lessons in ibd pathogenesis from new animal models of...

Lessons in IBD pathogenesisfrom new animal models of

spontaneous colitisR BALFOUR SARTOR MD

UNTIL RECENTLY, MOST EX-

perimental models of intestinalinflammation have been induced inlaboratory animals by exposure to avariety of toxins of questionable envir-onmental relevance (1,2) (Table 1).The exception is the cotton-top tamar-in, which spontaneously develops coli-tis and adenocarcinoma of the colon.However, this new world marmoset ison the endangered species list and isnot widely available. Existing animalmodels have provided very importantinsights into soluble mediators involv-ed in acute and, in some cases, chronicintestinal inflammation. However, theacute injury response to toxins is of lim-ited value to assist the understand-ingof basic immunoregulatory defectsrelevant to the pathogenesis of idio-pathic chronic inflammatory boweldisease (IBD) in humans.

Significant improvements in tradi-tional animal models have been madeby exploring the differential geneticsusceptibility of inbred rat strains tobacterial cell wall polymers (1,3). In-bred Lewis rats develop a biphasic,spontaneously relapsing granuloma-tous enterocolitis with associated fibro-sis, arthritis, granulomatous hepatitis,anemia and leukocytosis after intramu-ral injection of peptidoglycan-polysaccharide (PG-PS) polymers. The

IBD – BASIC SCIENCE: GENETICS, MARKERS AND MODELS

RB SARTOR. Lessons in IBD pathogenesis from new animal models of sponta-neous colitis. Can J Gastroenterol 1995;9(6):309-315. The recent explosionof transgenic and targeted gene deleted (knockout [KO]) rodents has yielded anumber of new animal models of spontaneous, chronic intestinal inflammationthat have provided novel insights into the pathogenesis of human inflammatorybowel disease (IBD). Spontaneous colitis resulting from deletion of genes encod-ing key immunoregulatory cytokines (interleukin [IL]-2, IL-10 and transforminggrowth factor [TGF]-beta) and T cell receptors (TCRs) demonstrates that an intactmucosal immune response prevents colitis. The TCR KO model incriminates Blymphocytes in spontaneous colonic inflammation – TCR KO with intact B cellscauses colitis, but simultaneous deletion of T and B cells does not. This model andinduction of colitis in severe combined immunodeficient (SCID) mice by consti-tution with one T cell subset (CD45RHhi), but prevention by addition of theCD45RBlo subset, strongly suggest that T cell subsets down-regulate inflammationin the normal, immunocompetent host. An essential role for normal luminal bac-teria in induction and perpetuation of enterocolitis is provided by the absence ofchronic intestinal inflammation in germ-free (sterile) IL-2 KO mice and humanleukocyte antigen (HLA)-B27 transgenic rats, and attenuated inflammation in IL-2and IL-10 KO mice raised under specific pathogen-free conditions. The fundamen-tal role of host genetic susceptibility in chronic intestinal inflammation and sys-temic manifestations is established by development of spontaneous colitis andperianal inflammation in C3H/HeJ Bir substrain mice and HLA-B27 transgenicrats. (Pour le résumé, voir page 310)

Key Words: Animal models, Colitis, Enterocolitis, Gene knockout, Genetic engineer-ing, Transgenic

Center for Gastrointestinal Biology and Disease, University of North Carolina, Chapel Hill,North Carolina, USA

Correspondence: Dr R Balfour Sartor, Division of Digestive Diseases and Nutrition, School ofMedicine, Department of Medicine, CB 7080, Room 326 Burnett-Womack Building, Universityof North Carolina at Chapel Hill, Chapel Hill, NC 27599-7080, USA. Telephone919-966-0149, fax 919-966-6842, e-mail [email protected]

This paper was presented at the Trends in Inflammatory Bowel Disease Therapy meeting,April 6 to 9, 1994, held in Victoria, British Columbia. This paper has also been published in SutherlandLR, et al, eds. Inflammatory Bowel Disease: Basic Research, Clinical Implications and Trends inTherapy. Boston, Dordrecht and London: Kluwer Academic Publishers, 1994

CAN J GASTROENTEROL VOL 9 NO 6 SEPTEMBER/OCTOBER 1995 309

chronic phase of PG-PS-induced ente-rocolitis is T lymphocyte-mediated andpersists for at least six months. In con-trast, Fischer F344 rats (major histo-compatibility complex [MHC]-matchedwith Lewis) and Buffalo rats develop

only transient intestinal injury and noextraintestinal inflammation. Al-though the PG-PS model has severalunique features, intramural injection ofPG-PS at the time of laparotomy is a te-dious and nonphysiological means of

inducing experimental and intestinalinflammation.

