Elevated Serum Eotaxin Levels in Patients WithInflammatory Bowel DiseaseAmparo Mir, M.D., Miguel Minguez, M.D., Jose Tatay, M.D., Isabel Pascual, M.D., Antonio Pena, M.D.,Vicente Sanchiz, M.D., Pedro Almela, M.D., Francisco Mora, M.D., and Adolfo Benages, M.D.Departments of Allergy and Clinical Immunology, and Gastroenterology, Hospital Clınico Universitario; andUnidad Mixta de Investigacion, Facultad de Medicina, Universidad de Valencia, Valencia, Spain

OBJECTIVE: Eotaxin is a recently characterized chemokinewith potent and selective chemotactic activity for eosino-phils. Previous studies indicating that eosinophils accumu-late and become activated in inflammatory bowel disease(IBD) led us to hypothesize that eotaxin is potentially in-volved in the pathophysiology of IBD and, therefore, thateotaxin would be increased in the serum of patients withIBD. The objective of this study was to test those assump-tions.

METHODS: We investigated 72 patients with IBD, 35 withulcerative colitis, and 37 with Crohn’s disease. A total of 27patients had active and 45 inactive disease; 26 were receiv-ing corticosteroids. Eotaxin serum levels were determinedby solid phase sandwich ELISA. Lymphocytes, monocytes,and granulocyte subpopulations were determined in freshblood samples with an automated autoanalyzer.

RESULTS: Serum eotaxin levels were significantly higher inpatients with Crohn’s disease and in those with ulcerativecolitis than in the control subjects (p � 0.0001). Patientswith inactive Crohn’s disease had significantly higher levelsof eotaxin than patients with inactive ulcerative colitis (p �0.05). We did not find significant differences for activity orinactivity of disease, nor for treatment with prednisone. Anegative correlation (p � 0.05) was found between eotaxinserum level and eosinophil counts in peripheral blood inpatients with Crohn’s disease.

CONCLUSIONS: There is an increased expression of eotaxinin IBD patients, suggesting that eotaxin may be involved inthe pathogenesis of IBD. This increase is more accentuatedin Crohn’s disease and negatively correlates with the eosin-ophil number in peripheral blood. Our data support theincreasing evidence that eosinophil are functionally in-volved in the pathophysiology of IBD. (Am J Gastroenterol2002;97:1452–1457. © 2002 by Am. Coll. of Gastroenter-ology)

INTRODUCTION

Eosinophils are present in the lamina propria of the GI tract,where they are believed to have an important role in the hostdefense against helminth infection (1). However, because

eosinophils are powerful proinflammatory cells, their accu-mulation and activation can produce tissue damage (2–4).After activation, eosinophils release inflammatory media-tors and specific cytotoxic proteins, such as the eosinophilcationic protein, eosinophil protein X, or major basic pro-tein, which can induce cytotoxic damage to various targetcells including epithelial bowel cells (5). In some boweldiseases such as eosinophilic gastroenteritis, eosinophiliccolitis and celiac disease eosinophils accumulate in themucosa and help to maintain the inflammatory response (6,7). Recent studies have shown increased levels of activatedeosinophils in the mucosa of patients with inflammatorybowel disease (IBD) (8–10), but their pathogenic role inhumans has not yet been demonstrated.

Eosinophils are generated in the bone marrow from plu-ripotential stem cells, and IL3, IL5, and GM-CSF regulatetheir development, accumulation, and function (11). Eosin-ophil trafficking into tissues is a complicated process thatinvolves sequential steps and is regulated by adhesion mol-ecules and chemoattractants (3). The mechanisms involvedin the migration of eosinophils to the GI tract are stillunknown and have only recently begun to be studied (12,13). An understanding of those mechanisms would signifi-cantly contribute to our knowledge of the pathogenesis ofmany diseases.

Chemokines are a large family of chemotactic cytokineswhose main role is the recruitment and activation of leuko-cytes during the inflammatory process (14). In addition, theymodulate several biological effects, such as modulate an-giogenesis, hematopoiesis, and T lymphocyte activation(15, 16). These cytokines act through interaction with dif-ferent receptors expressed on different kinds of leukocytes(17). Two subfamilies, CXC or � and CC or � chemokines,are distinguished according to the position of the first twocysteines, which are either separated by one amino acid orare adjacent (18).

Recently, eotaxin, a � chemokine with preferential che-motactic capacity for eosinophils, has been characterized(19–22). Eotaxin, unlike other chemokines, binds on onlyone receptor, CCR3, which is highly expressed on eosino-phils (23). It has no activity on neutrophils and monocytes,which do not seem to express CCR3 (24); but it does show

THE AMERICAN JOURNAL OF GASTROENTEROLOGY Vol. 97, No. 6, 2002© 2002 by Am. Coll. of Gastroenterology ISSN 0002-9270/02/$22.00Published by Elsevier Science Inc. PII S0002-9270(02)04043-1

a certain degree of chemotactic activity toward T helper type2 cells (Th2) (25) and basophils (26). These cells have alsobeen shown to express CCR3 (27). This explains whyeotaxin has the potential to mediate the migration of thesecells, although applied in vivo it only attracts eosinophils(22).

Eotaxin contributes to eosinophilic trafficking into base-line and inflammatory tissues (20, 22, 28). Besides stimu-lating chemotaxis, eotaxin induces eosinophlic aggregationand increases the expression of adhesion molecules, such asCD11b (29). This permits a firmer bond between the leu-kocyte and the endothelial cell and results in successfuladhesion and transmigration of eosinophlis to inflammatorytissues. Synergically with IL5, it can enhance the release ofeosinophils and their bone marrow progenitors, generating aquick eosinophilia (30).

In experimental animal models it has been found thateotaxin has a main role in allergic inflammation pathophys-iology (19, 20, 31). Several clinical studies have foundeotaxin to be involved in the pathogenesis of allergic dis-eases such as asthma (32, 33), allergic rhinitis (34), andatopic dermatitis (35), as well as in neoplasic (36), parasite(37), and bullous diseases (38). An increased expression ofeotaxin mRNA is shown in the intestinal mucosa of patientswith IBD (21); however, there are no other data regardingthe role of eotaxin in this disease.

The participation of eosinophils in Crohn’s disease andulcerative colitis, as demonstrated in recent studies (8–10,39), along with the important role of eotaxin in selectiverecruitment of eosinophils into tissues (12, 13, 21), led us tohypothesize a role for eotaxin in the pathogenesis of IBD.The aim of this study was to investigate eotaxin levels in theserum of patients with CD and UC, compared with healthycontrols, and to analyze the results depending on the stageof the disease and treatment with corticosteroids.

MATERIALS AND METHODS

Patient PopulationSera from 72 patients with IBD were studied. The groupincluded 37 men and 35 women between 17 and 76 yr ofage. The diagnosis for each patient was made using eitherclinical or radiographic studies, or histological criteria. Clin-ical activity was assessed using the Crohn’s Disease Activ-ity Index (CDAI) of Harvey and Bradshaw (40) and theTruelove-Witts index (41) for the patients with ulcerativecolitis. Patients with a clinical index �150 in the CDAI and

a clinical index �10 in the Truelove-Witts index wereclassified as having active disease. A total of 27 patients hadactive disease and 45 inactive. At the time of study, 26 of the71 patients were receiving oral steroids and 42 patients werereceiving 5-aminosalicylic acid. The clinical features ofpatients are shown in Table 1.

Patients With Crohn’s DiseaseA total of 37 patients with Crohn’s disease were studied.The group included 19 men and 18 women between 17 and76 yr of age. Nine patients had active disease and 28 inactiveaccording to criteria established by the CDAI (40). At thetime of study, 11 of the 37 patients were receiving oralsteroids and 19 patients were receiving 5-aminosalicylicacid.

