Gut MucosalenzymeactivityMucosalenzymeactivityforbutyrate oxidation Butyric acid CH3.CH2.CH2.COOH...

8
Gut 1996; 38: 886-893 Mucosal enzyme activity for butyrate oxidation; no defect in patients with ulcerative colitis E S Allan, S Winter, A M Light, A Allan Abstract Background-Butyrate is an important energy source for the colon and its metab- olism has been reported to be defective in ulcerative colitis. One mechanism for defective butyrate metabolism in patients with ulcerative colitis could be an enzyme deficiency in the P-oxidation pathway of butyrate. Aims-This study was undertaken to measure the activity of each enzyme involved in the 3-oxidation pathway of butyrate in colonic epithelium. Patients-Patients with ulcerative colitis (n=33), Crohn's colitis (n= 10), and control subjects with colorectal cancer or diver- ticular disease (n=73) were studied. Methods--Analysis was carried out using fluorometric and spectrophotometric tech- niques on homogenised epithelial biopsy specimens. Results-Significantly increased butyryl CoA dehydrogenase activity was found in mucosa from patients with ulcerative colitis (33.2 (28.3, 38.1) pumolIg wet weightl min:mean (95% CI)) compared with activity in mucosa from control patients (24.3 (20.9, 27.7) gmollg wet weightl min:mean (950/o CI)) p<0.02. No signifi- cant increase in activity of the enzymes butyryl-CoA synthetase, crotonase or hydroxybutyryl-CoA dehydrogenase was found in patients with ulcerative colitis. In contrast the mucosal thiolase activity was significantly lower in those patients with quiescent colitis (3.21 (2.61, 3.81) pmollg wet weightlmin:mean (95% CI)) when compared with control mucosa (5.69 (5.09, 6.29) pumollg wet weightlmin:mean (95% CI)) p<O0OOl. However, mucosal thiolase activity increases with the age of the donor patient and differences in the age range of the patient groups probably account for this finding. Conclusions-This study shows no sub- stantial deficiency of enzyme activity in the P-oxidation pathway of butyrate in the mucosa of patients with ulcerative colitis in histological remission. (Gut 1996; 38: 886-893) Keywords: short chain fatty acids, ulcerative colitis. There is evidence that a metabolic defect within colonic epithelial cells could contribute to the pathogenesis of ulcerative colitis. Colonic epithelial cells absorb luminal butyrate and oxidise it to provide energy through the citric acid cycle. l1 It was proposed by Roediger that metabolism of luminal butyrate may be abnormal in patients with ulcerative colitis and that this abnormality may result in energy deficiency at the cellular level,5 thus contributing to the colitis. By using in vitro metabolic studies, Roediger showed that butyrate was the fuel of choice for colonic epithelial cells compared with alter- native fuels such as glucose or glutamine.6 Roediger also found that when colonic cells were studied from patients with ulcerative colitis these cells oxidised butyrate as an energy source to a lesser extent than cells from control tissues.5 Furthermore the inability of colonic epithelial cells to utilise butyrate occurred irrespective of the activity of the colitis, sug- gesting that the abnormality might represent a primary defect in the mucosa and not be sec- ondary to the inflammation in the colonic tissue. Ireland and Jewell7 and Chapman et al4 confirmed that utilisation of butyrate is decreased in colonic cells from patients with quiescent ulcerative colitis. But Chapman et al were unable to confirm an earlier finding by Roediger2 that butyrate metabolism was greater in the descending colon compared with the ascending colon.8 Both groups found that 5-aminosalicylic acid had no effect on butyrate metabolism in colonic epithelial cells from normal controls.7 9 In contrast Finnie et al,3 and Clausen et al 10 were unable to demonstrate any defect of butyrate metabolism in mucosal biopsy specimens from patients with quiescent colitis compared with mucosa from normal controls. These authors suggested that the discrepancy with previous studies may have arisen either from the consequences of the epithelial cell isolation technique or because of the use of potentially toxic concentrations of butyrate.3 It is possible that abnormal butyrate oxida- tion may occur in the colonic epithelial cells of patients with quiescent ulcerative colitis. Such an abnormality may be of pathogenetic signifi- cance and therefore requires further study. This study set out to measure the activity of each enzyme involved in oxidation of butyrate (Fig 1). Methods Tissue samples Prior to assay all tissue samples were coded so that the biochemist performing the assays was unaware of the disease group of the donor patient. It was not possible to assay all enzymes in the oxidation pathway in each Departments of Biochemistry E S Allan S Winter Histopathology A M Light and Surgery, Good Hope Hospital Trust, West Midlands A Allan Correspondence to: M.r A Allan, Department of Surgery, Good Hope Hospital Trust, West Midlands B75 7RR. Accepted for publication 13 December 1995 886 on May 31, 2021 by guest. Protected by copyright. http://gut.bmj.com/ Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. Downloaded from

Transcript of Gut MucosalenzymeactivityMucosalenzymeactivityforbutyrate oxidation Butyric acid CH3.CH2.CH2.COOH...

  • Gut 1996; 38: 886-893

    Mucosal enzyme activity for butyrate oxidation;no defect in patients with ulcerative colitis

    E S Allan, S Winter, A M Light, A Allan

    AbstractBackground-Butyrate is an importantenergy source for the colon and its metab-olism has been reported to be defective inulcerative colitis. One mechanism fordefective butyrate metabolism in patientswith ulcerative colitis could be an enzymedeficiency in the P-oxidation pathway ofbutyrate.Aims-This study was undertaken tomeasure the activity of each enzymeinvolved in the 3-oxidation pathway ofbutyrate in colonic epithelium.Patients-Patients with ulcerative colitis(n=33), Crohn's colitis (n= 10), and controlsubjects with colorectal cancer or diver-ticular disease (n=73) were studied.Methods--Analysis was carried out usingfluorometric and spectrophotometric tech-niques on homogenised epithelial biopsyspecimens.Results-Significantly increased butyrylCoA dehydrogenase activity was found inmucosa from patients with ulcerativecolitis (33.2 (28.3, 38.1) pumolIg wet weightlmin:mean (95% CI)) compared withactivity in mucosa from control patients(24.3 (20.9, 27.7) gmollg wet weightlmin:mean (950/o CI)) p

  • Mucosal enzyme activity for butyrate oxidation

    Butyric acid CH3.CH2.CH2.COOH

    ATP

    AMP

    Butyryl CoA

    Crotonyl CoA

    Hydroxybutyryl Cl

    NA

    NAI

    Acetoacetyl CoA

    Acetyl CoA

    CH3.CH2.CH2.CO-S.CoA

    H2

    CH3.CH.CH.CO-S.CoA

    H20 -

    oA CH3.CHOH.CH2.CO-S.CoA

    D+

    DH

    CH3.CO.CH2.CO-S.CoA

    Butyryl.CoA.synthetase

    EC: 6.2.1.2.

    Butyryl.CoA.dehydrogenase

    EC: 1.3.99.2.

    Crotonase

    EC: 4.2.1.17.

    Hydroxybutyryl.CoA.dehydrogenase

    EC: 1.1. 1.35.

    Acetoacetyl.CoA.thiolase

    EC: 2.3.1.9.

    CH3.CO-S.CoA + CH3.CO-S.CoA

    Figure 1: Pathway of /3-oxidation of butyrate.

    mucosal specimen. Therefore as manyenzymes as possible were assayed dependingon the amount of mucosa available forhomogenisation from each patient. The effectof age, sex, or site of mucosa along the colon,or method of harvesting was assessed usingnormal mucosa.

    Sample preparationSamples of colorectal mucosa were transportedto the laboratory in ice cold saline (0. 15 MNaCl). Samples were examined and excessmoisture removed by blotting. Samples not forimmediate analysis were snap frozen and storedat -40°C. Lamina propria lymphocytes wereisolated and stored according to the method ofSachdev et al. l 1 Lymphocytes were thawedslowly at room temperature and enzymesassessed similarly to whole tissue samples.Mucosa was removed from the bowel wall by

    scraping with the edge of a clean microscopeslide and then weighed. Biopsy specimens wereweighed and disrupted in entirety. Previouslyisolated lamina propria lymphocytes wereweighed and then homogenised. All cells werehomogenised in enzyme extraction buffers at aratio of 100 mg/ml (wet weight/volume) byusing an Omni 5000 hand held homogeniser at40C.

    SubjectsMucosal specimens were taken from 73 subjectsundergoing investigation or treatment for colo-rectal cancer (n= 69) or diverticular disease(n=4). The median age of these patients was 67

    (range 28-9 1) years and 39 were male. Allmucosal specimens in these control patientswere taken from a site at least 10 cm from anymacroscopic abnormality and adjacent mucosalsamples were histologically normal. Sixty ofthese mucosal specimens were taken at laparo-tomy and 13 at rigid sigmoidoscopy.

    Mucosal specimens were taken from 33subjects with ulcerative proctocolitis, medianage 46 (range 18-83) years, 18 were men.Adjacent mucosal specimens were taken in eachcase for histological examination and assess-ment of severity of ulcerative colitis. Mucosalspecimens were taken from 13 patients atlaparotomy and 20 at rigid sigmoidoscopy.

    Mucosal specimens were taken from 10subjects with Crohn's colitis, median age 34(21-67) years, two were men. All of thesemucosal specimens were taken at laparotomy.All of these patients had coexisting small bowelCrohn's disease. The diagnosis of ulcerative orCrohn's colitis was established using conven-tional clinical, radiological, endoscopic, andhistological criteria.

    Enzyme extractionButyryl-CoA dehydrogenase and hydroxybu-

    tyryl-CoA dehydrogenase - mucosal homogenateswere extracted into 50 mM phosphate buffer(pH=7.0) containing 0.3 mM EDTA, 200 uMphenylmethylsulphonyl fluoride, and 2 mMdithiothreitol. Crude homogenate was reservedfor hydroxybutyryl-CoA dehydrogenase assay.Homogenate for butyryl-CoA dehydrogenaseassay was fractionated by addition of solidammonium sulphate to 40% w/v with mixingfor 30 minutes on ice and then centrifugation at5000xg at 4°C for 15 minutes. The supernatantwas assayed.

    Crotonase, acetoacetyl-CoA thiolase, andbutyryl-CoA synthetase were extracted into 10mM TRIS buffer (pH= 7.4) containing 1 mMEDTA, 3 mM magnesium chloride, and 300mM sucrose.

    Cytological assessment of cell suspensionsBefore complete disruption by homogenisa-tion, aliquots of mucosal cell suspensions fromthe normal colon of three patients were spreadon glass slides and sprayed with polyethyleneglycol, ethyl alcohol, and glacial acetic acidfixative and stained with Papanicolaou stain.The proportions of lymphocytes and epithelialcells in these preparations were then estimatedmicroscopically for 10 high power (X400)fields.

    Classification of mucosal samples from patientswith ulcerative proctocolitisRoutine haematoxylin and eosin stained paraf-fin wax sections were prepared from samples ofmucosa adjacent to those used for assay.Inflammatory changes were classified usingthe criteria of Truelove and Richards12(Appendix). Epithelial cell density wasassessed using the grading system described byGoodman et al. 13

    887

    on May 31, 2021 by guest. P

    rotected by copyright.http://gut.bm

    j.com/

    Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. D

    ownloaded from

    http://gut.bmj.com/

  • Allan, Winter, Light, Allan

    TABLE I Enzyme activity in normal biopsy specimens.Enzyme is expressed in ,umol/g wet weight/minute. Numbersofsamples and 95% confidence limits are also shown

    Enzyme Men Women

    Butyryl-CoA synthetaseMean 0-058 0.049Number 4 995% CI -0-062 to 0-178 0-019 to 0 079

    Butyryl-CoA dehydrogenaseMean 21-97 25-54Number 6 1295% CI 14-77 to 29-2 20-7 to 30-38

    CrotonaseMean 9-26 9-25Number 16 1995% CI 7-2 to 11-3 6-52 to 11-98

    Hydroxybutyryl-CoA dehydrogenaseMean 4-24 4.59Number 34 3095% CI 3.79 to 4-69 4-1 to 5-08

    ThiolaseMean 5-96 5-42Number 34 3195% CI 5-27 to 6-65 4-4 to 6-44

    Enzyme assaysAll chemicals and enzymes were purchasedfrom Sigma Chemical Company UK. Assayswere performed either spectrophotometricallyusing a Philips PU8720 spectrophotometer orfluorimetically using a Shimadzu RF-5001PCfluorimeter.

    All enzyme assays were optimised for humancolonic epithelial cells.

    Butyryl-CoA synthetase (BuCoAs) wasassayed according to the method of Lageweget al.'4 This enzyme catalyses the activation ofbutyrate to its CoA ester. The reaction wasassessed by coupling the reaction to acyl-CoAoxidase and peroxidase. The rate of peroxideformation was assessed fluorimetrically. Theintra-assay coefficient of variation was 9-6%.Buyryl-CoA dehydrogenase (BuCoADH) was

    assayed according to the method of Engel.'5BuCoADH catalyses the conversion ofbutyryl-CoA to crotonyl-CoA. The reaction

    Butyryl-CoA Synthetase

    TABLE ii Enzyme activity in ,umol/wet weight/min forcontrol subjects in samples collectedfrom rectal biopsy andsurgical specimens from rectum, left colon, and right colon

    Biopsy Surgical specimenspecimen

    Enzyme (rectum) Rectum Left colon Right colon

    Synthetase 0-0783 0.0375 0-0251BuCoA DH 23-41 32-24 21-50Crotonase 8-58 7.90 8-45 13-26HoBuCoADH 4-19 4-53 4-16 4-71Thiolase 4.13* 6-14 5.44 6-75Protein 3-71 4-22 3-62 3-53

    Statistical analysis of loglo transformed data shows no significantdifferences from synthetase, butyryl-CoA dehydrogenase,crotonase and hydroxybutyryl-CoA dehydrogenase. Thiolaseactivity, however, showed significant differences (*p=0.046)between rectal biopsy samples and surgical samples. Allsubsequent statistical analyses of thiolase activity wereperforned on like samples.

    was assessed by coupling this reaction todichlorophenolindophenol and phenazinemethosulphate. The rate of decreasingabsorbance was assessed spectrophotometri-cally at 600 nM.

    Crotonase was assayed according to themethod of Fong and Schultz.16 Crotonasecatalyses the conversion of crotonyl-CoA tohydroxybutyryl-CoA. The formation of themagnesium complex was assessed spectro-photometrically at 280 nm. The intra-assayand inter-assay coefficients of variation were4% and 13% respectively.

    Hydroxybutyryl-CoA dehydrogenase (Hobu-CoADH) activity was measured accordingto the method of Fong and Schulz.'7HobuCoADH catalyses the conversion ofhydroxybutyryl-CoA to acetoacetyl-CoA.However, in vitro the forward reaction rapidlyequilibrates thus the reaction was assayed inthe reverse direction. The rate ofreaction result-ing from nicotinamide adenine dinucleotide(NAD) formation was assessed spectrophoto-metrically at 340 nm. The intra-assay and inter-

    Butyryl-CoA dehydrogenase Crotonase130.20 r

    0.15 e-

    60

    40 F-

    0-10 |-

    30 r

    20 F-

    30K

    * 80

    * O

    Hydroxybutyryl-CoAdehydrogenase

    20 H

    10

    _4h O13

    0

    a 4+* |

    12

    10 e-

    o

    -I

    8

    itiII

    Thiolase

    18 r15

    a

    £1

    12

    9

    6

    3

    0

    +-e-c

    C,C

    Figure 2: Comparison ofenzyme activities between control subjects and ulcerative colitis patients classified as Truelove andRichards class 1. Control values are shown as O and uleerative colitis samples as *. The mean enzyme activity is shownby horizontal line.

    Ea)CD

    3._

    0)

    3.10E

    0

    a)ENcwL

    0.05

    0.00

    1211109876543210

    888

    .

    a12

    .

    on May 31, 2021 by guest. P

    rotected by copyright.http://gut.bm

    j.com/

    Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. D

    ownloaded from

    http://gut.bmj.com/

  • Mucosal enzyme activity for butyrate oxidation

    TABLE III Variation ofenzyme activity with age for control and ukerative colitis samples

    30-50 51-70 71-90

    Enzyme Control UC Control UC Control UC

    Synthetase 0.03 0-032 0-052 0-034 0-067 0-021BuCoA DH 29-31 28-84 27-25 24-31 18-91 25-45Crotonase 6-63 8-95 10-76 11-37 7.97 7-63HoBuCoA DH 4-51 3-85 4-33 3-13 4-66 2-96Thiolase 3-43 3-69 5-74 2-37 6-12 5-84

    assay coefficients of variation were 8.2% and16.7% respectively.Acetoacetyl-CoA thiolase (Thiolase) activity

    was assayed according to the method ofWilliamson et al.18 Thiolase catalyses theconversion of acetoacetyl-CoA to twoacetyl-CoA units. The rate of reaction due toacetyl-CoA formation was assessed spectro-photometrically at 303 nm. The intra-assaycoefficient of variation was 4%.

    Protein concentration was determined usingthe method of Bradford.'9 Protein concen-trations in control samples, ulcerative colitis,and Crohn's samples were compared. Nodifferences were seen between normal controlsamples and samples from patients with ulcer-ative colitis. Therefore no adjustment forprotein concentration was made. In the caseof Crohn's samples differences in protein

    100°

    10

    a)

    zC.).)0)

    0)

    a)

    E

    0.1 [_

    0.011 2 3

    Truelove and Richards classFigure 3: Comparison ofenzyme activity between Truelove and Richards classes. Butyryl-CoA synthetase (M), butyryl-CoA dehydrogenase (Ol), crotonase (M), hydroxybutyryl-CoA dehydrogenase (E), thiolase (u).

    concentration with the control samples wereobserved and therefore all results wereexpressed per gram of protein.

