Aiirotmiftumr?, Val 32, pp. 2 5 2 6 3 Reprints availsblc directly lrom the publisher Phntocopymg permined hy license only

Q 2000 OPA (Overwaa Puhlisher, Associalion) N.V. Published by liccnse under

the Harwood Academic Publishers imprint, part of Thc Gordon arid Breech Publishing Group

Printed i n Malaysia

Cultures of Human Colonic Epithelial Cells Isolated from Endoscopical Biopsies from Patients with Inflammatory

Bowel Disease. Effect of IFNy, TNFa and IL-lP on Viability, Butyrate Oxidation and IL-8 Secretion*

GITTE PEDERSEN"', TORBEN SAERMARK", KLAUS BENDTZENh and JBRN BRYNSKOV"

"Drpnrtment r$Mrtliral Gastroentrrology C, Herlev University Hospital, Copenhagen, DK-2730 Herlev ond "lnstitlrte f o r Itflammation Research, Righo.ypirdef, Nationtil Univer.sitv Hospiinl, DK-2200 N . Copenhagm, Denmark

(Received Jrmnarj 02, 2000; I n final form Mnv 0.5, 2000)

Cytokine-mediated impairment of viability and metabolic function of epithelial cells has been suggested as a possible early pathogenic event i n the development of inflammatory bowel disease (IBD). It is currently unknown whether pro-inflammatory cytokines have a direct effect on human nontransformed colonic epithelial cells. We investigated the effects of TNFa, IFNyand IL-Ip on viability, short chain fatty acid (butyrate) oxidation and IL-8 secre- tion in human colonic epithelial cell cultures in vitro obtained from macroscopically normal mucosa from IBD patients and controls. Colonic crypts were isolated from endoscopical biopsies by ultra-short (10 min) EDTAEGTA treatment, and exposed to TNFa, IFNy and IL-Ip for 24 hours. The combination of TNFa+IFNy induced a significant decrease in cell viability as judged by methyltetrazoleum (MTT) metabolism which decreased to median 68 % of unexposed cultures (P<O.OI). This effect was more pronounced than that observed after addition of TNFa (median 88 %) (P<O.05), but not IFNy alone (median 78 %), whereas IL-lp had no significant effect. Cells from IBD patients were significantly less sensitive to TNFa + IFNy exposure (median 74 %) compared to cells from controls (median 58 %) (P < 0.05). Butyrate oxidation, as measured by entrapment of 14C02, was not inhibited in cells exposed to TNFa + IFNy, neither from controls (median 112%) nor from IBD patients (median l08%), suggesting a relative increase of this specific metabolic function in living cells in response to immunoinflammatory stress. IL-8 levels in cell supernatants were increased by TNFa + IFNy, supporting the role of the epithelium in signalling between lumi- nal factors and mucosal immune cells. In conclusion, we report that TNFa and IFNy damage and influence human colonic epithelial cell function in vitro and that such mechanisms, if operative in vivo, also may be involved in the pathogenesis of IBD.

Keywords: Butyrate oxidation, Crohn's disease, Cytokines, Inflammatory Bowel Disease, Ulcerative colitis

* This paper is based on original work which in part has been published in abstract form and accepted for poster presentation at Digestive Disease Week (DDW) 1997, Washington, USA and as oral presentation at the 29th Annual Meeting of the Scandinavian Society for Immu- nology 1998, Copenhagen, Denmark and the 31 th Nordic Meeting on Gastroenterology 1999, Tromsfi, Norway.

t Corresponding author. Gitte Pedersen, Herlev University Hospital, 75 Herlev Ringvej, DK 2730 Herlev, Denmark. Tel.: +45 44883418. Fax: +45 44944056. E-mail: giped@herlevhosp.kbhamt.dk