Patients With Ulcerative ColitisA total of 35 patients with ulcerative colitis were evaluated.This group included 18 men and17 women between 24 and75 yr of age. Eighteen patients had active disease and 17inactive, according to criteria of Truelove-Wits (41). At thetime of the study 15 of 35 patients were receiving oralsteroids and 23 patients were receiving 5-aminosalicylicacid.

Control GroupA total of 38 healthy individuals, 20 men and 18 women,with an age range similar to that of the patient groups, wereincluded as the control group.

Informed consent to participate in the study was obtainedfrom each patient in accordance with the guidelines of theSpanish Department of Health.

Blood SamplesVenous blood samples (10 ml) were collected from patientsand healthy donors in Vaicutainer tubes (Becton Dickinson,Mountain view, CA). Serum samples were obtained in eachcase and cryopreserved at –80° until use.

Serum Eotaxin AssayEotaxin levels were determined by solid phase sandwichELISA, using the Cytoscreen Human Eotaxin immunoassay(Biosource International, Camarillo, CA). A monoclonalantibody specific for human eotaxin was coated onto thewells of the microtiter strips provided and 100 �l of serumadded to the wells. After incubation and washing of un-bound proteins, 100 �l of a biotinylated monoclonal secondantibody was added to each well. After another period of

Table 1. Clinical Features of Study Subjects

Ulcerative Colitis (n) Crohn’s Disease (n) Controls (n)

Number (male/female) 35 (18/17) 37 (19/18) 38 (20/18)Mean age (range), yr 44 (24–75) 40 (17–76) 46 (20–70)Active disease 18 9Inactive disease 17 28Treatment with oral steroids 15 11Treatment with 5-aminosalicylic acid 23 19

1453AJG – June, 2002 Elevated Serum Eotaxin in IBD

incubation and washing, streptavidin peroxidase was added.After incubation and washing, substrate solution and stopsolution were added. The colorimetric absorbance was readat 450 nm and compared with a standard curve to determinethe eotaxin concentration in each serum sample. The inten-sity of this colored product is directly proportional to theconcentration of eotaxin present in the original specimen.Each sample was measured in duplicate and the value ob-tained represents the mean of two samples.

Eosinophil CountsLymphocytes, monocytes, and granulocyte subpopulationswere counted from fresh blood samples by using an auto-mated autoanalyzer (Coulter). Eosinophil counts were ex-pressed as absolute numbers (eosinophils/mL).

Statistical AnalysisResults are expressed as mean � SEM. Parametric andnonparametric statistics were applied. Unpaired data wereanalyzed with Student’s t test and the Mann-Whitney rank-sum test. Spearman’s rank test was used to assess the cor-relation between immunological parameters. A significancelimit of p � 0.05 was used.

RESULTS

Serum Eotaxin Levels in IBD Patients and ControlsSerum eotaxin levels were significantly higher in patientswith IBD (mean 192 � 9) than in control subjects (mean118 � 8; p � 0.0001; Fig. 1). In general, there was nosignificant difference between the active (mean 191 � 18)and inactive form of the disease (mean 194 � 9).

Serum Eotaxin Levels in Patients With Crohn’s Diseaseand Patients With Ulcerative ColitisSerum eotaxin levels were significantly higher in patientswith Crohn’s disease (mean 198 � 11) and in patients withulcerative colitis (mean 184 � 13) than in control subjects(mean 118 � 8; p � 0001; Fig. 1). No statistically signif-icant difference was observed between serum eotaxin levelsin the group of Crohn’s disease and in ulcerative colitispatients. However, when clinical parameters were taken intoaccount, it was found that inactive Crohn’s disease patientshad significantly higher levels of eotaxin than the inactiveulcerative colitis group (mean 209 � 12 and 166 � 12,respectively) (p � 0.05; Fig. 2).

Effect of Disease Activity and Drug Treatment onSerum EotaxinWe evaluated the effects of disease activity and medicationon serum eotaxin. We did not find significant differences foractivity or inactivity of disease in Crohn’s disease nor inulcerative colitis. The small number of patients in eachgroup limits the power of this analysis. Serum eotaxinconcentration in patients who were taking oral steroidstended to be higher than those of patients with not takingoral steroids (mean 201 � 14 and 186 � 11, respectively).

Eosinophil Counts in Peripheral BloodEosinophil counts were not significantly different in patientswith Crohn’s disease (mean 1691 � 262) and ulcerativecolitis (mean 1910 � 398). Patients with active disease hadhigher eosinophil counts than the inactive patients (mean2138 � 490 and 1719 � 272, respectively) but the differ-ence was not statistically significant.

Figure 1. Serum eotaxin levels in patients with Crohn’s disease(CD), ulcerative colitis (UC), and control subjects. Data are pre-sented as mean � SEM. Filled circles indicate patients with activedisease; open circles indicate patients with inactive disease. Meanlevels were significantly higher in the CD and UC groups than inthe control group (p � 0.0001).

Figure 2. Serum eotaxin levels in patients with inactive disease.Data are presented as mean � SEM. Mean eotaxin levels weresignficantly higher in the Crohn’s disease group (CD) than in theulcerative colitis group (UC; p � 0.05) and control group (p �0001). Means for the UC group also differed from those of thecontrol group (p � 0.003).

1454 Mir et al. AJG – Vol. 97, No. 6, 2002

Serum Eotaxin Level and Eosinophil Counts inPeripheral BloodA negative correlation was found between eotaxin serumlevel and eosinophil counts in patients with Crohn’s disease(p � 0.05). No correlation was found between eotaxinserum level and eosinophil counts in ulcerative colitis pa-tients, or between eotaxin and other parameters as sedimen-tation rate (data not shown).

DISCUSSION

Eotaxin is a recently characterized chemokine that has beenimplicated in eosinophilic inflammation diseases. Severalstudies have proved that eotaxin is the main mediator inrecruitment, accumulation, and activation of eosinophils inthe lung in allergic asthma (32, 33) and is also involved incancerous (36) and parasitic diseases (37). Elevated plasmalevels of eotaxin are associated with acute asthma andimpaired lung function, which implies that eotaxin may beinvolved in the exacerbation of asthma (33). The first reportof circulating eotaxin in an infective disease in humans wasrecently made by Evans et al. (37). These authors reportedthat eotaxin was elevated in the serum of patients withcysticercosis and that this fact is consistent with the hypoth-esis that eosinophil-selective mediators have a role in thepathogenesis of cysticercosis (37).

Eotaxin is known to be expressed constitutively in the GItract, where it is responsible for eosinophilic basal homingand for host defense against parasites (12, 13, 21), but itscontribution to the pathophysiological mechanisms of IBDare not well known yet.

Recent studies show that eosinophils accumulate andbecome activated in IBD (8–10). The location of activatedeosinophils in areas with endoscopic evidence of recurrencein surgical specimens derived from patients with Crohn’sdisease supports the hypothesis that leukocytes take part inproducing tissue damage (39). On the other hand, severalstudies show increased levels of eosinophilic mediators infeces and rectal fluid from patients with ulcerative colitis (9,42, 43); and even though the pathobiological role of theeosinophil-derived mediators is not yet clear, there is in-creasing evidence that these cells could be involved in thepathogenesis of chronic IBD (8, 9). Defining the mecha-nisms that control recruitment of eosinophils to tissues isfundamental to understanding these disease processes.

In this study, we have investigated the levels of eotaxin inthe serum of patients with ulcerative colitis and withCrohn’disease, in their active or inactive form and under theeffect of drug treatment with glucocorticoids or withouttreatment. Our data show that the serum levels of eotaxin aresignificantly higher in patients with IBD compared withcontrol patients with no evidence of inflammatory disease.While the present article was in the process of revision,Chen et al. (44) published similar results in a group ofpatients with IBD. However, these authors find an increasein eotaxin only in patients with active disease. We found an

increased level of serum eotaxin even when consideringonly those patients with the inactive form of the disease.This fact indicates that in IBD there are immunologicalabnormalities that can be detected in the absence of clini-cally active disease.

On the basis of our present knowledge of its biology,eotaxin can be considered as a systemic indicator of localeosinophil chemotaxis (45). Given the established role ofeotaxin in the regulation of eosinophilic inflammation in theGI tract in healthy (12) and pathological states (21, 31), andthe increasing evidence of its role in tissue damage in IBD,the elevated concentration of eotaxin supports the hypoth-esis that eosinophil chemotaxis may be involved in diseasepathogenesis.

Taking into account the whole group of patients in dif-ferent clinical stages of the disease, we found no significantdifferences in serum levels of eotaxin between patients withCrohn’s disease and those with ulcerative colitis. However,considering only those patients with inactive disease, theeotaxin level is significantly higher in patients with Crohn’sdisease than in those with ulcerative colitis, even though theeotaxin levels of both groups are higher than those of thecontrol group.

Crohn’s disease and ulcerative colitis differ in clinical,endoscopic, and histological characteristics (46). Recentstudies suggest that IBD might result from an alteration ofthe T helper polarization. Ulcerative colitis is associatedwith Th2 polarization (47) and Crohn’s disease with a Th1polarization (48). Thus, the eotaxin increase that we foundin ulcerative colitis patients supports the hypothesis of a Th2polarization in this disease, as eotaxin is considered to be aTh2 chemokine (49). In contrast, the elevated serum eotaxinin Crohn’s disease patients is not consistent with the hy-pothesis of a Th1 prevalence in this disease. However, in anexperimental model of granulomatose lung inflammation, ithas been demonstrated that eotaxin is higher in eosinophil-rich type 2 granulomas induced by Schistosoma mansoniantigens than in type 1 granulomas induced by purifiedprotein derivative of Mycobacterium bovis (PPD) (50).However eotaxin depletion abrogates not only type 2 butalso type 1 inflammation, suggesting that eotaxin expressionis not limited to Th2 response but also promotes Th1 re-sponse (50). In addition, this same study showed that in vitrograded doses of eotaxin increased IFN� in PPD-sensitivelymph node cells.

We found a significant negative correlation between eo-sinophil counts in peripheral blood and eotaxin serum levelin patients with Crohn’s disease. Eotaxin is involved in theselective recruitment of eosinophils into the tissues, both atbaseline and in experimental models of eosinophilic GIinflammation (12, 31). Thus, a selective recruitment ofeosinophils to the GI tract cannot be ruled out as an expla-nation for the moderate decrease in these cells that has beenobserved in peripheral blood.

The differences between serum eotaxin levels in patientswith clinically active disease and remission were not rele-

1455AJG – June, 2002 Elevated Serum Eotaxin in IBD

vant, unlike the case of asthmatic patients, in whom highereotaxin levels are directly related to the severity of theasthma (45). Thus, according to our data, eotaxin cannot beconsidered a factor that indicates the clinical activity of thedisease. In the past few years, several biological markers forthe clinical activity of the disease have been described.However, the increase of these proteins during the activephase are detected in mucosa, feces, or intestinal fluid, notin serum. Eotaxin levels in bowel and mucosa may bedifferent, depending on the stage of the disease, as is thecase with some other inflammatory mediators. On the otherhand, the different drug treatments of the patients in ourstudy might have had an important influence on these re-sults. To address further the relevance of eotaxin as a markerof clinical activity, larger groups of patients in the samestage of the disease, and with the same treatment, areneeded.

