lable at ScienceDirect

Nutrition 28 (2012) 799–802

Contents lists avai

Nutrition

journal homepage: www.nutr i t ionjrnl .com

Basic nutritional investigation

Dietary a-linolenic acid–rich formula reduces adhesion molecules in rats withexperimental colitis

Ayman Ibrahim M.D., Ph.D. a, Moutaz Aziz M.D. b, Aktham Hassan M.D., Ph.D. a, Khaly Mbodji Ph.D. a,Elodie Collasse B.Sc. b, Moïse Coëffier Ph.D. a,c, Fr�ed�eric Bounoure Pharm.D. d,Guillaume Savoye M.D., Ph.D. a,e, Pierre D�echelotte M.D., Ph.D. a,c, Rachel Marion-Letellier Ph.D. a,*aAppareil Digestif Environnement Nutrition (ADEN EA 4311), Medicine University, I.F.R. 23, Institute of Biomedical Research, Rouen, Franceb Laboratory of Anatomo-Pathology, Rouen University Hospital, Rouen, FrancecNutrition Unit, Rouen University Hospital, Rouen, Franced In Cyclo SAS, Medicine University, Rouen, FranceeDepartment of Gastroenterology, Rouen University Hospital, Rouen, France

a r t i c l e i n f o

Article history:Received 12 July 2011Accepted 19 October 2011

Keywords:a-Linolenic acidColitisHeme oxygenase-1Intestineu-3 Polyunsaturated acidsAdhesion molecules

Fr�ed�eric Bounoure and Pierre D�echelotte are scientifi* Corresponding author. Tel.: þ33-0-2-3514-8245;

E-mail address: marion_rachel@hotmail.com (R. M

0899-9007/$ - see front matter � 2012 Published bydoi:10.1016/j.nut.2011.10.008

a b s t r a c t

Objective: The u-3 polyunsaturated fatty acid therapy in inflammatory bowel disease is focused onthe effects on fish oil–derived polyunsaturated fatty acids. We speculated that a vegetal oil rich ina-linolenic acid (ALA) might also inhibit colitis. Therefore, we evaluated whether dietary ALAwould decrease the expression of adhesion molecules by inducing the protective enzyme hemeoxygenase-1 (HO-1) in a rat colitis model.Methods: Colitis was induced at day 0 by an intrarectal injection of 2-4-6-trinitrobenzen sulfonicacid (TNBS), whereas control rats received the vehicle. Rats were fed an ALA-rich formula 450 mg $

kg�1 $ d�1, whereas the other colitic group (TNBS) and the control group were fed an isocaloriccorn oil formula for 14 d (from day �7 to day 7). The colonic expressions of intercellular adhesionmolecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), vascular endothelial growthfactor A receptor-2 (VEGFR2), and HO-1 were studied by immunohistochemistry.Results: The ALA-rich diet significantly decreased the expression of ICAM-1, VCAM-1, and VEGFR-2compared the TNBS group, but it did not affect the expression of HO-1.Conclusion: A vegetal ALA-rich formula decreases the expression of ICAM-1, VCAM-1, and VEGFR-2and independently of HO-1 in rats with TNBS-induced colitis. Further studies are required toevaluate its therapeutic potential in inflammatory bowel disease as an alternative to fish oil.

� 2012 Published by Elsevier Inc.

Introduction

Inflammatory bowel diseases (IBDs), including ulcerativecolitis and Crohn’s disease, are chronic inflammatory disordersthat affect the gastrointestinal tract. IBD is driven by leukocyteinfiltration into the gut mucosa that is mediated by the expres-sion of adhesion molecules on endothelial and inflammatorycells [1]. Adhesion molecules such as intercellular adhesionmolecule-1 (ICAM-1) or vascular cell adhesion molecule-1(VCAM-1) are upregulated in the intestinal mucosa of patientswith IBD and in experimental colitis [1]. Recent studies have

c consultants for In Cyclo.fax: þ33-0-2-3514-8226.arion-Letellier).

Elsevier Inc.

emphasized the role of angiogenesis in the pathogenesis of IBD,and vascular endothelial growth factor A and its receptor(VEGFR2) appear to be key mediators in this process [2]. Theadhesion molecules and angiogenic factor represent attractivetargets for the development of new IBD therapy [3].

The anti-inflammatory properties of u-3 polyunsaturatedfatty acids (PUFAs) aremainly attributed tovery long-chain PUFAs(eicosapentaenoic acid [EPA] and docosahexaenoic acid [DHA]).We previously demonstrated that fish oil–derived u-3 PUFAs candownregulate adhesion molecules and VEGFR2 in endothelialcells and colitic rats [4]. In contrast, the beneficial effect of theplant u-3 fatty acid, a-linolenic acid (ALA), is still controversial.We previously reported that ALA can downregulate interleukin(IL)-1–induced cytokine production in enterocyte-like Caco-2cells but is less potent than DHA and EPA [5]. In rats with 2-4-6-trinitrobenzen sulfonic acid (TNBS)-induced colitis, a dietary

A. Ibrahim et al. / Nutrition 28 (2012) 799–802800

ALA-rich formula improved colitis by decreasing inflammationand oxidative stress [6].

The present aimed to evaluate whether a dietary ALA-richformula can downregulate adhesion molecules and VEGFR2 inan experimental model of IBD.

Materials and methods

Experimental design

Diets were supplemented with a powder (provided by In Cyclo, Rouen,France) containing corn oil (control diet) or the ALA-rich camelina oil (Phytocos,Mourenx, France), a proprietary formulation based on cyclodextrin nano-encapsulation. The formula developed in the present study was used in a powderform instead of an oily compound to enable an easier mix with the rat diet. ALAmade up 0.8% and 28.8% of the total fatty acid content of the corn oil or ALAformula, respectively. Diets were supplemented from day �7 to day 7. Ratsreceived the ALA-rich diet (TNBS þ ALA group, 450 mg $ kg�1 $ d�1) or the cornoil diet (control and TNBS groups). The two diets were isocaloric.

Colitis induction

Colitis was induced by the administration of TNBS by the method previouslydescribed [6]. The animal care and experimentation complied with the Frenchand European Community regulations (Official Journal of the EuropeanCommunity L 358, December 18, 1986) and R. M.-L. has been authorized by theFrench government to use this rat model (authorization no. 76-106). Sprague-Dawley male rats (Janvier, Le Genest St. Isle, France) weighing 200 to 250 gwere randomized into three groups (n ¼ 8 per group): control, TNBS and TNBS þALA. Rats were sacrificed on day 7, and the distal portion of the colon wasremoved, cut longitudinally, and cleaned in ice-cold phosphate buffered saline toremove fecal residues. Fixed intestinal tissues were embedded in paraffin waxblocks for histologic studies.

Immunohistochemistry

Immunohistochemistry was performed as previously described [6]. Briefly,slides were incubatedwith the primary antibodies anti-rat ICAM at 1:50 (202403,Biolgend San Diego, CA, USA), rabbit anti-rat VCAM-1 at 1:100 (sc-8304, SantaCruz Biotechnology, Santa Cruz, CA, USA), rabbit VEGFR2 (ab2349, Abcam Paris,France), and rabbit anti-rat hemeoxygenase-1 (HO-1) at 1:200 (HC 3001, EnzoVilleurbane, France). The number of immunostained leukocytes was determinedby an observer blinded to the treatment allocation on five visual fields within themucosa (magnification 400�) [7]. The staining of endothelial cells was deter-mined semiquantitatively in a four-grade scale (0, negative; 1, slight staining; 2,moderate staining; 3, strong staining).

Statistical analysis

Statistical comparisonswere performed using Prism 5 (GraphPad, La Jolla, CA,USA). Data are expressed as mean � standard error of the mean. Results werecompared with the Mann-Whitney test. Values were considered statisticallysignificant at P < 0.05.

Results

Effects of colitis induction

We observed a constitutive expression of ICAM-1 in thecapillaries in themucosa of the control rats,whereasVCAM-1wasnot detected. TNBS rats had greater expressions of ICAM-1 (P ¼0.0009; Figs.1A, 2B) and VCAM-1 (P¼ 0.0014; Figs.1B, 2E) on theinflammatory cells in themucosa and submucosa comparedwiththe controls. A similar pattern of adhesion molecule expressionwas found at an endothelial level (data not shown). The TNBS ratshad stronger HO-1 (P < 0.001; Fig. 1C) and VEGFR2 (P ¼ 0.0081;Fig. 1D) immunostaining compared with the controls.