255

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

256 GITTE PEDERSEN et a/.

INTRODUCTION

Proinflammatory cytokines are important mediators of mucosal inflammation in the small and large intes- tine in patients with chronic inflammatory bowel dis- ease (IBD) - i.e. Crohn's disease (CD) and ulcerative colitis (UC).[ 'T~I Cytokines are pleiotropic mediators with a wide range of immunoinflammatory e f f e ~ t s , [ ~ ' ~ 1 but the precise mechanisms of cytokine-mediated mucosal injury are largely unknown. Deem et al (1991) showed that superna- tants of activated lamina propria lymphocytes were cytotoxic to a human colonic epithelial tumour cell line (HT-29) and that tumour necrosis factor (TNF)a and interferon (1FN)y were responsible for this effe~t.1~1 This has led to the hypothesis that TNFa and IFNy, which are both produced in excess in active lBD,lh-*l directly impair epithelial cell function, per- haps as an early event in mucosal inflamma- t i ~ n . [ ' ~ ~ ~ ' " ] Several studies have shown that TNFa and IFNy directly impair barrier function in trans- formed colonic epithelial cell line monolayers, but cytokine-mediated cytotoxicity has been a variable finding.",' '-I4] We have shown that TNFa combined with IFNy induce profound inhibition of butyrate oxi- dation, an essential metabolic function of colonic epi- thelium, both in HT-29 and in DLD-I cell lines, whereas the effect on cell viability varied between the cell lines studied.[ 's . 'hl

It is currently unknown how normal human differ- entiated colonic epithelium respond to cytokines since suitable in vitro models of representative human colonic epithelial cells have been lacking. By use of a new method to establish short-term cultures of normal human colonic epithelial cells from routine endoscop- ical we have studied the effects of TNFa, IFNy and IL-lp on viability, butyrate oxidation and secretion of IL-8 - a chemokine involved in signal- ling between the epithelium and immune cells in the lamina propria.['*] In addition, we also used this model to compare cytokine responses in colonic epi- thelial cells from IBD patients and non-inflamed con- trols.

MATERIALS AND METHODS

Patients

Biopsy specimens were obtained at routine colonos- copy from macroscopically normal mucosa in patients with established IBD according to standard- ised diagnostic criteria.[199201 A total of 26 patients with ulcerative colitis (UC) (median age 41 years, range 21-71, male/female 10/16) and IS patients with Crohn's disease (CD) (median age 45 years, range 24- 77, male/female 7/81 were included. Twenty-three UC patients had left-sided disease or proctitis, while three had more extensive disease, including two patients with pancolitis. In the CD group, five patients had ilia1 disease, four patients left-sided colonic disease, four patients right-sided colonic disease and two pan- colitis. Thirty-six of the IBD patients received 5-ami- nosalicylate treatment, while none received treatment with steroids at the time of inclusion. The control group consisted of 33 patients (median age 49 years, range 18-76, male/female 15/18> included in a polyp surveillance programme or patients undergoing a diagnostic examination showing no signs of neoplas- tic or inflammatory disease. Biopsies were taken after informed concept from the patients. The study was conducted in accordance with the Second Helsinki Declaration and approved by thc Scientific Ethical Committee of Copenhagen County.

Human Colonic Epithelial Cell Cultures

Five to seven biopsies were collected from endoacop- ically normal areas of the transverse or sigmoid colon using standard biopsy forceps. Epithelial cells were isolated using ultra-short (I0 min) EDTA/EGTA treatment and cultured as attached monolayers in 96-well plates coated with bovine dermal collagen. This method releases intact crypts with less than 5 % contaminating cells and approximately 50 5% viable cells after 24 hours as judged by the 3-(4,S-dimethyl- thiazol-2-yl)2,S-diphenyltetrazolium bromide) (MTT) metabolism, vital staining, and electron micros- copy.1'71

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

TNFu and IFNy EFFECTS ON HUMAN COLONOCYTES 257

1201 tt

I I 1 FIGURE 1 Effect of 1L-I b, TNFa and IFNy (all 1W8 M) on human colonic epithelial cells isolated from completely macroscopically normal mucosa. Viability in cultures was measured by 3-(4,5-dime- thylthiazol-2-yl)2,5-diphenyltetrazolium bromide (MTT) assay after 24 hours of exposure. Bars represent MTT values in cytokine exposed cultures as percentage of unexposed parallel cultures (medians, 25-75 percentiles). Bars repr:sents data obtained in cul- tures from 5 - 10 different individuals. p < 0.05, **p < 0.01 com- pared with parallel unexposed controls. "p < 0.05 compared with TNFa

Viability and Butyrate Oxidation

Viability was measured by a colorimetric assay based on conversion of MTT (Sigma, St. Louis, MO, USA) to its formazan by succinate dehydrogenase present in metabolically active cells. MTT (1 mg/ml) was added to the cultures after 18 hours and conversion was assessed during a 6-hour incubation period.'''] Meas- urements were performed in triplicates. Coefficient of variation (CV) for the assay was 5 %.