The fact that we also found an increase in eotaxin levelsin patients treated with corticoids adds to the importance ofthe rise in eotaxin concentration, given that corticoids act toinhibit the expression of eotaxin (51). Paradoxically, serumeotaxin concentration in patients who were taking oral ste-roids tended to be higher than those of patients not takingoral steroids. Lilly et al. (33) came to the same conclusionstudying a group of asthmatic patients. It is possible, as theyhypothesized, that the worse the disease is in glucocorticoid-treated patients, the higher the eotaxin serum levels are,because eotaxin serum levels are not neutralized by glu-cocorticoids.

In conclusion, we found that serum eotaxin is increased inCrohn’s disease and in ulcerative colitis, even if we consideronly those patients with the inactive form of the disease.This increase is significantly higher in Crohn’s disease andcorrelates negatively with the eosinophil number in periph-eral blood. Our data support the increasing evidence thateosinophils are functionally involved in the pathophysiol-ogy of IBD.

ACKNOWLEDGMENTS

We thank Dr. J. E. O’Connor for critical reading of themanuscript, and Ana Alonso and Isabel Sanz for experttechnical assistance.

Reprint requests and correspondence: A. Mir, M.D., Departa-mento de Medicina, University of Valencia, Avenida BlascoIbanez, 15, Valencia 46010, Spain.

Received Mar. 12, 2001; accepted Oct. 31, 2001.

REFERENCES

1. Capron M. Eosinophils, and parasites. Ann Parasitol HumComp 1991;66(Suppl 1):41–5.

2. Wardlaw AJ, Moqbel R, Kay AB. Eosinophils: Biology androle in disease. Adv Immunol 1995;60:151–64.

3. Rothenberg ME. Eosinophilia. N Engl J Med 1998;338:1592.

4. Gleich GJ. Mechanisms of eosinophil-associated inflamma-tion. J Allergy Clin Immunol 2000;105:651–63.

5. Gleich GJ, Frigas E, Loegering DA, et al. Cytotoxic propertiesof the eosinophil major basic protein. J Immunol 1979;123:2925–7.

6. Talley NJ, Kephart GM, McGovern TW, et al. Deposition ofeosinophil granule major basic protein in eosinophilic gastro-enteritis and celiac disease. Gastroenterology 1992;103:137–45.

7. Torpier G, Colombel JF, Mathieu-Chandelier C, et al. Eosin-ophilic gastroenteritis. Ultrastructural evidence for a selectiverelease of eosinophil major basic protein. Clin Exp Immunol1988;74:404–8.

8. Makiyama K, Kanzaki S, Yamasaki K, et al. Activated ofeosinophils in the pathophysiology of ulcerative colitis. JGastroenterol 1995;30(Suppl 8):64–9.

9. Raab Y, Fredens K, Gerdin B, et al. Eosinophil activation inulcerative colitis. Dig Dis Sci 1998;43:1061–70.

10. Bischoff SC, Mayer J, Nguyen QT, et al. Immunohistologicalassessment of intestinal eosinoohil activation in patients witheosinophilic gastroenteritis and inflammatory bowel disease.Am J Gastroenterol 1999;94:3521–9.

11. Takafuji S, Bischoff SC, De Week AL, et al. Interleukin 3, andinterleukin 5 prime human eosinophils to produce leukotrieneC4 in response to soluble agonists. J Immunol 1991;147:3855–61.

12. Matthews AN, Friend DS, Zimmerman N, et al. Eotaxin isrequired for the baseline level of tissue eosinophils. Proc NatlAcad Sci 1998;95:6273–8.

13. Mishra A, Hogan SP, Lee JJ, et al. Fundamental signals thatregulate eosinophil homing to the gastrointestinal tract. J ClinInvest 1999;103:1719–27.

14. Schall TJ, Bacon KB. Chemokines, leukocyte traffiking andinflammation. Curr Opin Immunol 1994;6:865–73.

15. Leek RD, Harris AL, Lewis CE. Cytokine networks in solidhuman tumors: Regulation of angiogenesis. J Leuk Biol 1994;56:423–35.

