both, for ulcerative colitis and pouchitis.32

In the present study, it was found that butyrate oxidation is decreased in mild and active pouchitis to the same degree as in UC (−79% for active UC and −68% for active pouchitis). This defect is probably secondary to the inflammation because butyrate oxidation in excluded pouches was normal and because we found a significant negative correlation between butyrate oxidation and histological activity score. The ileal metabolism of UC patients without pouches was not measured. In previous studies investigating the ileal mucosal biopsy samples in UC prior to operations, Finnie et al and Duffy et al found a similar metabolism of butyrate in controls and patients with ulcerative colitis, whereas Chapman et al found impaired butyrate oxidation compared with that in normal ileum and hypothesized that pouchitis is a manifestation of UC affecting ileal mucosa.9, 14, 16 Construction of an ileal fecal reservoir profoundly alters ileal luminal ecology, morphology of the epithelium, and availability of the mucosal metabolic substrate. Decreased butyrate oxidation was found in the ileal pouch mucosa of patients without active pouchitis, which was thought to be a consequence of changes in ecology and morphology.16 Moreover, the same authors found no metabolic difference in uninflamed ileum in Crohn's and UC, whereas butyrate oxidation in active UC as well as in Crohn's colitis was significantly decreased compared with that in control ileum and colon, indicating the absence of a panenteric abnormality.22

As the observed disturbance in oxidation in pouchitis was similar to that in UC, it is probably linked to evolution of the disease process. Together with the clinical and histological similarities of both diseases and the observation that pouchitis mainly develops in pouches constructed for UC, pouchitis may be considered a manifestation of UC in the ileum. As a fecal stream through the pouch is “necessary” to cause ileal inflammation, there is probably an important role for intestinal bacteria or their metabolites.31

In conclusion, our studies indicate there is an inflammation-related oxidation impairment in colonic mucosa in ulcerative colitis. This defect is not present in nondiseased colonic segments. The normal pouch mucosa metabolizes butyrate normally, but in the presence of pouchitis with colonic metaplasia, butyrate oxidation is impaired to the same extent as in active UC. This finding suggests that from a metabolic point of view, pouchitis and ulcerative colitis have similar characteristics.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES