www.e l sev i e r. com/ loca te / i n t imp
International Immunopharmacology (2008) 8, 20–27
Dextran sulphate sodium induces acute colitisand alters hepatic function in hamstersAgneta Karlssona, Åke Jägervall b, Madeleine Petterssonb,Ann-Katrin Anderssonb, Per-Göran Gillberg a, Silvia Melgar a,⁎
a Department of Integrative Pharmacology, AstraZeneca R&D Mölndal, Swedenb Department Molecular Pharmacology, AstraZeneca R&D Mölndal, Sweden
Received 5 July 2007; received in revised form 1 October 2007; accepted 1 October 2007
⁎ Corresponding author. AstraZenecaIntegrative Pharmacology, GI Biology,Tel.: +46 317065142; fax: +46 3177637
E-mail address: silvia.melgar@astra
1567-5769/$ - see front matter © 200doi:10.1016/j.intimp.2007.10.007
Abstract
Dextran sulphate sodium (DSS)-induced colitis in rodents is an experimental model for humaninflammatory bowel disease (IBD). The aim of this study was to characterize the effect of DSS inhamster colon and liver. DSS (2–5%) was administrated in the drinking water for 4–6 days. Clinicalsymptomswere recordeddaily, inflammatory and fatty acid-relatedmetabolicmarkerswere assessedin plasma, colon and liver. Six days of 3 or 5% DSS induced a severe wasting disease, whereas 2.5% DSSinduced a colonic inflammation without severe systemic adverse effects. The systemic inflammatoryresponse was characterized by an inverse production of albumin and the acute phase proteinhaptoglobin. The colonic inflammatory response was confined to the proximal colon, manifested by ahigh macroscopic inflammatory score, increased colon weight and expression of IL-1β, IL-6 and iNOS,infiltration of inflammatory cells and epithelial disruption. In contrast, only a low/mild inflammatoryresponse was observed in the distal colon of DSS-exposed hamsters. Significant hepatic-relatedmetabolic alterations were also observed, with elevation of plasma triglycerides and increased liverexpression of lipoprotein lipase and reduced expression of acyl-CoA oxidase and cytochrome P450A.Although liver weight was significantly reduced, no histopathological signs of inflammation or tissuedamage were observed. In summary, hamsters exposed to 2.5% DSS for 6 days develop acute colitisresembling murine DSS-induced colitis. In addition, DSS-exposed hamster showed alterations inhepatic fatty acids metabolism resembling human IBD, suggesting that the model can potentially beused for target discovery and validation of hepatic-related metabolic alterations.© 2007 Elsevier B.V. All rights reserved.
KEYWORDSDSS;Hamster;Hepatic function;Inflammatorybowel disease;Triglycerides;Lipoprotein lipase
R&D Mölndal, Department ofSE-431 83 Mölndal, Sweden.47.zeneca.com (S. Melgar).
7 Elsevier B.V. All rights reserved.
1. Introduction
Inflammatory bowel disease (IBD) encompasses two majorchronic diseases, ulcerative colitis (UC) and Crohn's disease(CD), which affects the gastrointestinal tract in humans. Flaresof remission and relapses with symptoms of bloody diarrhea,
mailto:[emailprotected]
21Acute colitis in hamsters
abdominal pain and rectal bleeding characterize the disease.The etiology of IBD is still unknown, but the pathogenesis islikely dependent on the interaction between local immunereactions and environmental factors in genetically susceptibleindividuals [1].
To this date, there are no specific animal models for eitherUC or CD. Instead, more than 30 experimental models havebeendeveloped, that reflect different features of the intestinalinflammation [1]. Among these, the dextran sulphate sodium(DSS) induced colitis in rodents is widely used, due to theconvenient induction of intestinal inflammation (DSS in drinkingwater ad libitum) and good reproducibility. The DSS-inducedcolitis in rodents resembles UC in humans due to the clinicalsymptoms, inflammatory markers and histopathologicalchanges [2,3]. The mechanism(s) by which DSS induces colitisis still unknown although, it has been suggested that it disruptsthe mucosal barrier function and/or alters the response ofmacrophages [2,4]. In rodents, the outcome of DSS-inducedcolitis is highly dependent on the genetic background and theinflammatory response is more prominent in the distal colon,although in some strains the inflammation can be found in theproximal colon or the cecum [3,5].
In addition to gastrointestinal alterations, IBD patients arelikely to develop extraintestinal alterations involving e.g., thejoints, the liver and the skin [6]. The occurrence of inflamma-tion in the joints or skin seems to correlate to the activity of gutinflammation, as the control of intestinal inflammation leads toits resolution. However, liver complications are generally notdependent on the activity of gut inflammation [6]. Few reportshave considered the effect that colitis or DSS-inducedinflammation may exert on the liver, considering inflammatoryresponses, changes in fatty acid metabolism or assessing liverinjury. There are, however, several reports on changes in lipidmetabolism, such as hypertriglycemia, accelerated lipolysis ordecreased hepatic fatty acid oxidation, that occurs during theacute phase response upon infection, inflammation or injury[7,8].