16. Taub DD, Turcovski-Corrales SM, Key ML. Chemokines a Tlymphocyte activation 1. B chemokines costimulate human Tlymphocyte activation in vitro. J Immunol 1996;156:2095–103.

17. Murphy PM. Chemokine receptors: Structure, function androle in microbial pathogenesis. Cytokine Growth Factor Rev1996;7:47–64.

18. Baggiolini M, Dewald B, Moser B. Human chemokines: Anupdate. Annu Rev Immunol 1997;15:675–705.

19. Jose PJ, Griffiths-Johnson DA, Collins PD, et al. Eotaxin: Apotent eosinophil chemoattractant cytokine detected in aguinea pig model of allergic airways inflammation. J Exp Med1994;179:881–7.

20. Rhotenberg ME, Luster AD, Lilly CM, et al. Constitutive andallergen-induced expression of eotaxin mRNA in the guinea-pig lung. J Exp Med 1995;181:1211–6.

21. Garcıa-Zepeda EA, Rothenberg ME, Ownbey RT, et al. Hu-man eotaxin is a specific chemoattractant for eosinophil cellsand provides a new mechanism to explain tissue eosinophilia.Nat Med 1996;2:449–56.

22. Ponath PD, Qin S, Ringler DJ, et al. Cloning of the humaneosinophil chemoattractant eotaxin. Expression, receptor bind-ing and functional properties suggest a mechanism for theselective recruitment of eosinophils. J Clin Invest 1996;97:604–12.

23. Daugherry BL, Siciliano SJ, De Martino JA, et al. Cloningexpression, and characterization of the human eosinophileotaxin receptor. J Exp Med 1996;183:2349–54.

24. Combadiere C, Ahuja DK, Murphy PM, et al. Cloning and

1456 Mir et al. AJG – Vol. 97, No. 6, 2002

functional expression of a human eosinophil CC chemokinereceptor. J Biol Chem 1995;16:491–4.

25. Sallusto FC, Mackay CR, Lanzavecchia A. Selective expres-sion of the eotaxin receptor CCR3 by human T helper 2 cells.Science 1997;277:2005–7.

26. Yamada HK, Hirai K, Miyamasu M. Eotaxin is a potentchemotaxin for human basophils. Biochem Biophys Res Com-mun 1997;231:365–9.

27. Uguccioni M, Mackay CR, Ochensberger B, et al. High ex-pression of the chemokine receptor CCR3 in human bloodbasophils: Role inactivtion by eotaxin, MCP-4 and other che-mokines. J Clin Invest 1997;100:1137–43.

28. Collins PD, Marleau S, Griffiths Johnson DA, et al. Cooper-ation between interleukin-5 and the chemokine eotaxin toinduce eosinophil accumulation in vivo. J Exp Med 1995;182:1169.

29. Tenscher K, Metzner B, Schopf E, et al. Recombinant humaneotaxin induces oxygen radical production Ca2� -mobiliza-tion, actin reorganization and CD 11b upregulation in humaneosinophls via a pertussis toxon-sensitive heterotrimeric gua-nine nucleotide-binding protein. Blood 1996;88:3195–9.

30. Palframan RT, Collins PD, Williams TJ, et al. Eotaxin inducesa rapid release of eosinophils and their progenitors from thebone marrow. Blood 1998;91:2240–8.

31. Hogan SP, Mishra A, Brandt EB, et al. A critical role foreotaxin in experimental oral antigen-induced eosinophilic gas-trointestinal allergy. Proc Natl Acad Sci USA 2000;97:6681–6.

32. Lamkhioued B, Renzi PM, Abi-Younes S. Increased expres-sion of eotaxin in bronchoalveolar lavage and airways ofasthmatic contributes to the chemotaxis of eosinohils to thesite of inflammation. J Immunol 1997;159:4593–601.

33. Lilly GM, Woodruff PG, Camargo CA, et al. Elevated plasmaeotaxin in patients with acute asthma. J Allergy Clin Immunol1999;104:786–90.