Effects of dietary ALA-rich formula

The dietary ALA-rich oil significantly decreased the expres-sions of ICAM-1 (P ¼ 0.032; Figs. 1A, 2C), VCAM-1 (P ¼ 0.0077;

Figs. 1B, 2F), and VEGFR2 (P ¼ 0.0036; Fig. 1D). HO-1 immu-nostaining was not significantly affected by the dietary ALA (P ¼0.9474; Fig. 1C).

Discussion

In the present study, we confirmed an upregulation of theadhesion molecules VCAM-1 and ICAM-1 in colitis [1,4]. We foundthat the dietary ALA-rich formula downregulates adhesion mole-cules in rats with TNBS-induced colitis. The downregulation ofadhesion molecules by very long-chain u-3 PUFAs has beendescribed in the intestine [4,8], but we found a downregulation ofadhesionmoleculesbydietaryALA inratswithTNBS-inducedcolitis.In dyslipidemic patients, supplementation with ALA for 12 wk hasbeen found to lead to a decrease of blood-soluble VCAM-1 [9].

Vascular endothelial growth factor A and its receptor VEGFR2are key components of the angiogenesis involved in the patho-physiology of IBD [2,10]. An upregulation of VEGFR2 has alsobeen reported in experimental colitis [2,4] and in IL-1b–treatedendothelial cells [4]. Because the inhibition of cyclooxygenase-2leads to an inhibition of angiogenesis [11], the consumption ofu-3 PUFAs could inhibit angiogenesis. Interestingly, DHApretreatment leads to a downregulation of VEGFR2 in endothe-lial cells in response to IL-1b [4]. We also confirmed this findingin colitic rats, where a dietary intervention with a fish oil rich inEPA and DHA normalized endothelial VEGFR2 levels [4]. In thepresent study, we observed an increased immunostaining ofVEGFR2 in colitic rats [4] that was downregulated by the dietaryALA-rich formula. The ability of the dietary ALA-rich formula todownregulate cyclooxygenase-2 may explain its inhibitory roleon angiogenesis in colitic rats [6].

The nutritional intervention with u-3 PUFAs in patients withIBD [12–14] has focused on the effects of fish oil–derived u-3PUFAs. Although compelling studies have demonstrated a bene-ficial effect of u-3 PUFAs in experimental models of IBD, theresults of clinical trials in IBD are disappointing. A possiblemechanism behind this discrepancy is the use of fish oil–derivedu-3 PUFAs. Indeed, we previously showed that EPA and DHAdownregulate cytokine production in enterocyte-like Caco-2cells through the peroxisome proliferator-activated receptor-g(PPAR-g) pathway [5]. Mutations in the PPAR-g gene have beenassociated with IBD [15,16], and different types of fat intake havebeen reported to interact with the polymorphism of IL-6, tumornecrosis factor-a, and PPAR-g to modulate the activity of Crohn’sdisease [17,18]. Therefore, a proportion of patients with IBD canbe considered non-responders to fish oil–derived PUFA therapy.Also of interest would be the effect of a nutritional interventionwith plant-derived u-3 PUFAs in IBD. In the present study, weshowed the beneficial effects of an ALA-rich formula in rats withTNBS-induced colitis. A comparison of fish oil with perilla oil inan experimental model of ileitis has proved the better efficacy ofperilla oil to inhibit the inflammatory response by a decrease ofproinflammatory cytokines and by amore pronounced inhibitionof adhesion molecules [19]. An ALA-rich oil may offer an alter-native to fish oil in IBD therapy.

Because HO-1 inhibits adhesion molecules in activatedendothelial cells and can be activated by u-3 PUFAs [20], wespeculated that the downregulation of adhesion molecules byALA would be mediated by the HO-1 induction, which was notthe case in the present study. The HO-1 upregulation may bespecific to fish oil–derived u-3 PUFAs, in particular DHA. Wepreviously showed that dietary ALA increases the EPA but not theDHA content in the erythrocyte fatty acid composition [6].Indeed, we are not aware of any study showing an upregulation

CONTROL TNBS TNBS+ALA

0

50

100

150

a

bc

IC

AM

-1

s

ta

in

in

g

(n

um

be

r o

f c

ells

/fie

ld

)

CONTROL TNBS TNBS+ALA

0

20

40

60

80

a

b

c

VC

AM

-1

s

ta

in

in

g

(n

um

be

r o

f c

ells

/fie

ld

)

CONTROL TNBS TNBS+ALA

0.0

0.5

1.0

1.5

2.0

2.5

a

b b

HO

-1

stain

in

g sco

re

CONTROL TNBS TNBS+ALA

0

1

2

3

4

a

b

c

V-E

GF

R2

sta

in

in

g s

co

re

A

C

B

D

Fig. 1. ICAM-1 (A), VCAM-1 (B), HO-1 (C), and VEGFR2 (D) staining in rats with TNBS-induced colitis. Results are expressed as mean � SEM, and means were compared usingthe Mann-Whitney non-parametric test. Means without a common letter differ. ALA, a-linolenic acid; HO-1, heme oxygenase-1; ICAM-1, intercellular adhesion molecule-1;TNBS, 2-4-6-trinitrobenzen sulfonic acid; VCAM-1, vascular cell adhesion molecule-1; VEGFR2, vascular endothelial growth factor A receptor-2.

Fig. 2. Inflammatory cells and microvessels positive for intercellular adhesionmolecule-1 (A–C) and vascular cell adhesion molecule-1 (D–F) in the colonicmucosa of rats with colitis induced by 2-4-6-trinitrobenzen sulfonic acid in thecontrol group (A, D), 2-4-6-trinitrobenzen sulfonic acid group (B, E), and 2-4-6-trinitrobenzen sulfonic acid plus a-linolenic acid group (C, F). Magnifications 400�.

A. Ibrahim et al. / Nutrition 28 (2012) 799–802 801

of at-shock protein by ALA. Interestingly, HO-1 can be induced byPPAR-g ligands such as DHA [21] or 5-aminosalicylic acid [22].We found that the DHA activated PPAR-g in enterocyte-likeCaco-2 cells, whereas ALA did not [5]. This may explain thelack of effect of ALA on HO-1. Because a dietary ALA-rich formulahas been found to inhibit nuclear factor-kB activation [6] thatmediates the gene expression of adhesion molecules [23], wespeculated that downregulation of adhesion molecules by a die-tary ALA-rich formula would be mediated through this pathway.Adhesion molecules are induced by proinflammatory cytokines[3,4]. A potential mechanism behind the inhibitory effect of anALA-rich formula on adhesion molecules may be mediatedthrough the inhibition of tumor necrosis factor-a, a key cytokinein intestinal inflammation. Indeed, a dietary ALA-rich formulawas found to inhibit colon mRNA levels and the production oftumor necrosis factor-a in colitic rats [6].

In conclusion, a dietary ALA-rich formula decreases theexpression of adhesion molecules and VEGFR2 in TNBS-inducedcolitis independently of HO-1. Further studies are required toevaluate its therapeutic potential in IBDas analternative tofish oil.

References

[1] Danese S. Inflammation and the mucosal microcirculation in inflammatorybowel disease: the ebb and flow. Curr Opin Gastroenterol 2007;23:384–9.

[2] Scaldaferri F, Vetrano S, Sans M, Arena V, Straface G, Stigliano E, et al.VEGF-A links angiogenesis and inflammation in inflammatory boweldisease pathogenesis. Gastroenterology 2009;136:585–95. e585.

[3] Ghosh S, Panaccione R. Anti-adhesion molecule therapy for inflammatorybowel disease. Therap Adv Gastroenterol 2010;3:239–58.

[4] Ibrahim A, Mbodji K, Hassan A, Boukhettala N, Coëffier M, Savoye G, et al.Anti-inflammatory and anti-angiogenic effect of long chain n-3 poly-unsaturated fatty acids in intestinal microvascular endothelium. Clin Nutr2011;30:678–87.

[5] Marion-Letellier R, Butler M, Dechelotte P, Playford RJ, Ghosh S. Comparisonof cytokine modulation by natural peroxisome proliferator-activatedreceptor gamma ligands with synthetic ligands in intestinal-like Caco-2cells and human dendritic cellsdpotential for dietary modulation ofperoxisome proliferator-activated receptor gamma in intestinal inflamma-tion. Am J Clin Nutr 2008;87:939–48.

[6] Hassan A, Ibrahim A, Mbodji K, Coëffier M, Savoye G, D�echelotte P, et al. Analpha-linolenic acid–rich formula reduces oxidative stress and inflamma-tion by regulating NF-kappaB in rats with TNBS-induced colitis. J Nutr2010;140:1714–21.

A. Ibrahim et al. / Nutrition 28 (2012) 799–802802

[7] German AJ, Hall EJ, Day MJ. Analysis of leucocyte subsets in the canineintestine. J Comp Pathol 1999;120:129–45.