In the present study, we have characterized the pro-inflammatory effects of DSS in hamsters and its potentialinterest as a new IBD model. For this purpose, we assessedclinical symptoms, local (colonic) and systemic inflammatorymediators and colonic histology in hamsters exposed to DSS. Inaddition, systemic inflammatory responses and hepatic func-tionality were also assessed.
2. Materials and methods
2.1. Animals
Specific pathogen free, male Syrian Hamsters (Mesocricetus auratus,6–8weeks old,weighting 80–100 g)were obtained fromHarlan (Horst,The Netherlands). The hamsters were housed individually and kept atthe animal house facilities at AstraZeneca R&D Mölndal, Sweden,under controlled environmental conditions (50% humidity; 12:12 hlight:dark cycles) and fed with a standard pellet diet (R3 pellets,Lactamin, Sweden) and tap water ad libitum. Hamsters wereacclimatized for at least 2 weeks before entering the study. Thepresent study was approved by the Local Animal Research BoardCommittee (No. 66-2006, Göteborg, Sweden).
2.2. Induction of colitis
Dextran sulphate sodium (45 kD;TdBConsultancyAB,Uppsala, Sweden)was added to tap water at different concentrations (2 to 5% (w/v)) and
given to the hamsters for 4 to 6 days. Fresh DSS solutionswere prepareddaily and the body weight and the general health condition, includingfecal consistency and fecal bleeding, were recorded daily. The totalnumber of animals included in the 2.5% DSS exposed group was 12–27hamsters and in the healthy control group 4–16 hamsters.
2.3. Tissue and plasma sampling
Colon and liver tissue and plasma samples were collected as spe-cified. At the end of each time point, animals were anesthetized byinhalation of isoflurane (Abbot Scandinavia AB, Solna, Sweden),blood was drawn by retroorbital puncture, and euthanasia wasapplied thereafter. Blood was collected in tubes containing EDTA orheparin; plasma was obtained by centrifugation, frozen and kept at−80 °C until analyses. The intestines were excised and carefullyrinsed with saline. The colon was cut in proximity to the ileocecalvalve and the rectum, and the length was measured. The colon wasdivided into three pieces, distal, middle and proximal. The colonpieces were opened and divided into 2 longitudinal sections, whereone piece was rolled as a “Swiss roll”, fixed in Zinc-formalin solution(pH 7.4, Histolab Products AB, Göteborg, Sweden) and embedded inparaffin for histology analysis and the second piece was frozen inliquid nitrogen for RNA analysis. A piece of liver was collected fromeach animal, and divided into two parts, one piece used for histolo-gical evaluation and the other piece used for RNA analysis as de-scribed for the colonic tissue.
2.4. Assessment of inflammation
Clinical assessment of inflammation included daily monitoring of bodyweight and general health condition, as previously mentioned. At nec-ropsy, the macroscopic appearance of the colon (inflammatory score)was evaluated after the same criteria used to evaluate DSS-inducedcolonic inflammation in mice [9]. Three cm of the most distal andproximal colon were evaluated according to: stiffness (scale 0–2),edema (scale 0–3), visible ulcerations (0–1) and thickness (0–4). Basedon these scores an overall inflammatory score was obtained, with themaximal score being 10 [9]. In addition, the stool consistency (diarrheascore) and visible fecal bloodwas scored separately on a scale of 0 to 3,as described in the mouse DSS model [3].
2.5. Histology
Five μm thickness tissue sections of distal, middle and proximalcolon were stained with hematoxylin/eosin (H&E) to evaluate thedegree of inflammation. The stained tissue was analyzed in a blindedfashion by A-K. A. and M.P. using a standard microscope (Zeiss,Germany). The pathophysiology of the tissue was characterized bythe presence of ulcerations, inflammatory cells (polymorphonuclear,mononuclear and plasma cells), signs of edema, crypt loss, surfaceepithelial cell hyperplasia, goblet cell reduction and signs ofepithelial regeneration. Similar histology evaluation was performedon H&E stained liver sections.
2.6. Analysis of local and systemic inflammatory markers
The levels of the acute phase proteins haptoglobin and serum amy-loid A (SAA) were determined using a Cobas Bio centrifugal analyzerwith a commercial reagent kit for haptoglobin (Tridelta TP 801)and a commercial solid phase sandwich ELISA for SAA (TrideltaDevelopment, Ltd, Ireland) according to the manufacturer's ins-tructions [3].