34. Cristodoulopoulos P, Wright E, Frenkel S, et al. Monocytechemotactic proteins in allergen-induced inflammation in thenasal mucosa: Effect of topical corticosteroids. J Allergy ClinImmunol 1999;103:1036–44.

35. Yawalkar N, Uguccioni M, Scharer J, et al. Enhanced expres-sion of eotaxin and CCR3 in atopic dermatitis. J Invest Der-matol 1999;113:43–8.

36. Jundt F, Anagnostopoulos J, Bommert K. Hodgkin/Reed-Sternberg cells induce fibroblasts to secrete eotaxin, a potentchemoattractant for T cells and eosinophils. Blood 1999;94:2065–71.

37. Evans CA, Garcıa HH, Hartmell A, et al. Elevated concentra-

tions of eotaxin and interleukin-5 in human neurocysticerco-sis. Infect Immun 1998;66:4522–5.

38. Wakugawa M, Nakamura K, Kino H. Elevated levels ofeotaxin and interleukin-5 in blister fluid of bullouspemphigoid: Correlation with tissue eosinophilia. Br J Der-matol 2000;143:112–6.

39. Dubucquoi S, Janin A, Klein O, et al. Activated eosinophilsand interleukin 5 expression in early recurrence of Crohn’sdisease. Gut 1995;37:242–6.

40. Harvey RF, Bradshaw JM. A simple index of Crohn’s diseaseactivity. Lancet 1980;1:514.

41. Truelove SC, Witts LJ. Cortisone in ulcerative colitis, prelim-inary report on a therapeutic trial. Br Med J 1954;2:375–8.

42. Berstad A, Borkje B, Riedel B, et al. Increased fecal eosino-phil cationic protein in inflammatory bowel disease. Hepato-gastroenterology 1993;40:276–8.

43. Carlson M, Raab Y, Peterson C, et al. Increased intraluminalrelease of eosinophil granule proteins EPO, ECP, EPX, andcytokines in ulcerative colitis and proctitis insegmental perfu-sion. Am J Gastroenterol 1999;94:1876–83.

44. Chen W, Paulus B, Shu D, et al. Increased serum levels ofeotaxin in patients with inflammatory bowel disease. Scand JGastroenterol 2001;36:515–20.

45. Nakamura H, Weiss ST, Israel E, et al. Eotaxin and impairedlung function in asthma. Am J Respir Crit Care Med 1999;160:1952–6.

46. Stenson WF. Inflammatory bowel disease. In: Yamada T,Alpers DH, Laine L, et al, eds. Textbook of gastroenterology,3rd ed. Philadelphia: Lippincott, 1999:1775–839.

47. Boirivant M, Fuss IJ, Chu A, et al. Oxazolone colitis: Amurine model of T helper cell type 2 colitis treatable withantibodies to interleukin 4. J Exp Med 1998;188:1929–39.

48. Parronchi P, Romagnani P, Annunziato F, et al. Type 1 T-helper cell predominance, and interleukin-12 expression in thegut of patients with Crohn’s disease. Am J Pathol 1997;150:823–32.

49. Doucet C, Brouty-Boye D, Pottin-Clenceau C, et al. Interleu-kin (IL) IL4, and IL13 act on human lung fibroblasts: Impli-cation in asthma. J Clin Invest 1998;101:2129–39.

50. Ruth JH, Luckacs NW, Warmington KS, et al. Expression andparticipation of eotaxin during mycobacterial (type 1) andschistosomal (type 2) antigen-elicited granuloma formation.J Immunol 1998;161:4276–82.

51. Stellato C, Matsukura S, Fal A, et al. Differential regulation ofepithelial-derived C-C chemokine expression by IL-4 and theglucocorticoid budesonide. J Immunol 1999;163:5624–32.

1457AJG – June, 2002 Elevated Serum Eotaxin in IBD