Butyrate oxidation was measured using a miniature C02-collection system devised for these cell cultures. In brief, epithelial cells were seeded in a central, sealed well (approximately 10' cells/well) in growth

medium containing 1 mM butyrate and 0.125 yCi I4C-labelled butyrate. Generated I4CO2 was collected during 24 hours of culture and trapped by addition of I ml NaOH (1 M) in an outer surrounding chamber. The NaOH solution was collected and mixed with S ml scintillation fluid before counting. Measure- ments were performed in triplicates and CV was I5 % for this assay.

IL-8 Secretion

IL-8 secretion was measured in cell culture superna- tants collected after 18 hours. The IL-8 levels were measured by ELISA (R&D, Abingdon, UK). Meas- urements were performed in duplicates and CV was 20 %.

C ytokines

Human recombinant IFNy ( l o 5 unitlml), TNFa and IL-1 p ( lo6 units/ml) (Sigma) were added alone or in combination as a bolus at the beginning of incubation of cell cultures. MTT metabolism and butyrate oxida- tion were performed after 24 hours of culture at 37 "C. Results obtained in cytokine experiments are expressed as percentage of parallel unstimulated cul- ture from the same patient.

Statistical Analyses

Results are presented as medians with 25-75 percen- tiles unless otherwise stated. Significance was tested by the Wilcoxon or Mann-Whitney tests. P<O.OS (two-sided) was considered significant.

RESULTS

Effect of TNFa + IFNy on Viability of Human Colonic Epithelial Cells

Since TNFa + IFNy is the most powerful inducer of cell damage in HT-29 colonic epithelial cell monolay- ers,[159161 initial concentration-effect studies were car-

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

258 GITTE PEDERSEN el al.

ried out using this combination. Addition of TNFa + IFNy (both at M) induced a small but significant decrease in MTT values to median 81 % (50-91) (n=7) (P<0.05) of unexposed controls after 24 hours. Increased cytokine concentrations resulted in a further decline in median MTT values to 64 % (55-74) (P<O.O1) at lo-* M (n = 8) and 63 % (48-82) (P<0.02) at lop7 M (n=7). Since the effect of the two highest concentrations did not differ (P = NS), M was used in the subsequent experiments.

Effect of Single and Combined Cytokines (lo-' M) on Cell Viability

As shown in Fig. 1, the addition of TNFa or IFNy resulted in a significant decrease in MTT values, whereas IL-lP exposure only induced a minor non- significant inhibition. The combination of TNFa and

A 0.58

g! 0.4 7 3 0

v)

t

CI - - 0.3 * .-

= 0.2

0.1

0.0

.. 8 rn

.3'

v

v +

v

v

0

0

4

-?- t 0

IFNy induced a more pronounced decrease in MTT value, which was additive rather than synergestic compared to the effect of cytokines added alone. The combination of TNFa, IFNy and IL- 1 p had no addi- tional effect, and the combination of TNFa and IFNy was used in subsequent experiments.

Effects of TNFa + IFNy (lo-' M) on Viability and Butyrate Metabolism in Human Colonic Epithelial Cells from IBD and Control Patients

Fig. 2 (panel A) shows that baseline viability in cell cultures from controls, UC and CD patients were identical as judged by MTT values after 24 hours. Similarly, no significant differences were observed in spontaneous butyrate oxidation rates between cultures from controls and patient groups (Fig. 2, panel B).