    Statistical methodsComparisons in enzyme activity betweengroups of patients were assessed using anunpaired t test with two tail p values on logl0transformations of the enzyme data. Otherstatistical tests are as indicated in the text.

    EthicsApproval for the study was given by theNorth Birmingham Research and EthicsCommittees.

    Results

    Enzyme activity in histologically normal mucosaTable I summarises enzyme activity in normalbiopsy specimens. No differences in enzymeactivity were found between men and womenfor any of the enzymes tested (two samplet test). One way analysis of variance showedno site related difference in enzyme activitybetween samples from the rectum, left or rightcolon. There was increased thiolase activitywith age when patients were subdivided intothree groups: those below 50 years, thosebetween 50 and 70 years, and those over 70years of age. Statistical analysis (one wayanalysis of variance) on logl0 transformed datashowed significant differences in enzymeactivity between groups (F=4-83: p=0 01).No other enzyme activity varied significantlywith age. Thiolase activity also varied signifi-cantly between rectal biopsy and laparotomysamples with the biopsy specimens givinglower values (two sample t test p=0 008).Table II gives these results. All furthercomparisons were stratified into biopsy andrectal excision samples. No other enzymeactivity varied with the mode of collection.

    ProteinFifty four specimens were assayed for proteincontent. The mean (950/O CI) protein concen-tration was 3.79 (3.52, 4.06) g/g wet weighttissue. No differences were seen across the agerange of patients tested (45-91 years).Comparative analysis by site of origin ofsample showed no differences between rectum,left colon or right colon (Table II).

    Enzyme activity in ulcerative colitis mucosaButyryl-CoA synthetase - 18 patients donated

    biopsy specimens, eight were male and 10 werefemale. The median age was 55 years and theage range spanned 20 to 80.Buyl-CoA dehydrogenase - 24 patients

    donated biopsy samples, 15 were male andnine were female. The median age was 39years and the age range spanned 18 to 83.

    Crotonase - 20 patients donated biopsyspecimens, 13 were male and seven female.

    889

    1

    on May 31, 2021 by guest. P

    rotected by copyright.http://gut.bm

    j.com/

    Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. D

    ownloaded from

    http://gut.bmj.com/

  • Allan, Winter, Light, Allan

    The median age was 38 years and the age rangespanned from 18 to 83.

    Hydroxybutyryl-CoA dehydrogenase - 24patients donated biopsy specimens, 13 weremale and 11 female. The median age was 39years and the age range spanned 18 to 83.

    Thiolase - 20 patients donated biopsysamples, 13 were male and seven were female.The median age was 38 years and the age rangespanned 18 to 72.

    Comparison ofenzyme activity in ulcerativecolitic mucosa with histologically normal mucosaFor each enzyme studied a comparison wasmade between the enzyme activity in controlbiopsy specimens and the enzyme in Trueloveand Richards (T and R) class 1 biopsy speci-mens (two sample t test on loglo transformeddata). The mean (95% CI) butyryl CoAdehydrogenase activity (pumol/g wet weight!min) in normal mucosa was 24.3 (20.9, 27.7)n=18 compared with 33.2 (28.3, 38.1) n=10in biopsy specimens of T and R class 1(p=002). The mean thiolase activity was 5-69(5.09, 6.29) n=65 in normal mucosacompared with 3.21 (2.61, 3.81) n=9 insamples ofT and R class 1 (p=0.001). Therewas no significant difference in butyryl-CoA

    100 r

    synthetase, crotonase, or hydroxybutyryl-CoAdehydrogenase activity between control and Tand R class 1 biopsy specimens (Fig 2).The mean thiolase activity in T and R class 1

    biopsy specimens was significantly lower thanthe mean thiolase activity in control samples.Within theT and R class 1 biopsy specimens themedian age of the donor patients was 37 (range28-62) years compared with a median age of 67(range 28-85) years in patients with normalmucosa. In view of this the youngest ninecontrol patients: median age 49 (range 28-52)years were selected from the control group andthiolase activity in these biopsy specimens wascompared with the thiolase activity in the T andR class 1 samples. No significant difference wasfound (two sample t test on loglo transformeddata p=06). Table III shows these data.