Recent advances in molecular bio-logical techniques have made it practi-cal to overexpress or delete genes ofinterest in rodents as well as in viruses,bacteria, yeast and cultured cells. Over-expression (transgenic) and deletion(knockout [KO]) of specific human leu-kocyte antigen (HLA), cytokine and Tcell receptor (TCR) genes in rodentshave provided important and some-times surprising models of spontaneouscolitis (Table 2). These new models, incombination with a spontaneous mu-tation in C3H/HeJ mice, restorationof T cell subsets to immunocompro-mised mice and induction of intestinalinflammation by PG-PS, indomethacinor trinitrobenzene-sulphonic acid ininbred rodent strains, have providednovel insights into mechanisms ofchronic intestinal and systemic inflam-mation.

This paper describes clinical, histo-logical and immunological features ofthese animal models, concentrating onthe newer, genetically engineered ro-dents that spontaneously developchronic intestinal inflammation.These new rodent models provide thetools to dissect the basic mechanisms ofchronic intestinal and systemic inflam-mation and provide relevant insightsinto the pathogenesis of IBD.

C3H/HeJ BIR SUBSTRAIN MICEIn 1994 Sundberg and colleagues

(4) described a heritable form of colitisin mice. Over the past 10 years it be-came apparent that individualC3H/HeJ mice episodically developeddiarrhea and perianal ulceration withno demonstrable pathogens. Selectivebreeding of mice exhibiting diarrheaestablished a pedigree in which there isa 50% incidence of spontaneous colitis.There is a male predominance (67%incidence in males, 31% in females)and seasonal variation (occurring moreoften in winter than in summer). Thetime course of inflammation is bipha-sic. The most reproducible inflamma-tion develops at three to five weeks ofage with spontaneous resolution inmost mice. Some mice exhibit chronicinflammation, which is failure to re-

La pathogenèse des MII tirée de nouveaux modèles animaux decolite spontanée

RÉSUMÉ : La récente multiplication de rongeurs transgéniques dont un gène ci-blé a été éliminé (knockout [KO]) a donné lieu à un nombre de nouveaux modèlesanimaux d’inflammation intestinale chronique spontanée qui donnent unéclairage nouveau à la pathogenèse des maladies inflammatoires de l’intestinchez l’humain (MII). La colite spontanée résultant de l’élimination de gènes re-sponsables de l’encodage des cytokines immunorégulatrices clés (interleukineIL-2, IL-10 et facteur de croissance transformant [TGF]-bêta) et les récepteurs descellules T (TCR) démontrent qu’une réponse immunitaire intacte au niveau de lamuqueuse prévient la colite. Le modèle TCR KO élimine les lymphocytes B dansl’inflammation spontanée du côlon. Les TCR KO avec cellules B intactes provo-quent la colite, contrairement à l’élimination simultanée des cellules T et B. Cemodèle et l’induction de la colite chez des souris atteintes d’une immunodéfi-cience combinée congénitale grave avec un sous-ensemble de cellules T(CD45RHhi), qui peut être prévenue par l’ajout de la sous-série CD45RBlo, suggèrentfortement que les sous-ensembles de cellules T régissent à la baisse l’inflamma-tion chez l’hôte immunocompétent normal. Le rôle central des bactéries nor-males dans l’induction et la perpétuation de l’entérocolite est rendu possible parl’absence d’inflammation intestinale chronique chez des souris IL-2 KO non con-taminées (stériles) et chez des rats transgéniques (HLA)-B27 soumis à l’antigèneleucocytaire humain et par l’atténuation de l’inflammation chez des souris IL-2 etIL-10 KO dans des conditions spécifiques, sans organisme pathogène. Le rôle fon-damental de la sensibilité génétique de l’hôte dans l’inflammation intestinalechronique et dans les manifestations systémiques a pu être confirmé par l’installa-tion de colite et d’inflammation périanale spontanée chez des sous-espèces desouris Bir C3H/HeJ et des rats transgéniques HLA-B27.