6

250

f!! 200 2 N ? 150

100

50

3

0 F

E a 0

0 uc CD Controls

(n=10) (n=5) (n=12)

0

0 0

uc CD Controls (n =5) (n= 5) (n = 12)

FIGURE 2 Panel A. Comparison of colonic epithelial cell viability in cultures from ulcerative colitis (UC), Crohn's disease (CD) and control patients. Median MTT absorbency values after 24 hours of culture for each group are illustrated by vertical lines. N-values refer to numbers of individual patients in each group. (P = NS between controls and UC or CD) Panel B. Comparison of butyrate metabolism in cell cultures obtained from UC, CD and control patients. Assessment of butyrate oxidation rate is based on I4CO2 liberation after 24 hours in culture. Medians are shown as vertical lines. N-values refer to numbers of individual patients in each group. (P = NS between controls and UC or CD)

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

TNFu and IFNy EFFECTS ON HUMAN COLONOCYTES 259

TABLE I The effect of tunlour necrosis factor (TNF)u combined with interferon (1FN)yon interleukin-8 (1L-8) secretion

IL-8 concentration (pg / m l ) median (range)

Unstirnuluted T N F d I F N y P

IBD-patients (n = 7) 260 (I9 - 837) 334 (39 - 802) NS

Control-patients (n = 8) 246(19-496) 5 I5 (19 - 788) < 0.02

P NS NS NS = not significant.

Following exposure to TNFa + IFNy, MTT metab- olism decreased in cultures from both control and IBD patients, but IBD cells were significantly less sensitive (Fig. 3). No differences in cytokine-induced MTT responses between cultures from UC patients (n = 10) and CD patients (n = 5 ) were encountered (P=NS). As judged by baseline viability in the cul- tures (Fig 2), the observed differences in cytokine sensitivity was not explained by differences in sur- vival in vitro. Surprisingly, butyrate oxidation and MTT metabolism was not inhibited in parallel, as butyrate oxidation increased sligthly in cells from both controls and IBD patients although these eleva- tions did not reach significance.

Assessment of MTT metabolism and butyrate oxida- tion in parallel in the same patients revealed that butyrate oxidation / MTT ratios increased significantly in TNFa + IFNy exposed cultures compared to unex- posed cultures from control patients (median 501 x lo2 cpm / MTT unit (range 190-771 x lo2) versus 290 x lo2 cpm / MTT unit (200-423 x lo2), n = 6) (P < 0.05). Similarly, a cytokine-mediated increase of butyrate oxidation / MTT ratios was observed in cul- tures from IBD patients, although this did not reach significance (median 368 x lo2 cpm / MTT unit (range 281-714 x 10') versus 270 x lo2 cpm / MTT unit (165-619 x lo')), n = 7)(P = 0.06). In separate sets of control experiments, sonication (10 seconds) or Tri- ton-X (0.05 76) treatment was used to permeabilise the cell membranes. Butyrate oxidation in these cultures did not increase compared to values in untreated con- trols, suggesting that the observed relative increase of butyrate metabolism compared to MTT values was not explained simply by cytokine-induced cell membrane damage with enhanced non-limited transport and oxi- dation of butyrate (data not shown).

RiS3 Butyrate T

(n = 12) (n = 15) (n = 10) Controls IBD

FIGURE 3 Comparison of TNFu + IFNy (both lo-' M) responses in colonic epithelial cell cultures from control patients and patients with IBD. MTT metabolism and butyrate oxidation was measured in cell cultures after cytokine exposure for 24 hours. N-values refer tu numbers of individual patients in each group. :% P < 0.01 com- pared with parallel unexposed cultures. # P < 0.05 compared with MTT values in IBD cell cultures

TNFa + IFNy - Induced IL-8 Secretion

While IL-8 secretion was not detectable after 4 hours, it increased gradually during the next 14 hours of cul- ture. As shown in Table I, both control and IBD cells secreted IL-8 spontaneously after 18 hours in culture. The IL-8 secretion increased significantly in culture supernatants from controls challenged by TNFa + IFNy. In contrast TNFa + IFNy treatment induced only a minor nonsignificant increase of IL-8 levels in cultures from IBD patients.