    Differences in enzyme activity in Truelove andRichard classesEnzyme activity was compared between eachof the three Truelove and Richards classes(one way analysis of variance on loglo trans-formed data). Butyryl-CoA dehydrogenaseactivity was significantly different between Tand R classes. Hydroxybutyryl-CoA dehydro-genase activity was also significantly differentbetween T and R classes with enzyme activityfalling with increasing inflammation in theadjacent mucosal samples (Fig 3).

    Correlation of enzyme activity with epithelial celldensityBiopsy specimens from patients with ulcerativecolitis were classified according to their epithe-lial cell density. There was a trend towardsdecreasing enzyme activity with decreasingcell density, but between the grades 1, 2, 3there was no significant difference (Fig 4). Nospecimens were classified as class 4.

    Correlation of enzyme activity with drugs takenby patientThe mean enzyme activity in ulcerative colitispatients taking either oral or intravenous corti-costeroids (n= 8) was compared with theenzyme activity in those patients taking neithertopical nor systemic corticosteroids (n=9). Anunpaired t test on the log1o transformed datashowed no significant difference in any enzymeactivity.A similar analysis comparing the enzyme

    activity in patients taking any 5-aminosalicyliccontaining drug orally or rectally (n= 17) withthe enzyme activity in those patients takingno 5-aminosalicylic containing drug eitherorally or rectally (n= 5) showed no significantdifferences.

    Enzyme activity in the mucosa ofpatients with3 colonic Crohn 's disease

    Butyyl-CoA dehydrogenase - samples fromIcell nine patients were assayed for butyryl-CoAcse ( dehydrogenase activity. The mean activity was

    1-33 (0-83, 1.83) p.mol/g protein/min.

    10 i

    nccC.)

    ._

    0)

    0

    c

    0E=L

    0.1 r-

    0.01 IH

    I I0 2

    Epithelial cell densityFigure 4: Comparison ofenzyme activity in tissue classified according to epithelialdensity. ButyIyl-CoA synthetase (E), butyryl-CoA dehydrogenase (Ol), crotonahydroxybutyryl-CoA dehydrogenase (s), thiolase (m). H=hypertrophied crypt.

    F=EL-iF==]EL-3==1a__AF=:=1

    890

    1

    on May 31, 2021 by guest. P

    rotected by copyright.http://gut.bm

    j.com/

    Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. D

    ownloaded from

    http://gut.bmj.com/

  • Mucosal enzyme activity for butyrate oxidation

    1.4r

    1.2 F-

    'E 0.8c

    0

    0.

    0 0.6

    _-

    0.4 _

    0.2

    0 WOWOI0%1A I,,,,5Y% Q,/,%ABucoA DH *Crotonase *HoBucoA DH Thiolase

    Figure 5: Comparison ofenzyme activity in Crohn 's and control tissue. The enzymeactivity is expressed in 1ttmollg protein/minute. Crohn 's disease (E), control (W).*p

  • Allan, Winter, Light, Allan

    E

    ._

    0

    0

    E

    4

    2

    BucoA DH Crotonase HoBucoA DH Thiolase

    Figure 6: Enzyme activities in lamina propria lymphocytes from control (E), ulcerativecolitis ( D), and Crohn 's disease (0) tissues. Enzyme activity is expressed as ,umol/g wetweight tissue/min.

    certainly due to the increased age of the controlgroup, with correspondingly increased enzymeactivity, compared with the younger age of thecolitic group. Furthermore thiolase activity isrequired to catalyse the final step in the oxida-tion of butyrate and it is probable that butyryl-CoA synthetase controls the rate limiting stepof the oxidation pathway.

    This study therefore shows a rise in enzymeactivity in minimally inflamed mucosa frompatients with ulcerative colitis compared withnormal mucosa.There was a tendency for all enzyme activity

    studied to fall in parallel with falling epithelialcell density in the mucosa of patients withulcerative colitis. This type of relation waspreviously described as a property of thecolonic epithelial cell enzyme glucosaminesynthetase.13 This is probably a generalisedresponse to epithelial cell depletion and may beshared by many colonic epithelial cell enzymesin patients with ulcerative colitis.

    Previous studies show that 5-aminosalicylicacid has no effect on butyrate metabolism incolonic epithelial cells from normal controls.This study complements this finding bysuggesting that neither corticosteroid nor5-aminosalicylic acid treatment have anyeffect on the enzyme activity of the oxidationpathway of butyrate.