Messenger RNA expression of IL-1β, IL-6, TNF and iNOS was de-termined on 3 cm distal and proximal colonic tissue hom*ogenate andliver tissue hom*ogenates using real time PCR (Taqman®). Samplesfrom colon and liver were hom*ogenized with Mixer Mill (RetschGmbH, Haan, Germany) in Trizol (Invitrogen AB, Lidingö, Sweden),
Table 1 Primers used in the study
Marker Forward primer Reverse primer Comments
ACO 5′GGAGATGCAGCTCGGTGTCT 3′CCAAAATCTGTGGTTCTGGTTCACYP4 5′CATGGCCTTCCGTGTTCCTA 3′CTTTCTCCAGGCGACATGTGAIL-1β 5′GGCTGATGCTCCCATTCG 3′CACGAGGCATTTCTGTTGTTCAIL-6 5′TGGAACTTCCGGTGATACAAATAAATGA 3′CATTGTTCGTCACAAACTCCAGGTAGATiNOS 5′CCTGCCAGCTTGGAGTTCAC 3′ATCGAAGCGGCCATAGCLPL 5′TTTAACTACCCCCTGGACAATGTC 3′ACCTTCTTGTTGGTCAGACTTCCT Applied from [30]TNF 5′GCCTCTTCTCCTTCCTGCTT 3′ATGGAGCCGATGATAGGGTTβ-actin 5′GCACAGGCCTTTCGCAGCTCTTTCTTC 3′CGTCATCCATGGCGAACTGGTG Applied from [31]
22 A. Karlsson et al.
followed by total RNA isolation, according to the manufacturesinstructions (Applied Biosystems, Stockholm, Sweden). Five hundredng of DNAase treated total RNA was used for each cDNA synthesis,using the High capacity cDNA archive kit (Applied Biosystems). Eachsample was run in triplicates with 4 ng of template in each reaction.Data was normalized using β-actin as internal standard, and ex-pressed as fold change relative to healthy controls. The primerswere designed from cDNA sequences in Primer Express (AppliedBiosystems) and purchased from Operon (Cologne, Germany). Theprimers used in the study are depicted in Table 1. Whenever pos-sible, the primers were placed in the intron–exon junction.
Figure 1 Clinical symptoms and macroscopic colonic inflammatorexposure. a) Body weight was recorded daily and the body weightweight (day 0). b) At the day of necropsy, the presence of diarrhea aMaterials and methods. c) The colonic inflammatory score was based(0–1) and thickness (0–4) obtaining a maximal score of 10. The finproximal and distal colon. d) The weight (mg) of 3 cm of the proximalNumber of DSS-exposed hamsters n=26–27; healthy hamsters n=15
2.7. Detection of markers of fatty acid metabolism in liverand colon hom*ogenates
The lipid metabolic status was determined by assessing the hepa-tic and colonic level of expression of several enzymes involved infatty acid metabolism, namely lipoprotein lipase (LPL), acyl-CoAoxidase (ACO) and cytochrome P450 (CYP4A). Expression levels weredetermined on liver and colon tissue hom*ogenates using real timePCR (Taqman) as described above. Primers used are depicted inTable 1.
y markers in hamsters with acute colitis after 6 days of 2.5% DSSchange was calculated as the percent change from the startingnd visible fecal blood was scored in a scale of 0–3 as described inon: stiffness (scale 0–2), edema (scale 0–3), visible ulcerations
al score was based on the assessment of a 3 cm segment of theand distal colon was assessed as a marker of inflammation. (a–d)–16. ⁎⁎⁎ pb0.001 between DSS-exposed and healthy animals.
Table 2 Changes in plasma markers during DSS-inducedacute colitis
Plasma marker a Healthy b DSS c
Triglycerides (Tg) (mM) 1.7±0.3 6.0±0.9 ⁎⁎⁎
Cholesterol (mM) 3.1±0.1 3.8±0.3Glucose (mM) 8.4±1.0 10.3±0.8Alanin aminotransferase (ALAT)(μkat/L)
1.2±0.2 0.9±0.5
Aspartate aminotransferase(ASAT) (μkat/L)
1.2±0.2 1.4±0.4
Albumin (g/L) 36.7±0.5 26.4±1.1 ⁎⁎⁎
Haptoglobin (g/L) 0.4±0.03 d 0.9±0.1 e
a Plasma was collected from healthy and DSS-treated hamstersand analyzed for the seven markers as described in Materialsand methods. The values are expressed as mean±SEM.b n=4 for Tg, cholesterol, glucose, ALAT, ASAT.c n=8 for Tg, cholesterol, glucose, ALAT, ASAT.d n=16 for healthy hamsters.e n=27 for DSS-treated hamsters.⁎⁎⁎ pb0.001 compared to healthy animals.