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

260 GITTE PEDERSEN et nl.

DISCUSSION

Epithelial cells appear to play an active role in the pathogenesis of chronic intestinal inflammation and the general concept that these cells are innocent bystanders is no longer valid. Hence, epithelial cells secrete or express a variety of proinflammatory cytokines, growth factors, chemokines, chemokine receptors, arachidonic acid metabolites and nitric oxide.11,22-2s1 Studies in cell line models have shown that these cells also respond in vitro to cytokines, such as TNFa and IFNy, leading to impaired barrier func- tion and chemokine expression.1'2,2~,2s1 Therefore, it has been proposed that proinflammatory cytokines, produced by infiltrating immune cells during active stages of inflammation, directly damage or kill epi- thelial cells and that these mechanisms may be involved in the pathogenesis of human IBD. Unfortu- nately, the vast majority of data have been collected from studies using transformed epithelial cell line models, which may not be fully representative for human intestinal epithelium.[9% 12,'s,251 We report for the first time that TNFa + lFNy directly impair the viability and integrity of differentiated human colonic epithelial cells isolated from macroscopically normal mucosa. MTT metabolism declined in cultures fol- lowing TNFa + IFNy exposure for 24 hours, and this combination was significantly more cytotoxic to the cells than TNFa alone, but only slightly more toxic than IFNy alone. Thus, the strong synergistic effect observed in HT-29 colonic epithelial cell monolayers after 48 was not observed in these cul- tures. Indeed, l 0-fold higher cytokine concentrations were required to induce a marked decline in MTT metabolism in differentiated human colonic epithelial cells, suggesting that these cells are somewhat less sensitive to cytokines than HT-29 cells. However, the data corroborate previous studies in HT-29 cell lines showing that IFNy seems to play a pivotal role in cytokine-mediated colonic epithelial cell

IL- 1 p has many biological effects in common with and it has also been implicated in the

pathogenesis of IBD.r31 We found that IL-Ip had no effect on cell viability and also failed to augment the

injuryp15,2h.27 I

cytotoxic effect of TNFa + IFNy. These data are in accordance with previous findings in HT-29 cell mono layer^,^'^"^^ indicating that IL- 1 p is largely non-toxic to human colonic epithelial cells in vitro.

In the present study, biopsy specimens obtained at routine endoscopy were used for establishment of short-term cell cultures. This approach is attractive, because it provides the opportunity to investigate representative colonic epithelial cells obtained from a broad spectrum of patients, including early stages of IBD and control patient^."^] The epithelial cells from normal IBD mucosa proved to be less sensitive to TNFa + IFNy -induced cell damage than cultures from controls (Fig. 3). Differences in baseline sur- vival between cultures from IBD- and control patients cannot explain this observation (Fig. 2) but it might reflect intrinsic defence mechanisms triggered in IBD cells as a result of exposure to toxic luminal factors or high local cytokines levels in vivo. This notion is sup- ported by the demonstration of increased TNFa and IL- 1 p levels, even in uninvolved IBD mucosa.LY3301 Increased epithelial crypt cell renewal in colonic IBD mucosa may also result in a higher proportion of young cells, less sensitive to toxic factors when cul- tured in v i t v ~ . ' ~ ' l Intestinal epithelial cell lines have been shown to respond to thermal or oxidative stress by inducing heat shock proteins. Currently it is unknown whether similar or other protective mecha- nisms are induced in colonic epithelial cells following cytokine-mediated injury.r321

Oxidation of butyrate, a short-chain fatty acid, is an essential energy source required for maintenance of the functional integrity of normal human colonic epi- thelial cells. In the present study, spontaneous butyrate oxidation was identical in cultures obtained from UC, CD patients and controls. Since cells were isolated from endoscopically normal mucosa, our data support the view that impaired butyrate oxidation pre- viously observed in active UC may be the conse- quence of chronic inflammation, rather than a primary pathogenic mechanism.L'3-"1 Surprisingly, butyrate oxidation was not affected by TNFa + lFNy expo- sure, neither in IBD-nor in control patient cultures. This observation contrast findings in colonic epithe- lial cell lines (HT-29 and DLD- 1 ) where butyrate oxi-

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

TNFa and IFNy EFFECTS ON HUMAN COLONOCYTES 26 1

dation decreases in parallel with or prior to decrease in MTT m e t a b o l i ~ m . [ ’ ~ ~ ’ ~ 1

In contrast to other cytotoxicity assays based on assessment of cell membrane integrity (e.g. lactate dehydrogenase release, cellular trypan blue staining, 51Cr relase), the MTT test allows detection of early cellular damage such as impaired mitochondria1 function, which occurs before loss of membrane inte-