    Enzyme activity in colonic mucosa frompatients with Crohn's colitis was raised com-pared with that from the mucosa of controlsubjects. This rise in activity was significant forthe enzymes butyryl-CoA dehydrogenase andthiolase. This rise in activity mirrors theincreased activity seen in Truelove andRichards class 1 mucosa from patients withulcerative colitis. It probably occurred becauseof selection of the least inflamed tissue in thecolon for assay. Previous studies of the enzymeglucosamine synthetase13 report a rise in theactivity of this enzyme in the mucosa ofpatients with Crohn's colitis even at sites dis-tant from the macroscopically evident disease.The rise in enzyme activity found in mucosafrom patients with Crohn's colitis is probablynot caused by an increased cell turnover in theCrohn's tissue because cell proliferation in rec-tal biopsy specimens from patients with rectalCrohn's disease is normal irrespective of thedegree of inflammation in the rectum.2' Itseems more probable that the disease processtriggers a generalised reactive increase in theactivity of many intracellular enzymes. Thisfinding suggests that the observed rise inglucosamine synthetase activity in patientswith Crohn's colitis is part of a more wide-spread response to inflammation in the colonicepithelial cell.

    This study used biopsy specimens and notepithelial cell preparations to study colonicepithelial cell enzyme activity and thereforeconsideration must be given to sources ofenzyme activity within the samples but notarising from the colonic epithelial cells them-selves. Although colonic wall muscle does notmetabolise butyrate3 the mucosal homogenateused for enzyme assay contained a mixtureof both colonic epithelial cells and laminaproprial lymphocytes. Cytological assessmentof these cell mixtures suggested that laminalproprial lymphocytes accounted for only 10%of the cells in the mixture. Further studies onisolated laminal proprial lymphocytes showeda small but detectable enzyme activity fromthese cells. This activity was similar irrespec-tive of whether the cells were from controlsubjects or patients with inflammatory boweldisease. They are therefore likely to have madea minimal but relatively constant contributionto the total enzyme activity in the homogenatesirrespective of the source of the specimens.

    In conclusion, this study suggests that ifabnormalities of butyrate metabolism occur inthe colonic epithelium of patients with ulcera-tive colitis then the mechanism of this abnor-mality does not entail decreased enzymeactivity in the 3-oxidation pathway forbutyrate. None the less an abnormality ofbutyrate metabolism in patients with ulcera-tive colitis remains possible. It seems unlikelythat supply of butyrate to colonic epithelialcells is ever in danger because studies ofcolonic luminal bacterial flora in patients withulcerative colitis show few changes.22 In addi-tion absorption of butyrate in the remissionphase of the disease is no different fromabsorption in control subjects.23 However,other abnormalities of intracellular butyrate

    892

    on May 31, 2021 by guest. P

    rotected by copyright.http://gut.bm

    j.com/

    Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. D

    ownloaded from

    http://gut.bmj.com/

  • Mucosal enzyme activity for butyrate oxidation 893

    metabolism may occur such as abnormalintracellular transport of butyric acid, alter-ation of mitochondrial oxidation or abnormalincorporation of butyrate products into thecitric acid cycle for energy production.This study was supported by a grant from the West MidlandsRegional Health Authority and the Sutton Municipal Charities.The isolated lamina proprial lymphocytes were a generous giftfrom Dr D P Jewell, The Radcliffe Infirmary, Oxford. Dr P JHanson, Department of Pharmaceutical Sciences, University ofAston gave invaluable advice on aspects of the enzymes assays.Mrs J H Plant typed the manuscript.

    1 Cummings JH, Pomare EW, Branch WJ, Naylor CPE,MacFarlene GT. Short chain fatty acids in the humanlarge intestine, portal, hepatic and venous blood. Gut1978; 28: 1221-7.

    2 Roediger WEW. Role of anaerobic bacteria in the metabolicwelfare of the colonic mucosa in man. Gut 1980; 21:793-8.

    3 Finnie IA, Taylor BA, Rhodes JM. Ileal and colonic epithe-lial metabolism in quiescent ulcerative colitis: increasedglutamine metabolism in distal colon but no defect inbutyrate metabolism. Gut 1993; 34: 1552-8.

    4 Chapman MAS, Grahn MF, Boyle MA, Hutton M, RogersJ, Williams NS. Butyrate oxidation is impaired in thecolonic mucosa of sufferers of quiescent ulcerative colitis.Gut 1994; 35: 73-6.

    5 Roediger WEW. The colonic epithelium in ulcerative coli-tis: an energy deficiency disease? Lancet 1980; ii: 712-5.

    6 Roediger WEW. The utilisation of nutrients by isolatedepithelial cells of the rat colon. Gastroenterology 1989; 83:424-9.

