Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

P112. Different production of soluble HLA-G antigens by peripheral blood mononuclear cells in ulcerative colitis and Crohn's disease

A. Zelante1, R. Rizzo2, R. Borgoni3, C. Galuppi1, V. Cifalà1, O. Baricordi2, S. Gullini1

1Gastroenterology Unit, University Hospital “Sant'Anna”, Ferrara, Italy; 2Department of Experimental and Diagnostic Medicine – Section of Medical Genetics, University of Ferrara, Ferrara, Italy; 3Department of Statistics, University of Milano-Bicocca, Milan, Italy

HLA-G antigens are non classical MHC class I molecules characterized by tolerogenic and anti-inflammatory properties. HLA-G molecules have a low allelic polymorphism in comparison with classical HLA genes and a restricted tissue distribution. Detectable levels of sHLA-G molecules have been reported in a percentage of plasma samples from healthy subjects and significant variations in the concentrations have been evidenced in different pathological conditions. Cytokines, hormones and immunosuppressant drugs induce HLA-G production.

In this study we have evaluated 27 Ulcerative Colitis (UC – F 17 – M 10 – median age 50 ys) and 22 Crohn's disease (CD – F 12 – M 10 – median age 55 ys) treated with multiple drugs (azathioprina, corticosteroids, mesalazine). We have analyzed plasma samples and PBMCs (peripheral blood mononuclear cells) with or without LPS-activation for the production of HLA-G molecules and compared with the results obtained in a previous cohort of 20 UC (F 9 – 3M 8; median age 54 ys) and 17 CD (female 11, male 9; median age 51 ys) patients without immunosuppressant drugs. PBMCs have been extracted by Ficoll centrifugation from whole peripheral blood and activated with 10 ng/ml of LPS for 48 hours. The culture supernatants and the plasma samples have been analyzed in triplicate by ELISA system. The levels of HLA-G secretion remain higher in non-activated PBMC cultures from CD patients versus UC patients even in treated cohort (p = 0.011 – Wilcoxon). UC treated patients presented an increased HLA-G secretion in both non-activated (p = 0.057 Mann-Whitney) and LPS-activated PBMC cultures (p = 0.004 Mann-Whitney) in comparison with the second cohort of UC patients. On the contrary, LPS-activated PBMC cultures from treated CD and UC patients released similar amounts of HLA-G (p = 0.075 Wilcoxon). These data suggest that therapy is able to control the hyper-activation of CD PBMCs and the unbalanced microenvironment of UC patients in the presence of an inflammatory stimulus as LPS. Age, gender, activity, localization or duration of the disease and previous surgery have not influenced the sHLA-G levels (p = NS; Wilcoxon). We have also analyzed sHLA-G levels in plasma samples from treated CD and UC patients and compared the results with the HLA-G secretion by the correspondent non-activated PBMC cultures. There is a strict correlation between in vitro and plasmatic levels of sHLA-G in both CD (r = 0.693; p = 0.003; Spearman) and UC (r = 0.911; p = 0.0002; Spearman) patients. Therapy is able to modify the production of HLA-G molecules in CD and UC and we propose the analysis of sHLA-G levels in CD and UC patients during the follow-up. The ability of HLA-G to distinguish between UC and CD patients and to monitor the therapy as a plasmatic biomarker would be of extreme interest in establishing an individualized treatment.