23Acute colitis in hamsters
2.8. Analysis of plasma markers of hepatic functionality
The plasma levels of triglycerides (Tg), cholesterol, glucose, alanineaminotransferase (ALAT), and albumin were assayed using commer-cial available systems as previously described [10]. Aspartate ami-
Figure 2 Colonic expression of inflammatory markers in hamstersexpression of a) IL-1β, b) IL-6, c) iNOS and d) TNF was analyzed in theColonic hom*ogenates were prepared and the mRNA expression anaexpressed as fold change relative to healthy controls, mean±SEM.healthy hamsters n=12–14. In distal colon: number of DSS-exposed
notransferase (AST) activity in plasma was measured by using acommercial reagent system (AS 7905, Randox Laboratories Ltd,United Kingdom) according to the manufacturer's instructions.
2.9. Statistics
Data are presented as mean±S.E.M. Mann–Whitney's U-test wasused for statistical analysis; a value of pb0.05 was considered sig-nificant (Graph Pad Software, Inc).
3. Results
3.1. Exposure to 2.5% DSS for 6 days induces clinicalsymptoms and colonic and systemic inflammation inhamsters
Pilot studies showed that hamsters exposed to 5% or 3% DSS for 4–6days developed a severe wasting disease with severe body weightloss (N22%), hunched posture and bloody diarrhea. A reduction inthe concentration of DSS to 2%, given to the hamsters for 5 days,resulted in a milder inflammatory response compared to 3 and 5%DSS exposure. Moreover, the response was rapidly attenuatedwhen the 2% DSS was removed after 5 days exposure and the ham-sters were allowed to drink water for 2 or 5 days. Based on theseobservations, in subsequent experiments animals were exposedto 2.5% DSS for 6 days. The 2.5% DSS-induced response wascharacterized by an affected general health condition with body
with acute colitis after 6 days of 2.5% DSS exposure. The mRNAproximal and distal colon of healthy and DSS-treated hamsters.lyzed by PCR as described in Materials and methods. Data areIn proximal colon: number of DSS-exposed hamsters n=19–26;hamsters n=12; healthy hamsters n=8.
Figure 3 Histology sections of inflamed proximal colon inhamsters with acute colitis after 6 days of 2.5% DSS exposure.Representative H&E tissue sections of proximal (a, c, e, g) anddistal colon (b, d, f, h) from a hamster exposed to 6 days of 2.5%DSS (a–d) and a healthy hamster (e–h). Inset in (a–b) and (e–f)are shown in a higher magnification in (c–d) and (g–h). The colonis rolled as “Swiss role” with the most proximal part of the colonsegment localized to the middle of the role. Original magnifica-tion 1.25× (a–b; e–f); 10× (c–d; g–h).
Figure 4 Hepatic markers in hamsters with acute colitis after6 days of 2.5% DSS exposure. a) The weight (g) of the whole liverwas assessed at the time of necropsy fromhealthy andDSS-treatedhamsters. b) The liver mRNA expression of pro-inflammatorymediators IL-1β (IL-1), IL-6, iNOS and TNF and (c) the fatty acidenzymes lipoprotein lipase (LPL), acyl-CoA oxidase (ACO) andcytochromeP450 (CYP4A)was analyzed in liver of healthy andDSS-treated hamsters. Liver hom*ogenates were prepared and themRNA expression analyzed by PCR as described in the Materialsand methods. Data are expressed as fold change relative tohealthy controls, mean±SEM. (a) Number of DSS-exposed ham-sters n=26; healthy hamsters n=15. (b–c) Number of DSS-exposedhamsters n=9–12; healthy hamsters n=5–8.
24 A. Karlsson et al.
weight loss along the progression of the disease and with asignificant appearance of diarrhea/loose feces with visible fecalblood at the day of termination (Fig. 1a–b). A systemicinflammatory response could also be demonstrated by anelevation in the acute phase protein haptoglobin (p=0.07 in DSSvs. healthy animals; Table 2) and in SAA (healthy animalsb11.8μg/ml (lower detection level); DSS-treated hamsters N190 μg/ml(maximal detection level)). In addition, albumin, a negative acutephase protein, was also significantly decreased in DSS-treatedhamsters compared to healthy animals (Table 2).
3.2. Increased pro-inflammatory cytokines and infiltrationof inflammatory cells in the proximal colon of hamstersexposed to 2.5% DSS for 6 days
Macroscopic examination of the colon at the time of necropsyindicated that the colonic inflammatory response was locatedpredominantly in the proximal colon, with a significantlyincreased inflammatory score and colon weight, while only amild effect, although still significant, was observed in the distalarea of the colon (Fig. 1c–d). The length of the colon in DSS-treated hamsters was also significantly reduced (18.4±0.6 cm)compared to healthy animals (28.2±0.7 cm; pb0.001). Wefurther analyzed the mRNA expression of the pro-inflammatorycytokines IL-1β, IL-6, TNF and iNOS in both the distal and the
proximal colon. As described for the macroscopic inflammatoryscore, a significantly higher expression of IL-1β, IL-6 and iNOSwas found in the proximal colon compared to the expressionlevels in healthy animals (Fig. 2a–c). However, in the distal co-lon, regardless the presence of a macroscopic moderate in-flammation, none of the pro-inflammatory cytokines assessed
Table 3 mRNA expression of fatty acid enzymes in colontissue of hamsters with DSS-induced acute colitis
Marker Proximal colon a Distal colon a
Healthy b DSS c Healthy d DSS e
LPL 1.64±0.44 0.55±0.12 ⁎⁎ 2.16±1.54 0.42±0.58ACO 0.66±0.13 0.44±0.09 1.22±0.82 0.22±0.06CYP4A 1.03±0.19 1.11±0.15 0.75±0.14 0.97±0.24a Three cm of proximal and distal colon was collected from
healthy and DSS-treated hamsters, hom*ogenized and RNAextracted for analysis of the expression of lipoprotein lipase(LPL), acyl-CoA oxidase (ACO) and cytochrome P450 (CYP4A) asdescribed in Materials and methods. Data are expressed as foldchange relative to healthy controls, mean±SEM.b n=11–12/marker.c n=19/marker.d n=8/marker.e n=12–13/marker.⁎⁎ p=0.01 compared to healthy animals.
25Acute colitis in hamsters
was elevated (Fig. 2a–c). The mRNA expression of TNF was onlymarginally increased in the proximal colon, without reachingstatistical significance (p=0.282 in DSS-exposed vs. healthy ham-sters; Fig. 2d).
Furthermore, the inflammatory responses were evaluated byhistology using H&E stained tissue sections of the proximal,middle and distal colon. In the proximal colon, the mucosashowed a severe inflammatory response with infiltration ofinflammatory cells often found in closed proximity to ulcera-tions, and luminal and crypt cell degeneration as compared tohealthy animals (Fig. 3a,c,e,g). In addition, a moderateinfiltration of inflammatory cells was found along the proximalcolon, with focal mild infiltrations of polymorphonuclear cellsand eosinophils in the muscle layer. The middle colon was partlyaffected with a focal infiltration of inflammatory cells in themucosa accompanied by epithelial degeneration, whereas thesubmucosa and muscle layers were only mildly affected (datanot shown). Finally, no/very low infiltration of inflammatorycells and no epithelial disruption was observed in the distal colonof DSS-exposed hamsters compared to healthy animals (Fig. 3b,d,f,h).
3.3. DSS-exposure elevates plasma triglycerides and liverLPL expression, decreases liver weight but does not induceliver pro-inflammatory cytokines
Liver weight was significantly reduced in animals exposed to 6days of 2.5% DSS (Fig. 4a). Nevertheless, no significanthistolopathological changes were observed in liver of DSS-exposed hamsters compared to healthy animals. Further effectson hepatic functionality were assessed by determining theexpression of several pro-inflammatory cytokines and metabolicmarkers. In agreement with the lack of histological changes,plasma levels of ALAT and AST or the local expression of theinflammatory mediators IL-1β, iNOS and TNF were not signifi-cantly different between healthy and DSS-exposed animals;although IL-6 expression was significantly down regulated in DSS-exposed hamsters compared to healthy animals (Fig. 4b andTable 2). Interestingly, genes involved in fatty acid metabolismwere affected by inflammation, significantly inducing LPLexpression and reducing CYP4A expression and down-regulating
ACO mRNA expression (p=0.09 compared to healthy controls;Fig. 4c). Plasma levels of triglycerides were increased in DSS-exposed hamster compared with those in healthy animals (Table2). The other metabolic parameters (cholesterol and glucose)were essentially comparable between DSS-treated and healthyhamsters, although an elevation in glucose levels was found inthe DSS-treated compared to healthy hamsters (p=0.12; Table2).
3.4. DSS exposure reduces LPL mRNA expression in theinflamed colon of hamsters
As for the liver, the expression of enzymes related to lipids' me-tabolism was also altered in the colon. LPL expression in the pro-ximal colon was significantly decreased in DSS-treated hamsterscompared to healthy animals, with a similar tendency for ACOexpression (p=0.11; Table 3), while CYP4A expression was un-changed (Table 3). No changes in LPL, ACO or CYP4A expressionwere detected in the distal colon of DSS-treated compared tohealthy hamsters (Table 3).