    7 Ireland A, Jewell DP. 5-aminosalicylic acid (5-ASA) has noeffect on butyrate metabolism in human colonic epithelialcells. Gastroenterology 1989; 102: A176.

    8 Chapman MAS, Grahn MF, Giamondo P, O'Connell PR,Onwu D, Hutton M, et al. A new technique to measuremucosal metabolism and its use to map substrate utilisa-tion in the healthy human large bowel. BrJ'Surg 1993; 80:445-9.

    9 Chapman MA, Grahn MF, Williams NS. 5-aminosalicylicacid increases colonic mucosal oxidation of glucose[Abstract]. BrJSurg 1993; 80: 648.

    10 Clausen MR, Mortensen PB. Kinetic studies of coloncytemetabolism of short chain fatty acids and glucose inpatients with ulcerative colitis. Gut 1995; 37: 685-9.

    11 Sachdev GK, Dalton HR, Hoang P, Dipaglo MC, Crotty B,Jewell DP. Human colonic intraepithelial lymphocytessuppress in-vitro immunoglobin synthesis by autologousperipheral blood lymphocytes and lamina propria lym-phocytes. Gut 1993; 34: 257-63.

    12 Truelove SC, Richards WCD. Biopsy studies in ulcerativecolitis. BMJ 1956; 1: 1315-8.

    13 Goodman MJ, Kent PW, Truelove SC. Gluco-saminesynthetase activity of the colonic mucosa in ulcera-tive colitis and Crohn's disease. Gut 1977; 18: 219-28.

    14 Lageweg W, Steen I, Tager JM, Wanders RJA. A fluorimet-ric assay for Acyl-CoA synthetase activities. AnalytBiochem 1991; 197: 384-8.

    15 Engel P. Butyryl-CoA dehydrogenase from Megasphaeraelsdenii, Methods Enzymol 1981; 71: 359-66.

    16 Fong JC, Schulz H. Purification and properties of pig heartcrotonase and the presence of short chain and long chainenoyl Coenzyme A Hydratases in pig and guinea pigtissues. 7 Biol Chem 1977; 252: 542-7.

    17 Fong JC, Schulz H. On the rate - determining step of fattyacid oxidation in heart. Inhibition of fatty acid oxidationby 4-pentanoic acid. J Biol Chem 1978; 253: 6917-22.

    18 Williamson DH, Bates MW, Page MA, Krebs HA.Activities of enzymes involved in acetoacetate utilisationin adult mammalian tissues. Biochem J 1971; 121: 41-7.

    19 Bradford MH. A rapid and sensitive method for the quanti-tation of microgram quantities of protein utilising theprinciple of dye binding. Analyt Biochem 1976; 72:248-54.

    20 Allan A, Bristol JB, Williamson RCN. Crypt cell productionrate in ulcerative proctocolitis: differential increment inremission and relapse. Gut 1985; 26: 999-1003.

    21 Winslet MC, Allan A, Poxon V, Youngs D, Keighley MRB.Faecal division for Crohn's colitis: a model to study therole of the faecal stream in the inflammatory process. Gut1994; 35: 236-42.

    22 Keighley MRB, Arabi Y, Dimock F, Burdon DW, AllanRN. The influence of inflammatory bowel disease onintestinal microflora. Gut 1978; 19: 1099-104.

    23 Roediger WEW, Hayworth M, Willoughby P, Piris J, MoreA. Luminal ions and short chain fatty acids as markersof functional activity of the mucosa in ulcerative colitis.J Clin Pathol 1982; 35: 323-6.

    Appendix Summary of Truelove and Richardsclassification for histological severity of ulcerative colitis

    No significant inflammationInflammation NilCrypt abscesses AbsentEpithelium IntactGland architecture Glands reduced in number

    Stunted and atrophicMild to moderate inflammationInflammation Increased chronic inflammatory cell

    Mild increase in neutrophilsVariable intensity

    Crypt abscesses OccasionalEpithelium Increased proliferation

    Intact (apart from crypt abscess)Gland architecture Irregular shape and densitySevere inflammationInflammation Chronic inflammatory cell infiltrateCrypt abscesses Plentiful+erosions of crypt epithelium

    Inflammation extending into laminapropria

    Epithelium Small breaches and erosionsIrregular and often cuboid cells

    adjacent to erosionsGland architecture Markedly irregular in density and

    shape with gland destruction

    on May 31, 2021 by guest. P

    rotected by copyright.http://gut.bm

    j.com/

    Gut: first published as 10.1136/gut.38.6.886 on 1 June 1996. D

    ownloaded from

    http://gut.bmj.com/