Assuming that the MTT test also in the present model reflects the number of viable cells, our results suggest that butyrdte oxidation is increased in primary human colonic epithelial cells in response to TNFa + IFNy exposure. This notion is supported by a previous study showing increased activity of enzymes in the P-oxidation pathway ( i .e . butyryl-CoA and hydroxybutyryl-CoA) in mucosa from UC patients with active disease.[401 Control experiments using sonication or detergent to disrupt cell membrane integrity showed that the relative increase of butyrate oxidation was not explained by increased non-limited transport of butyrate over damage cell membranes. It is therefore possible that differentiated colonic epithe- lial cells increase butyrate oxidation by enzyme acti- vation or promotion of enzyme synthesis in response to cytokine-induced cellular stress. The MTT test is, however, both an index of cellular integrity and cell number13*] and an alternative explanation is that butyrate metabolism is more resistant to cytokine induced damage than MTT reduction in vitro. IL-8 is one of the best characterised members of

the C-X-C family of chemokines which participate in immunoinflammatory responses through chemoat- traction and activation of leucocytes.’’ 8,241 Spontane- ous IL-8 secretion has been reported in colonic epithelial cells isolated from resected IBD m u ~ o s a . ~ ~ ~ ~ Our data, showing that cultures of human colonic epithelial cells from biopsies spontaneously secrete IL-8, extend these observations and provide further evidence for a role of the colonic epithelium in immune signalling in the human intestine. We were also able to extend previous findings in transformed colonic epithelial cell lines,[427431 by showing that proinflammatory cytokines increase IL-8 secretion in normal differentiated colonic epithelial cells in vifro. A similar tendency was also observed in IBD cultures

although the predefined levels of significance was not reached. It remains unresolved whether the nonsignif- icant P-value represents a statistical type-2 error or actually reflects differences in the immunological capacity between IBD and control cell cultures.

In conclusion, we report that TNFa and IFNy directly influence human colonic epithelial cell func- tion in vitro and that such mechanisms, if operative in vivo, also may be involved in the pathogenesis of IBD.

Acknowledgements

The authors wish to thank employees at Dept. Medi- cal Gastroenterology C and Dept. Surgical Gastroen- terology D, Herlev Hospital for assistance in collection of biopsies and Mrs. B. Deibjerg and Mrs. A. Pedersen for their technical assistance in the labo- ratory.

Grant Support

P.C. Petersen’s Foundation, Brdr. Hartmann’s Founda- tion, J. Madsen and Spouse 0. Madsen’s Foundation, 0. Buhl Olesen and Spouse E. Buhl Olesen’s Founda- tion, D. Hansen’s Foundation, S.R. Moran’s Founda- tion, V. Binder and P. Riis’ Foundation, Danish Hospital Foundation for Medical Research, Region Copenhagen. Faroe Islands and Greenland, the Dan- ish Biotechnology Programme and the University of Copenhagen, Faculty of Medicine.

References Sartor RB. Cytokines in inte5tinal inflammation: Parhophysi- ological and clinical considerations. Gastroenterology 1994; 106:533-539. D’Haens G, Van Deventer S , Van Hogezand R, Chalmers D, Kothe C, Baert F et al. Endoscopical and histological healing with anti-tumor necrosis factor antibodies in Crohn’s disease: A european multicenter trial. Gustroenrerology 1999: 116: 1029-1034. Brynskov J, Nielsen 0, Ahnfelt-Rmne I , Bendtzen K. Cytokines (immunoinflammatory hormones) and their natu- ral regulation in inflammatory bowel diseases (Crohn’s dis- ease and ulcerative colitis). A review. Dig Di.7 1994; 12290- 304. Bendtzen K . Cytokines and natural regulators of cytokines. Onrnunol Left 1994: 43:ll 1-123. Deem RL. Shanahan F, Targan SR. Triggered human mucosal T-cells release tumour necrosis factor-alfa and inter-