4. Discussion
DSS-induced colitis was first described in mice and is current-ly a well-established model for IBD [2,3]. Herein, we presenta careful characterization of DSS-induced effects in the colonand liver function in hamsters. Data obtained showed that inhamsters, DSS exposure induced clinical symptoms, systemicinflammatory reaction and colonic inflammation charac-terized by increased local pro-inflammatory mediators andhistopathological changes. These observations are similarto those described in either mice or rats receiving DSS(reviewed in [11,12]). However, in addition to the colonicinflammation, hamsters exposed to DSS exhibited significantalterations in liver function, particularly in relation tothe fatty acid metabolic pathways, without any obvious ef-fect on pro-inflammatory mediators or histopathologicalalterations.
In this study, we showed that DSS induces an acute colonicinflammation in hamsters, with a concentration of DSS simi-lar to the described for mice or rats [2,3,13]. Our preliminaryobservations also suggest that the inflammatory process israpidly resolved after DSS withdrawal. In mice, this pheno-menon seems to be strain related, with some murine strainsshowing a rapid healing (BALB/c), whereas others progress toa long-lasting chronic inflammatory process (C57BL/6) [3].Whether a similar phenomenon is present in hamsters and ifit is also strain-related deserves further studies. However,our observations indicate that hamsters, at least the strainused in the current study, might behave as BALB/c ratherthan C57BL/6 mice.
Interestingly, hamsters showed a predominant inflamma-tory reaction in the proximal colon. This contrasts with thatusually observed in themost usedmurine strains (i.e., BALB/cand C57BL/6), in which inflammatory responses predominatein the distal colon. The macroscopic location of inflammationin the proximal colon was confirmed histologically and wasassociated to an increased local expression of inflammato-ry markers, namely IL-1β, IL-6 and iNOS, further supportingthe notion of an ongoing acute inflammation. The histo-pathological changes observed (mucosal ulcerations, crypt
26 A. Karlsson et al.
degeneration, epithelial cell loss and infiltration of poly-morphonuclear and mononuclear cells) resembled thoseobserved on day 5 of 5 or 3% DSS exposure in either BALB/cor C57BL/6 mice, respectively [3]. Nevertheless, the distalcolon of DSS-exposed hamsters showedmacroscopical signs ofmild inflammation. However, no histopathological alterationsor changes in the expression of inflammatory mediators weredetected in this part of the colon. At the moment, we do nothave any explanation for the regional differences betweenhamsters and mice, although it might be a strain-relatedresponse as previously reported in mice [5].
In addition to the local (colonic) inflammatory reaction,hamsters exposed to DSS presented a systemic acute in-flammatory response, as reflected by the increased levels ofhaptoglobin and SAA and decreased levels of albumin[14,15]. A similar systemic response was found in DSS-induced acute colitis in mice [10], however some differencesbetween mice and hamster should be considered. In contrastto mice, the levels of haptoglobin in healthy hamsters arerelatively high (5–10 fold that reported in healthy mice).Exposure to DSS in hamsters induced a 2–3 fold increases inhaptoglobin levels, compared to a 10–20 fold increase ob-served in mice [3].
It is well established that IBD is associated to severalhepatobiliary disorders e.g., primary sclerosing cholangitis,cirrhosis and cholangiocarcinoma [6]. According to theresults obtained in the current study, hamsters exposed toDSS for 6 days displayed also significant alterations in hepaticfunction, with a reduction in total liver weight and changesin the expression of metabolic enzymes, but withouthistopathological changes or up regulation of inflammatorymediators. In line with this data, rats exposed to 3% DSS for7 days showed no significant liver injury, although alterationsin hepatic P450 enzymes were reported [16]. On the otherhand, dinitrobenzene sulphonic acid induced colitis in micewas accompanied by leukocyte infiltration in the liver,without other signs of liver injury [17]. These observations,together with the present results in hamsters, suggest alimited involvement of the liver in the colitis process, whichis in agreement with the notion that in humans livercomplications do not usually correlate to the degree of gutinflammation [6,17]. Nevertheless, liver enzymes implicatedin fatty acid metabolism were significantly altered inhamsters with acute colitis. These changes correlated witha similar alteration in colonic tissues and also with anincrease in plasma triglycerides. Interestingly, a decrease infatty acid synthase expression was recently reported in UCpatients [18] and fat wrapping has been considered amacroscopic hall mark in Crohn's disease patients [19].