TABLE 1Established animal models of intestinal inflammation

Mucosal Intermediate Transmural

Dextran sulphate sodium

(mouse)

Acetic acid (rat) Trinitrobenzene-sulphonic acid

(rat, mouse)

Immune complex

(rabbit)

Peptidoglycan-polysaccharide

(rat)

Carrageenan (guinea pig) Indomethacin (rat, dog)

Cotton-top tamarin

TABLE 2New rodent models of spontaneous intestinal inflammation

Spontaneous mutation Genetically engineered Reconstituted

C3H/HeJ Bir mouse HLA-B27�2� transgenic rat CD45RBhi

�SCID mouse

IL-2 deleted mouse

IL-10 deleted mouse

�� TCR deleted mouse

TGF-�1 deleted mouse

HLA-B27�2� Human leukocyte antigen B27 �2 microglobulin; IL Interleukin; SCID Severe combined immuno-

deficient; TCR T cell receptor; TGF Transforming growth factor

310 CAN J GASTROENTEROL VOL 9 NO 6 SEPTEMBER/OCTOBER 1995

SARTOR

solve either the acute phase or sponta-neous reactivation of chronic in-flammation. Clinical manifesta-tions of inflammation include occa-sional diarrhea, hemoccult-positivestools and perianal ulceration. Inflam-mation is routinely found in the cecumnear the ileocecal valve, but can in-volve the left colon. Histological char-acteristics include focal linear ulcerswith infiltration of mononuclear cellsand neutrophils. Extraintestinal mani-festations have not yet been reported.

The mechanism(s) of spontaneouscolitis in the C3H/HeJ Bir substrainmouse have not yet been determined.Presumably the propensity to developinflammation is the result of spontane-ous mutation of a specific gene. At-tempts to identify this gene are in prog-ress, and if successful will provideimportant clues to direct investigationsof specific gene mutations in IBD pa-tients. The C3H/HeJ parent strain isinteresting because it is resistant tobacterial lipopolysaccharide (LPS) orendotoxin. Like the parent strain,macrophages from C3H/HeJ Bir micefail to produce interleukin (IL)-1 or tu-mour necrosis factor (TNF)-alpha in re-sponse to in vitro LPS stimulation (5).

HLA-B27 TRANSGENIC RATSHLA-B27/�2 microglobulin (�2�)

transgenic rodents were developed todetermine whether experimental invivo expression of this class I MHC

molecule would drive a systemic in-flammatory response that resemblesthe human HLA-B27 syndrome. HLA-

B27/ �2� transgenic mice fail to exhibitclinical evidence of spontaneous in-flammation, but are more susceptible toexperimental infection with Yersinia

species (6). However, transgenic ratsthat expressed high copy numbers of thehuman HLA-B27/�2� gene develop asystemic syndrome of polyarthritis, der-matitis, hair loss, nail changes, myocar-ditis, epididymitis, progressive wasting,gastritis and colitis (7). Surprisingly,diarrhea usually precedes other mani-festations and is clinically evident in50% of transgenic rats by 10 weeks ofage and in almost 100% by 20 weeks ofage (8). The chronic diarrhea is notgrossly bloody and is associated with in-

termittent perianal ulceration (unpub-lished observations). The colon isgrossly thickened and granular but hasno discrete ulceration, nodularity orstrictures. Histologically there is an in-filtration by mononuclear cells into thelamina propria, with occasional cryptabscesses in severe cases. Inflammationis confined to the mucosa in almost allrats. Crypts are hyperplastic with reac-tive atypia and goblet cell depletion.The majority of rats exhibit focal areasof gastric inflammation in the laminapropria and submucosa with a prolifera-tion of mucus-neck cells and grandularectasia. In the joints, neutrophils accu-mulate in the joint space; the synoviumis hyperplastic, infiltrated with mono-nuclear cells, and pannus erodes thebone.

Several studies have been performedto determine the mechanisms of pro-gressive intestinal and systemic inflam-mation in this model. The susceptibil-ity to inflammation correlated with thelevel of human HLA-B27/�2� expressionon lymphoid cells (8). Eighty per centof hemizygous transgenic rats in twolines that express more than 50 copiesof recombinant HLA-B27/�2� per celldeveloped clinically evident colitis andsystemic disease by 10 to 16 weeks ofage. Rats from lines expressing lowcopy numbers of B27 failed to develop

inflammation even in the homozygous

state, whereas homozygous but not het-

erozygous offspring of mid-level line

developed clinical disease. Interest-

ingly, there is no apparent difference in

frequency or onset of colitis or systemic

inflammation in Lewis or Fischer rat

lines that express equivalent copy

numbers of the transgene. The lack of

clinically apparent spontaneous in-

flammation in transgenic mice express-

ing high copy numbers of the same con-

struct is difficult to explain.