[8] Ramakers JD, Mensink RP, Schaart G, Plat J. Arachidonic acid but noteicosapentaenoic acid (EPA) and oleic acid activates NF-kappaB andelevates ICAM-1 expression in Caco-2 cells. Lipids 2007;42:687–98.

[9] Rallidis LS, Paschos G, Papaioannou ML, Liakos GK, Panagiotakos DB,Anastasiadis G, Zampelas A. The effect of diet enriched with [alpha]-linolenic acid on soluble cellular adhesion molecules in dyslipidaemicpatients. Atherosclerosis 2004;174:127–32.

[10] Danese S, Sans M, de la Motte C, Graziani C, West G, Phillips MH, et al.Angiogenesis as a novel component of inflammatory bowel disease path-ogenesis. Gastroenterology 2006;130:2060–73.

[11] Wang D, Wang H, Brown J, Daikoku T, Ning T, Shi O, et al. CXCL1 induced byprostaglandin E2 promotes angiogenesis in colorectal cancer. J Exp Med2006;203:941–51.

[12] Feagan BG, Sandborn WJ, Mittmann U, Cohen A, Dallaire C, Ponich TP, et al.Omega-3 free fatty acids for the maintenance of remission in Crohndisease: the EPIC randomized controlled trials. JAMA 2008;299:1690–7.

[13] De Ley M, de Vos R, Hommes DW, Stokkers P. Fish oil for induction ofremission in ulcerative colitis. Cochrane Database Syst Rev 2007;17:CD005986.

[14] Turner D, Shah PS, Steinhart AH, Zlotkin S, Griffiths AM. Maintenance ofremission in inflammatory bowel disease usingomega-3 fatty acids (fish oil):a systematic review and meta-analyses. Inflamm Bowel Dis 2011;17:336–45.

[15] Sugawara K, Olson TS, Moskaluk CA, Stevens BK, Hoang S, Kozaiwa K, et al.Linkage to peroxisome proliferator-activated receptor-[gamma] in SAMP1/

YitFc mice and in human Crohn’s disease. Gastroenterology 2005;128:351–60.

[16] Andersen V, Christensen J, Ernst A, Jacobsen BA, Tjønneland AA, Krarup HB,Vogel UU. Polymorphisms in NF-kappaB, PXR, LXR, PPARgamma and risk ofinflammatory bowel disease. World J Gastroenterol 2011;17:197–206.

[17] Guerreiro CS, Ferreira P, Tavares L, Santos PM, Neves M, Brito M, Cravo M.Fatty acids, IL6, and TNFalpha polymorphisms: an example of nutrigeneticsin Crohn’s disease. Am J Gastroenterol 2009;104:2241–9.

[18] Ferreira P, Cravo M, Guerreiro CS, Paula Moura Santos, Manuela Neves,Miguel Brito and Mar~Alia Cravo. Fat intake interacts with polymorphisms ofCaspase9, FasLigand and PPARgamma apoptotic genes in modulatingCrohn’s disease activity. Clin Nutr 2011;29:819–23.

[19] Matsunaga H, Hokari R, Kurihara C, Komoto S, Nakamura M, Tsuzuki Y,Kawaguchi A, et al. Omega-3 polyunsaturated fatty acids ameliorate theseverity of ileitis in the senescence accelerated mice (SAM)P1/Yit micemodel. Clin Exp Immunol 2009;158:325–33.

[20] Soares MP, Seldon MP, Gregoire IP, Vassilevskaia TT, Berberat PO, Yu JJ, et al.Heme oxygenase-1 modulates the expression of adhesion moleculesassociated with endothelial cell activation. J Immunol 2004;172:3553–63.

[21] Marion-Letellier R, Dechelotte P, Iacucci M, Ghosh S. Dietary modulation ofperoxisome proliferator-activated receptor gamma. Gut 2009;58:586–93.

[22] Horv�ath K, Varga C, Berk�o A, P�osa AA, L�aszlo FF and Whittle BJ. Theinvolvement of heme oxygenase-1 activity in the therapeutic actions of 5-aminosalicylic acid in rat colitis. Eur J Pharmacol 2008;581:315–23.

[23] Collins T, Read MA, Neish AS, Whitley M, Thanos D, Maniatis T. Tran-scriptional regulation of endothelial cell adhesion molecules: NF-kappa Band cytokine-inducible enhancers. FASEB J 1995;9:899–909.