Changes in fatty acid metabolism in the DSS-treatedanimals were associated to an elevation in liver LPLexpression and a decrease in liver CYP4A and ACO expres-sion. Similar changes have been observed in other models ofinflammatory conditions [20,21] and are likely to result in anincreased production of triglycerides with reduced use [22],which might contribute to the elevated plasma levels oftriglycerides detected in the DSS-exposed animals. In modelsof lipopolysaccharide (LPS)-induced acute inflammation,hepatic fatty acid synthesis is stimulated within hours afterexposure, resulting in increased plasma Tg, free cholesteroland SAA. This response has been observed in mice, rats andhamsters, with hamsters being the most sensitive species to
LPS [8,23]. However, in contrast to those results in hamsters,mice with acute DSS colitis have decreased Tg levels [10].The discrepancy in the Tg pattern is most likely due todifferences in liver metabolism between mice and hamsters.Interestingly, increased levels of Tg have been reported inIBD patients [24,25]. Furthermore, lipoproteins metabolismin hamsters is comparable to humans, due to similarities inlipoproteins and bile acids metabolism [26]. Taken together,these observations suggest that hamster might be a bettermodel than other rodent species to reproduce the hepaticalterations, particularly those related to fatty acids meta-bolism, observed in IBD.
Murine and rat models of colitis are normally used in theevaluation of new potential therapies for IBD. However, insome instances, other species might be better suited fortarget validation and evaluation due to e.g., differences intarget expression or target activity between humans and theclassical murine and rat models. Examples of such targets areendothelin [27] or neutrophil elastase (NE) [28]. Forexample, a higher anti-NE activity of protease inhibitorshas been reported in rodents compared to humans, whilehamsters and humans have similar activity [28]. Thus, the IBDmodel described herein can be a useful alternative to themore classical models when characterizing targets for whichhamsters might represent better human-like characteristicsthan mice or rats e.g., in hepatic and metabolic responses.
In summary, we have shown that, in hamsters, 6 days ofDSS exposure induces acute colitis with clinical symptomsand local and systemic inflammatory responses resemblingthose reported in the murine DSS colitis model and in humanIBD [1–3,29]. In addition, compared to the responses in miceor rats, hamsters presented unique hepatic and metabolicalterations reminiscent of the changes seen in IBD patientsand as such, the current model should be suitable for targetdiscovery and validation of IBD hepatic-related metabolicalterations.
Acknowledgments
We gratefully acknowledge the skilful contribution of mem-bers in the IBD group, Integrative Pharmacology; Mrs.Liselotte Hallengren and her group for taking care of thehamsters and Dr. Anders Elmgren and his group for bio-analytical analysis. The authors want to thank Dr. VicenteMartinez for comments on the manuscript.
References
[1] Podolsky DK. Inflammatory bowel disease. [see comment]N EnglJ Med 2002;347:417–29.
[2] Okayasu I, Hatakeyama S, Yamada M, Ohkusa T, Inagaki Y, NakayaR. A novelmethod in the induction of reliable experimental acuteand chronic ulcerative colitis in mice. Gastroenterology 1990;98:694–702.
[3] Melgar S, Karlsson A, Michaelsson E. Acute colitis induced by dex-tran sulfate sodium progresses to chronicity in C57BL/6 but not inBALB/c mice: correlation between symptoms and inflammation.Am J Physiol Gasterointest Liver Physiol 2005;288: G1328–38.
[4] Ni J, Chen SF, Hollander D. Effects of dextran sulphate sodium onintestinal epithelial cells and intestinal lymphocytes. Gut 1996;39:234–41.
27Acute colitis in hamsters
[5] Mahler M, Bristol IJ, Leiter EH, Workman AE, Birkenmeier EH,Elson CO, et al. Differential susceptibility of inbredmouse strainsto dextran sulfate sodium-induced colitis. Am J Physiol 1998;274:G544–51.
[6] Adams DH, Eksteen B. Aberrant homing of mucosal T cells andextra-intestinal manifestations of inflammatory bowel disease.Nat Rev Immunol 2006;6:244.
[7] Hardardottir I, Grunfeld C, Feingold KR. Effects of endotoxin onlipid metabolism. Biochem Soc Trans 1995;23:1013–8.
[8] Khovidhunkit W, Kim MS, Memon RA, Shigenaga JK, Moser AH,Feingold KR, et al. Thematic review series: the pathogenesis ofatherosclerosis. Effects of infection and inflammation on lipidand lipoprotein metabolism mechanisms and consequences tothe host. J Lipid Res 2004;45:1169–96.
[9] Larsson AE, Melgar S, Rehnstrom E, Michaelsson E, Svensson L,Hockings P, et al. Magnetic resonance imaging of experimentalmouse colitis and association with inflammatory activity.Inflamm Bowel Dis 2006;12:478–85.
[10] Melgar S, BjursellM,GerdinAK, SvenssonL,MichaelssonE,BohloolyYM. Mice with experimental colitis show an altered metabolismwith decreased metabolic rate. Am J Physiol Gasterointest LiverPhysiol 2007;292:G165–72.
[11] Jurjus AR, Khoury NN, Reimund JM. Animal models of inflam-matory bowel disease. J Pharmacol Toxicol Methods 2004;50:81–92.
[12] Byrne FR, Viney JL. Mouse models of inflammatory bowel dis-ease. Curr Opin Drug Disc Dev 2006;9:207–17.