Clinical disease can be transferredby bone marrow cells fromHLA-B27-positive donors to irradiated

B27-negative recipients (9). Disease

can also be transferred by fetal liver

cells, but not by mature lymph node or

spleen cells, indicating that repopula-

tion of the recipient by B27-positive

progenitor cells is necessary for diseaseinitiation and that donor cells do not

need to be derived from rats with ac-

tive inflammation. Moreover, disease

remitted in B27-positive, irradiated re-

cipients following engraftment with

B27-negative cells. These studies dem-

onstrate the colitis is mediated by im-

munocytes and that HLA-B27

expression by nonimmune cells

(epithelial, endothelial, etc) is neither

sufficient nor necessary for colitis in-

duction.

Several studies demonstrate a fun-damental role of ubiquitous conven-tional flora in the pathogenesis ofHLA-B27 associated colitis and arthritis

(10,11). HLA-B27/�2� transgenic ratsraised in a sterile (germ-free) environ-ment fail to develop clinically apparentdiarrhea or arthritis and have no eleva-tion of colonic myeloperoxidase or IL-1

compared with rats housed under con-ventional conditions with no demon-strable bacterial pathogens. Theincidence of arthritis in HLA-B27 trans-

genic rats repopulated with specific

pathogen-free bacterial flora is lessthan in those raised in a conventionalrodent facility. Despite the absence ofcolitis and arthritis, germ-free trans-genic rats develop hair loss, dermatitisand epididymitis to the same degree as

conventional B27-positive rats. A role

for anaerobic bacteria is suggested by

attenuation of colitis and gastritis by

metronidazole (12).

IL-2 KO MICEIL-2 is a key immunoregulatory cyto-

kine produced by activated T helper(TH)1 lymphocytes whose principal ac-tivities are to stimulate proliferationand activation of effector and regula-tory lymphocytes (13). Mice homozy-gous for the mutated IL-2 gene (IL-2 –/–)develop normally for three to fourweeks before exhibiting a progressivewasting syndrome (14). Approximately50% of mice die by nine weeks of agewith evidence of splenomegaly, lym-phadenopathy and severe hemolyticanemia. Surviving mice developchronic diarrhea, episodic bloodystools and rectal prolapse, which be-come clinically evident between sixand 15 weeks of age. Progressive diseaseresults in death by 10 to 25 weeks.These mice develop pancolitis, with


New animal models of spontaneous colitis

the left colon affected to a greater de-gree, but no evidence of small intesti-nal inflammation. The inflamed colonis thickened with crypt hyperplasia butinflammation is confined to the mu-cosa. The mucosa is ulcerated, crypt ab-scesses are present and the laminapropria is infiltrated by mononuclearcells and neutrophils. In advancedstages, epithelial dysplasia is noted.Amyloidosis develops in the liver,spleen and kidney.

Immunopathogenesis of colitis inthis model has been explored by immu-nohistochemical staining. The numberof T and B cells within the lamina pro-pria dramatically increases in coliticmice which had up to 100 times moreT lymphocytes than controls. In thelate stages of disease, colonic B cells arereduced. T lymphocytes in inflamedcolons are activated, as indicated bythe CD44 (activated memory T cells)and CD69 (early T cell activation)markers. T cells within inflammatoryfoci have increased proliferation rates.Lamina propria immunoglobulin (Ig)G1 and IgA secreting cells are dramati-cally increased and serum IgA is ele-vated 16 to 64 times the control values.Anticolon antibodies are increased un-til late stages of inflammation, but arepart of a generalized autoantibody re-sponse.

Bacterial and viral pathogens werenot detected and there was no evidenceof horizontal disease transmission innormal litter mates. However, normalluminal bacteria were implicated by noclinical or histological evidence of coli-tis in three IL-2 deficient germ-freemice followed for five months in a ster-ile environment. KO mice raised under

specific pathogen-free conditions de-veloped no clinical signs of colitis, buthad evidence of mild histological andimmunological abnormalities by 17 to20 weeks of age.

IL-10 DEFICIENT MICEIL-10, a product of TH2 lymphocytes,

is believed to have predominantly im-munosuppressive activity by virtue ofits ability to downregulate macrophagenatural killer cells, lymphocyte activa-tion and cytokine production (13). IL-

10 deficient mice spontaneously de-velop chronic inflammation of theirentire intestinal tract, with the mostsevere inflammation in the duodenum,proximal jejunum and right colon (15).These mice develop a progressivewasting syndrome with anemia whichbegins by three to four weeks of age andresults in 30% mortality by five to sixweeks. Male and female mice areequally affected. The proximal smallbowel is markedly thickened due tomucosal and crypt hyperplasia. Thecrypt architecture is deranged anddysplastic with a decrease in Peyer’spatches. Superficial erosions arepresent with exudate. Away from ulcer-ated areas the lamina propria is infil-trated by mononuclear cells, neutro-phils and occasional multinucleatedgiant cells. Inflammation diminishes inthe distal small bowel, with atrophy ofthe ileal mucosa. Colonic lesions arepredominantly atrophic. Liver histol-ogy is normal but the bone marrowdemonstrates an increased myeloid:erythroid ratio.

The majority of plasma cells in in-flamed intestine produced IgA or IgG1,and expression of MHC class II mole-

cules was increased on small intestinaland colonic epithelial cells. SerumIgG1 and IgA levels were increased inIL-10 –/– mice. The antibody responseto immunization with a T cell-dependent antigen was similar in IL-10

deleted and control mice. However, invivo nippostrongylus infection stimu-lated a TH1 (increase in interferon[IFN]-gamma) lymphokine response inIL-10 –/– mice versus the usual TH2 (IL-4,IL-5) response in controls. Interest-ingly, spleen cells from the KO miceproduced 20-fold more IL-6 and TNF-�than normal mice following LPS stimu-lation. These results suggest that defi-cient production of IL-10, whichsuppresses activation and cytokinesynthesis by macrophages, natural kil-ler cells and T cells, could lead tochronic stimulation by luminal bacte-rial components. This hypothesis issupported by the observation that IL-10

deficient mice reared under specificpathogen-free conditions have onlyisolated left-sided colitis with no histo-logical evidence of small intestinal in-flammation.

�� TCR DEFICIENT MICEThe TCR is a heterodimer compris-

ing either � + � or � + � subunits. Anti-gen binding to the TCR complex inconjunction with accessory ligandbinding and costimulatory cytokines isrequired for antigen-specific activationof T lymphocytes. Mutation of variouscomponents of the TCR has been ele-gantly studied by Mombaerts et al (16).Deletion of the � or � chains of the TCR

in mice induces spontaneous colitis,with onset by three to four months ofage and mortality by six to 12 months.Mice develop nonbloody diarrhea,weight loss and rectal prolapse due to apancolitis which affects the left sidemore than the right side. The colon isquite thickened due to massive crypthyperplasia, goblet cell depletion,acute and chronic inflammation of thelamina propria and crypt abscesses.Macroscopic mucosal ulcerations andrectal bleeding are not seen. The smallintestine is grossly and histologicallynormal, but mesenteric lymph nodesare enlarged and focal hepatic inflam-

TABLE 3Pathogenesis of colitis in �� TCR deficient mice

Strain CD4+ �� T cells B cells Colitis

Wild type + + No

TCR � KO + + No

Nude – + No

RAG-1 KO – – No

TCR � KO – + Yes

TCR � KO – + Yes

TCR � KO x � KO – + Yes

Class II MHC KO – + Yes

KO Knock out; MHC Major histocompatibility complex; RAG Recombination-activating gene; TCR T

cell receptor


SARTOR

matory cell infiltrates are present insome mice.

The authors examined multiple de-letion constructs to investigate mecha-nisms of immunodeficiency leading tocolitis in this model (Table 3). Thecommon denominator in mice thatdeveloped colitis is the absence of ��

T lymphocytes and the presence offunctional B cells. An important clue isthe lack of colitis in mice with mutatedrecombination-activating genes(RAG-1) and in severe combined immu-nodeficient (SCID) mice, which are to-tally deficient in mature T and Blymphocytes. These results suggest thatspontaneous colitis in this model is theresult of lack of appropriate �� T cell-mediated suppression of B cells. Analy-sis of the �� TCR mutation in severalinbred mouse strains demonstratedthat 129/Sv and C3H/He were permis-sive and Balb/c were more resistant.

LYMPHOCYTES SUBSETRECONSTITUTED

IMMUNODEFICIENT MICEThe SCID mouse and nude rat can be

populated with immunocytes withoutrejection, which permits the in vitrostudy of the activity of various T cellsubpopulations. Based on studies byPowrie and Mason (17) in nude rats,several groups have described colitis re-sulting from injection of T lymphocytesubsets divided on the basis of bindingdensity of the monoclonal antibodyCD45RB into SCID mice (18,19).CD45RBhi lymphocytes have beenshown to secrete profiles of TH1 lym-phokines whereas CD45lo cells secreteTH2 lymphokines. Injection of unfrac-tionated cells or combined CD45RBhi

and CD45RBlo subsets caused no disease.However, injection of only CD45RBhi

lymphocytes induced chronic wastingbeginning 10 to 30 days after reconsti-tution with 30% mortality at 12 weeks.Intestinal inflammation is predomi-nantly in the colon (left side more af-fected than right side) but was mild inthe small intestine. Histological exami-nation of the small intestine revealedfocal mononuclear infiltrates in thelamina propria. Colon inflammationwas characterized by focal ulceration,chronically thickened mucosa with

crypt hyperplasia and a mononuclearcell infiltrate of the lamina propriawith occasional neutrophils and trans-mural inflammation. Mild periportalinfiltrates were seen in the liver.