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

262 GITTE PEDERSEN et al.

feron-gamma which kill human colonic epithelial cells. Clin Exp Immunol 1991; 83:79-84. Radford-Smith G, Jewel DP. Cytokines and inflammatory bowel disease. Baillieres Clin Gastroenterol 1996; 10: 15 1- 164. Plevy S , Landers C, Prehn J, Carramanzana NM, Deem RL, Shearly D et al. A role for TNFa and mucosal T helper-I cytokines in the pathogenesis of Crohn's disease. J lmmunol 1997; 159:6276-6282. Schreiber S , Nikolaus S , Hampe J, Hamling J, Koop I, Groessner B et al. Tumour necrosis factor alpha and inter- leukin 1 beta in relapse of Crohn's disease. LanL~er 1999; 353:459-461, McKay DM, Baird A. Cytokine regulation of epithelial per- meability and ion transport. Guf 1999; 44:283-289. Fiocchi C. Inflammatory Bowel Disease: Etiology and patho- genesis. Gasrroentcrologv 1998; 115: 182-205. Di Paolo MC, Merrett MN, Crotty B, Jewel DP. S-aminosali- cylic acid inhibits the impaired epithelial barrier function induced by gamma interferron. Gut 1996; 38:llS-I 19. Mahraoui L, Heyman M, Plique MT, Desjeux JF. Apical effect of diosmectite on damage to the intestinal barrier induced by basal tumour necrosis factor-alfa. Cur 1997: 40:339-343. Madara JL, Stafford J. Interferon-gamma directly affects bar- rier function of cultured intestinal epithelial monolayers. J Clin Invest 1989; 83:724-727. Jarry A, Muzeau F, Laboisse C. Cytokine effects in a human colonic colonic goblet cell line. Dig Dis Sci 1992; 37:l 170- 1178. Pedersen G, Saermark T, Bendtzen K, Giese B, Horn T, Brynskov J. Cytokine-induced damage to human colonic epi- thelial cells: viability and metabolism. Scan J Immuriol 1998; 47: A 34. Pedersen G, Saermark T, Bendtzen ti, Giese B, Horn T, Brynskov J. Cytokine-induced impairment of short chain fatty acid oxidation and viability in two human colonic epi- thelial cell lines. Cytokirie (in press). Pedersen G, Saermark T, Giese B, Hansen A. Drag B, Bryn- skov J . A simple method to establish short-term cultures of normal human colonic epithelial cells from endoscopical biopsies. S a n d J Gastroenterol ( in press). MacDermott RP, Sanderson I, Reinecker H. The central role of chemokines (chemotactic cytokines) in the immunopatho- genesis of ulcerative colitis and Crohn's disease. Inf7umm B C J W ~ Dis 1998; 454-67. Langholz E, Munkbolm P, Davidson M, Binder V. Couse of ulcerative colitis: analysis of changes in disease activity over years. Gtistroenrerology 1994; 1 0 7 5 1 I . Munkholm P, Langholz E, Nielsen 0, Kreiner S , Binder V. Incidence and prevalence of Crohn's disease in the county of Copenhagen, 1962-87: A sixfold increase in incidence. Scand J Gastroenterol 1992; 27:609-614. Liu Y, Peterson DA, Kimura H. Schubert D. Mechanism of cellular 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra~~)- lium bromide (MTT) reduction. J Neurochem 1997; 69581- 593. Christ A, Blumberg R. The intestinal epithelial cell: immuno- logical aspects. Springer Sernin immunoparhol 1997; 18:449461. McKay DM, Philpott DJ, Perdue MH. In vitro models in inflammatory bowel disease research - a critical review. Ali- ment Pharmacol Ther 1997; 11:70-80.

[24] Dwinell MB, Eckmann L, Leopard JD, Varki NM, Kagnoff ME Chemokine receptor expression by human intestinale epithelial cells. Gastroenterology 1999; 117:359-367.

[25] Madsen K, Lewis SA, Tavernini MM, Hibbard J, Fedorak RN. Interleukin 10 prevents cytokine-induced disruption of T84 monolayer barrier integrity and limits chloride secretion. Ga.rtroenterology 1997; 113: 15 1-1 59.

[261 Xu X, Fu X, Plate J, Chong A. IFN-y induced cell growth inhibition by Fas-mediated apopto Requirement of Stat1 protein for upregulation of Fas and FasL expression. Cancer Res 1998; 58:2832-2837.