[13] Sasaki S, Hirata I, Maemura K, Hamamoto N, Murano M, ToshinaK, et al. Prostaglandin E2 inhibits lesion formation in dextransodium sulphate-induced colitis in rats and reduces the levelsof mucosal inflammatory cytokines. Scand J Immunol 2000;51:23–8.
[14] Don BR, Kaysen G. Serum albumin: relationship to inflammationand nutrition. Sem Dial 2004;17:432–7.
[15] Wang Y, Kinzie E, Berger FG, Lim SK, Baumann H. Haptoglobin, aninflammation-inducible plasma protein. Redox Report 2001;6:379–85.
[16] Masubuchi Y, Horie T. Endotoxin-mediated disturbance of hepaticcytochrome P450 function and development of endotoxin tole-rance in the rat model of dextran sulfate sodium-induced expe-rimental colitis. Drug Metab Dispos 2004;32: 437–41.
[17] Scott JR, Fox-Robichau AE. Hepatic leukocyte recruitment in amodel of acute colitis. Am J Physiol Gastrointest Liver Physiol2002;283:G561–656.
[18] Heimerl S, Moehle C, Zahn A, Boettcher A, Stremmel W,Langmann T, et al. Alterations in intestinal fatty acid metabo-lism in inflammatory bowel disease. Biochim Biophys Acta MolBasis Dis 2006;1762:341.
[19] Peyrin-Biroulet L, Chamaillard M, Gonzalez F, Beclin E, DecourcelleC, Antunes L, et al. Mesenteric fat in Crohn's disease: a patho-genetic hallmark or an innocent bystander? Gut 2007;56:577–83.
[20] Singh I, Paintlia AS, Khan M, Stanislaus R, Paintlia MK, Haq E,et al. Impaired peroxisomal function in the central nervoussystem with inflammatory disease of experimental autoim-mune encephalomyelitis animals and protection by lovastatintreatment. Brain Res 2004;1022:1–11.
[21] Warren GW, Van Ess PJ, Watson AM, Mattson MP, Blouin RA.Cytochrome P450 and antioxidant activity in interleukin-6knockout mice after induction of the acute-phase response.J Interferon Cytokine Res 2001;21:821–6.
[22] Rip J, Nierman MC, Ross CJ, Jukema JW, Hayden MR, KasteleinJJP, et al. Lipoprotein lipase S447X: a naturally occurring gain-of-function mutation. Arterioscler Thromb Vasc Biol 2006;26:1236–45.
[23] Feingold KR, Hardardottir I, Memon R, Krul EJT, Moser AH,Taylor JM, et al. Effect of endotoxin on cholesterol biosynthe-sis and distribution in serum lipoproteins in Syrian hamsters.J Lipid Res 1993;34:2147–58.
[24] Levy E, Rizwan Y, Thibault L, Lepage G, Brunet S, Bouthillier L,et al. Altered lipid profile, lipoprotein composition, and oxidantand antioxidant status in pediatric Crohn disease. Am J Clin Nutr2000;71:807–15.
[25] Kosaka T, Yoshino J, Inui K,Wakabayashi T, Okushima K, Kobayash*t, et al. Impact of lipoprotein lipase gene polymorphisms onulcerative colitis. World J Gastroenterol 2006;12: 6325–30.
[26] Sullivan MP, Cerda JJ, Robbins FL, Burgin CW, Beatty RJ. Thegerbil, hamster, and guinea pig as rodent models for hyperli-pidemia. Lab Anim Sci 1993;43:575–8.
[27] Shiota N, f*ckamizu A, Takai S, Okunishi H, Murakami K, MiyazakiM. Activation of angiotensin II-forming chymase in the cardio-myopathic hamster heart. J Hypertens 1997;15:431–40.
[28] Schulz H, Mohr M, Meyer M. Elastase and papain inhibition byserum of mammals. Scand J Clin Lab Invest 1989;49:381–8.
[29] Dieleman LA, Ridwan BU, Tennyson GS, Beagley KW, Bucy RP,Elson CO. Dextran sulfate sodium-induced colitis occurs in severecombined immunodeficient mice. Gastroenterology 1994;107:1643–52.
[30] Guo Q, Sahoo SP, Wang PR, Milot DP, Ippolito MC, Wu MS, et al. Anovel peroxisome proliferator-activated receptor alpha/gammadual agonist demonstrates favorable effects on lipid homeostasis.Endocrinology 2004;145:1640–8.
[31] Domenech VS, Nylen ES, White JC, Snider Jr RH, Becker KL,Landmann R, et al. Calcitonin gene-related peptide expressionin sepsis: postulation of microbial infection-specific responseelements within the calcitonin I gene promoter. J Investig Med2001;49:514–21.