IFN-� mRNA tissue concentrationswere substantially higher than normalvalues in SCID mice reconstituted withCD45RBhi cells, but IL-10 levels were notincreased (19). IL-4 levels were unde-tectable in all mice, and TNF-� was ele-vated in reconstituted mice with andwithout colitis. These results indicatethat CD45lo lymphocytes can suppressintestinal inflammation, which is me-diated by the reciprocal CD45hi popula-tion, and suggest that a disregulationbetween TH1 and TH2 lymphokine pro-duction may contribute to chronic in-testinal inflammation. These studiesgraphically illustrate the essential roleof competent immunosuppressive cellsin homeostasis of the intestinal mu-cosa.

TRANSFORMING GROWTHFACTOR-BETA

DEFICIENT MICETransforming growth factor

(TGF)-beta is a macrophage productthat has pleiotropic immunological ac-tivities including monocyte chemo-taxis and stimulation of collagen syn-thesis by intestinal smooth musclecells and fibroblasts (13). An impor-tant immunoregulatory activity of thismolecule is suppression of lymphocyteproliferation. TGF-�1 KO mice have in-creased in utero mortality, but appearto develop normally until weaning(three weeks of age), when they de-velop a wasting syndrome and systemicinflammation consisting of myocar-ditis, bronchoalveolar inflammation,pancreatitis and dermatitis (20,21).One of the two reported TGF-�1 defi-cient lines develops diarrhea and hasincompletely described colitis (20).The mechanisms of chronic systemic

inflammation and colitis in this modelremain obscure, but may be related todefective suppression of lymphocytes.Of interest, many of the organs in-volved are colonized by ubiquitous bac-teria.

SUMMARY ANDCONCLUSIONS

Chronic intestinal inflammationthat develops in response to spontane-ous gene mutations, targeted deletionor overexpression of specific genes orselective manipulation of lymphocytessubsets can provide important insightsinto the mechanisms of intestinal in-flammation (22). Several broad immu-nological mechanisms that mediatespontaneous colitis are listed in Table4. The similarities of chronic intestinalinflammation in these models to idio-pathic ulcerative colitis and Crohn’sdisease suggest that similar mecha-nisms may be active in human IBD.

A theme common to many of thesemodels is that intestinal inflammationdevelops as a result of defective immu-nosuppression. Models of TGF-�1 KO,�� TCR deletion and CD45RBhi recon-stitution of SCID mice clearly demon-strate the consequences of ineffectiveT lymphocyte suppression and suggestthat T cell suppression is critical to nor-mal mucosal homeostasis. The disor-dered balance between TH1 and TH2lymphokine profiles in IL-10 KO andCD45RBhi

� SCID mice further docu-ments the importance of effective im-munoregulation by lymphocytes. Howthese animal studies relate to the pro-posed TH1 profile of Crohn’s diseaseand TH2 for ulcerative colitis remainsto be determined, but it is of interestthat IL-10 –/– and reconstituted SCID

mice have small intestinal inflamma-tion and TH1 lymphokine profiles. Evi-dence from TCR deletion studies andIL-2 KO mice strongly suggest that in theabsence of regulatory T cells, uncon-

TABLE 4Immunological mechanisms of spontaneous colitis in genetically engineered ro-dents

Defective T cell suppression: CD45RBhi

� SCID, ��TCR –/–, TGF-�1 –/–

� TH1/TH2 lymphokines: IL-10 –/–, CD45RBhi

� SCID

� B cell response: IL-2 –/–, ��TCR –/–

IL Interleukin; SCID Severe combined immunodeficient; TCR T cell receptor; TGF Transforming growth factor;

TH T helper



trolled B lymphocyte activation is ca-pable of mediating mucosal damage.