1271 Ossina N, Cannas A, Powers V, Fitzpatrick PA, Knight JD, Gilbert JR et al. Interferon-y modulates a pS3-independent apoptotic pathway and apoptosis-related gene expression, J Biol Chem 1997; 272:16351-16357.

1281 Bendtzen K, Mandrup-Poulsen T, Nerup J , Nielsen J, Din- arello C, Svenson M. Cytotoxicity of human pl 7 inter- leukin-I for pancreatic islets of Langerhans. Science 1986;

1291 Bendtzen K. Immune hormones (cytokines); pathogenic role in autoimmune theumatic diseases and endocrine diseases. Autoimmunity 1989; 2:177-189.

1301 Reimund J-M, Wittersheim C, Dumont S , et al. Increa5ed production of tumor necrosis factor alfa, interleukin- 1 beta, and interleukin-6 by morphologically normal intestinal biop- sies from patients with Crohn's disease. Gut 1996; 39:684- 689.

1311 Gibson PR, Barkla DH. Mucosal metabolism and prolifera- tion. In: Allan RN, Rhodes JM, Hanauer SB, Keighley MRB, Alexander-Williams J, Fazio VW. ed. Inflammtrfory Bowel Disrase. 3 Edn. New York: Churchill-Livingstone, 1997:201-214.

[32] Musch M, Ciancio M, Sarge K , Chang E. Induction of heat shock protein 70 protects intestinal epithelial IEC- I8 cells from oxidant and thermal injury. A m .I P/iJ.tiol 1996;

[33] Clausen MR, Mortensen PB. Kinetic studies on colonocyte metabolism of short chain fatty acids and glucose in ulcera- tive colitis. Gut 199.5; 37:684-689.

1341 Roedinger W. The colonic epithelium in ulcerative colitis: an energy defiency disease'? Lancet 1980: 21:793-798.

[35] Finnie IA , Taylor BA, Rhodes JM. Ileal and colonic epithe- lial metabolism in quiescent ulcerative colitis: increased glutainine metabolism in distal colon but no defect in butyrate metabolism. Gut 1993; 34: 1552-15S8.

[36] Simpson E, Chapman M, Dawson J, Berry I, Macdonald I , Cole A. In vivo measurement of colonic butyrate metabolism in patients with quiescent ulcerative colitis. Cur 2000; 4673- 77.

[37] Chapman M, Grahn MF, Boyle MA, Hutton M. Rogers J, Williams NS. Buiyrote oxidation is impaired in colonic mucosa of sufferers of quiescent ulcerative colitis. Cur 1994; 35:73-76.

1381 Watson AJM, Askew JN. Sandle GI. Characterisation ofoxi- dative injury to an intestinal cell line (HT-29) by hydrogen peroxide. Gut 1994; 35:1575-1581.

1391 Schulze-Osthoff K, Bakker A, Vanhaesebroeck B, Beyaert R, Jacobs WA, Fiers W. Cytotoxic activity of tumor necrosis Factor is mediated by early damage of mitochondria1 func- tions. J Biol Chern 1992; 2675317-5323,

[401 Allan ES, Winter S , Light AM, Allan A. Mucosal enzyme activity for butyrate oxidation; no defect in patients with ulcerative colitis. Gut 1996; 38:886-893.

232: 1.545-1547.

270:C429-C436.

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.

TNFa and IFNy EFFECTS ON HUMAN COLONOCYTES 263

1411 Gibson PR, Rosella 0. Interleukin 8 secretion by colonic [43] Yang S, Eckmann L, Panja A, Kagnoff M. Differential and crypt cells in vitro: response to injury supressed by butyrate regulated expression of C-X-C, C-C and C-chemokines by and enhanched in inflammatory bowel disease. Gut 1995; human colon epithelial cells. Gastroenrerology 1997;

[42j Warhurst A, Hopkins S, Warhurst G . Interferon y induces dif- ferential upregulation of a and p chemokine secretion in colonic epithelial cell line. Cur 1999; 42:208-213.

37536-543. 113: 1214-1223.

Aut

oim

mun

ity D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 10

/30/

14Fo

r pe

rson

al u

se o

nly.