A second concept is that specificgene mutations can cause spontaneousintestinal inflammation and that clini-cally and histologically similar lesionscan arise from different immunoregula-tory defects. These observations illus-trate the limited repertoire of mecha-nisms of tissue injury, suggest that avariety of immunoregulatory defectsmay induce human IBD and imply thatthere may be multiple etiologies forthese disorders. Moreover, the etiologi-cal heterogeneity among ulcerative co-litis and Crohn’s disease patients maybe responsible for clinical subsets of pa-tients with predictable patterns of dis-ease, complications and response tomedications.

A third concept is that host geneticsusceptibility determines aggressive-ness, chronicity and complications ofintestinal inflammation. The effect ofgenetic susceptibility is illustrated bythe C3H/HeJ Bir mouse and differen-tial responses to experimental intesti-nal inflammation in Lewis versusFischer rats in the PG-PS andindomethacin models (3,23), andC3H/ He or 129/Sv versus Balb/c micein the �� TCR KO model (16). Identifi-cation of genetic mechanisms of defec-tive immunoregulation in susceptiblerodents could direct the search for ge-netic susceptibility factors in IBD pa-tients.

Fourth, transfer studies in HLA-

B27/�2� transgenic rats demonstratethat bone marrow-derived cells canmediate chronic intestinal and sys-

temic inflammation. The inflamma-tory response of these cells is geneti-cally programmed because progenitorcells can be derived from fetuses that donot exhibit evidence of intestinal orsystemic inflammation. Moreover, dis-play of HLA-B27 on host epithelial cellsis not sufficient for disease induction.This observation suggests that in-creased expression of MHC moleculesduring experimental or idiopathic in-flammation is a consequence of the in-flammatory reaction and is notnecessary for chronic intestinal inflam-mation. Resolution of spontaneous in-testinal inflammation in HLA-B27 ratsafter irradiation and successful bonemarrow engraftment from nontrans-genic rats raise the possibility of thera-peutic bone marrow transplantation inIBD patients.

Finally, attenuation of spontaneousintestinal inflammation in IL-2 and IL-

10 deficient mice raised in a pathogen-free environment and lack of evidenceof colitis in IL-2 KO mice and HLA-

B27/�2� mice living in germ-free (ster-ile) conditions dramatically illustratethe critical pathogenetic role of ubiqui-tous luminal bacterial constituents.These results parallel those observed inindomethacin-induced enterocolitiswhere gnotobiotic Lewis rats populatedwith specific pathogen-free flora de-velop chronic mid-small bowel ul-cers but germ-free littermates have noevidence of chronic inflammation(24). Because microbiological condi-tions vary between each rodent facility,direct comparisons of the clinical fea-tures of each spontaneous colitis model

must be made under similar conditions,optimally using animals colonized withthe same flora. It is unknown whetherspontaneous colitis develops as the re-sult of luminal bacteria invading tissuesof immunocompromised hosts (IL-2 andTCR KO mice), as a hyperactive re-sponse to bacterial components (LPS,PG-PS) in rodents that lack effectiveimmunosuppressor cytokines (IL-10,TGF-� KO mice) or as an exaggeratedantibody response to ubiquitous bacte-rial antigens (IL-2, TCR KO models andCD45hi

� SCID mice).These genetically engineered ro-

dent models have only recently beendescribed so their usefulness has beenonly superficially exploited. Similartypes of spontaneous intestinal inflam-mation will almost certainly be discov-ered as new KO and transgenic rodentsare generated to study other immuno-regulatory molecules. Application ofsophisticated molecular and immuno-logical techniques and further geneticmanipulation by selective crossbreed-ing with well characterized mousestrains should generate important in-sights and new enthusiasm to exploremechanisms of chronic intestinal in-flammation.

ACKNOWLEDGEMENTS: This workwas supported by United States PublicHealth Service grants DK 40249, DK 47700and DK 34987 and the Crohn’s and ColitisFoundation of America. The authorgratefully acknowledges the expert secretarialassistance of Robert Tuttle and Shirley Wil-lard.

REFERENCES1. Sartor RB. Animal models of intestinal

inflammation: Relevance to IBD.In: MacDermott RP, Stenson W, eds.Inflammatory Bowel Disease. NewYork: Elsevier, 1991:337-54.

2. Stenson WF. Animal models ofinflammatory bowel disease. In: TarganSR, ed. Inflammatory Bowel Diseasefrom Bench to Bedside. Baltimore:Williams & Wilkins, 1992:180-92.

3. McCall RD, Haskill S, Zimmerman E,et al. Tissue IL-1 and IL-1 receptorantagonist expression in enterocolitisin resistant and susceptible rats.Gastroenterology 1994;106:960-72.

4. Sundberg JP, Elson CO, Bedigian H,et al. Spontaneous, heritable colitis in

a new substrain of C3H/HeJ mice.Gastroenterology 1994;107:1726-35.

5. McCabe RP, Mills T, Ridwan B, et al.Immune reactivity of C3H/HeJ micewith spontaneous colitis.Gastroenterology 1993;104:A656.

6. Nickerson CI, Luthra HS, David CS.Role of enterobacteria and HLA-B27in spondyloarthropathics: Studies withtransgenic mice. Ann Rheum Dis1990;49:426-33.

7. Hammer RE, Maika SD, RichardsonJA, Tang J, Taurog JD. Spontaneousinflammatory disease in transgenic ratsexpressing HLA-B27 and human �2�:An animal model of HLA-B27-associated human disorders. Cell1990;63:1099-112.

8. Taurog JD, Maika SD, Simmons WA,Breban M, Hammer RE. Susceptibilityto inflammatory disease in HLA-B27transgenic rat lines correlates with thelevel of B27 expression. J Immunol1993;150:4186.

9. Breban M, Hammer RE, RichardsonJA, Taurog JD. Transfer of theinflammatory disease of HLA-B27transgenic rats by bone marrowengraftment. J Exp Med1993;178:1607-16.

10. Taurog JD, Richardson JA, Croft JT, etal. The germfree state preventsdevelopment of gut and jointinflammatory disease in HLA-B27transgenic rats. J Exp Med1994;180:2359-64.


SARTOR

11. Rath HC, Bender DE, Grenther T,et al. Normal bacteria stimulatecolonic, gastric and systemicinflammation in HLA-B27/�2�transgenic rats. Gastroenterology1995;108:A899. (Abst)

12. Rath HC, Bender DE, Holt LC, et al.Metronidazole attenuates colitis inHLA- B27/�2� trangenic (TG) rats:A pathogenic role for anaerobicbacteria. Clin Immunol Immunopathol1995;76:545.

13. Sartor RB. Viewpoints on digestivediseases. Cytokines in intestinalinflammation: Pathophysiologicaland clinical considerations.Gastroenterology 1994;106:533-9.

14. Sadlack B, Merg H, Schorle H, et al.Ulcerative colitis-like disease in micewith a disrupted interleukin-2 gene.Cell 1993;75:253-61.

15. Kuhn R, Lohler J, Rennick D,Rajowsky K, Muller W. Interleukin-10-deficient mice develop chronicenterocolitis. Cell 1993;75:263-79.

16. Mombaerts P, Mizoguchi E, Grusby M,et al. Spontaneous development of

inflammatory bowel disease in T cellreceptor mutant mice. Cell1993;75:275-82.

17. Powrie F, Mason D. Ox-22high CD4+ Tcells induce wasting disease withmultiple organ pathology: Preventionby the Ox-22low subset. J Exp Med1990;172:1701-8.

18. Morrisey PJ, Charrier K, Braddy S,Liggitt D, Watson JD. CD4+T cells that express high levels ofCD45RB induce wasting diseasewhen transferred into cogenicsevere combined immunodeficientmice. Disease development isprevented by cotransfer of purifiedCD4+ T cells. J Exp Med1993;178:237-44.

19. Powrie F, Leach MW, Mauze S, CaddleLB, Coffman RL. Phenotypicallydistinct subsets of CD4+ T cells induceor protect from chronic intestinalinflammation in C B-17 SCID mice.Int Immunol 1993;5:1461-71.

20. Kulkarni AB, Huh CG, Becker D,et al. Transforming growth factor �1null mutation in mice causes excessive

inflammatory response and early death.Proc Natl Acad Sci USA1993;90:770-4.

21. Shull MM, Ormsby I, Kier AB, et al.Targeted disruption of the mousetransforming growth factor-�1gene results in multifocalinflammatory disease. Nature1992;359:693-9.

22. Strober W, Ehrhardt RO. Chronicintestinal inflammation: Anunexpected outcome in cytokine or Tcell receptor mutant mice. Cell1993;75:203-5.

23. Sartor RB, Bender DE, Holt LC.Susceptibility of inbred rat strains tointestinal and extraintestinalinflammation induced byindomethacin. Gastroenterology1992;102:A690.

24. Sartor RB, Bender DE, Grenther T,Holt LC. Absolute requirement forubiquitous luminal bacteria in thepathogenesis of chronic intestinalinflammation. Gastroenterology1994;106:A767.


Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

lessons in ibd pathogenesis from new animal models of...

Documents