Eugénia CM Santos, Cecília MMC Mendes, Mª. Cecília M Nunes, João Castro e Melo Serviço de Imunologia, Hospital Geral de Santo António, 4000 Porto, Portugal.

INTRODUCTION AND OBJECTIVES - Abundant evidence suggests that auto- immune phenomena have an important role in the immunopathogenesis of HIV infection; among these, anti-lymphocyte antibodies are believed to be of relevance, although its origin, target cell populations and its pathogenic effects are poorly understood. Therefore, in the course of our work in an ongoing study of HIV infection, we became interested in the evaluation and characterization of these antibodies; preliminary results thereof are presented in this work.

MATERIALS AND METHOD - Population : For this preliminary study 35 patients in various Walter-Reed (WR) stages of HIV infection were selected (WR-1, n=14; WR-2, n=5; WR-3/4, n=11; WR5, n=2; WR-6, n=3). Fifteen healthy blood donors were used as a control group. Disease was staged by an extended clinical and laboratory study protocol. Lymphocyte populations : Peripheral blood cells, analysed by dual and triple colour Flow Cytometry (FCM), were characterized as (CD3/CD4)+, (CD3/CD8)+, (CD3/HLA-DR)+, (CD3)-/(CD16 and/or CD56)+, CD19+, surface membrane immunoglobulin (Smlg)+ (under conditions assuring removal of nonspecifically cell-bound lg). Anti-lymphocyte antibodies (ALA) :ALA were detected by FCM as cells (marked as for Smlg) positive for surface immunoglobulin (Sig) in excess of B lymphocytes, either directly on patient's lymphocytes or on lymphocytes of blood-group o control individuals previously incubated with plasma of patients; distribution of ALA by lymphocyte sub-populations was evaluated by simultaneous analysis of Slg and of T and B cell markers.

RESULTS - ALA were detected directly on the cells of 26 (75%) patients as a Smlg+ population significantly higher than that of B (CD19+ ) cells; in the control group, the fraction of Smlg+ cells was always closely coincident with that of B lymphocytes. Dual marker studies showed that in these patients CD4 and CD8 cells had a significant proportion (always >10%) positive for surface immunoglobulin (ALA+); this fraction was always essentially absent in cells of the control group. In some patients, all of the lymphocytes were positive for surface immunoglobulin. Control lymphocytes pre-incubated with plasma of 5 ALA+ patients (selected for this study) were always positive for surface immunoglobulin with values and a distribution by sub-populations very similar to those observed in the cells of HIV patients. Correlation between the presence of ALA and several clinical and immunological markers (WR stage, fraction and number of lymphocytes and of its subpopulations, serum immnoglobulin levels, presence and level of auto-antibodies), studied by linear regression and analysis of covariance revealed only significant association between the presence or the titer of anti-nuclear antibodies (ANA) and the fraction of CD4+/Smlg+ cells (p= 0.012) and between the proportions of Smlg CD4+/Smlg+ and of CD8+/Smlg+ cells (p= 0.013).

CONCLUSIONS - Our results clearly show that many HIV infected patients have in the peripheral blood antibodies reactive with control lymphocytes and have lymphocytes with these ALA specifically bound to their surface; these data indicate that ALA are not directed against viral components. The main T cells populations, either in patients or healthy control individuals, are always the target for these ALA; in some patients ALA are reactive with all of the lymphocytes (B, T and NK). In the present phase of our work, ALA do not appear to be related to stage of the disease nor to the number of CD4 cells (in contrast with reports correlating CD4 depletion with ALA); the observed correlation between the presence of the titer of ANA may be related to the recognized polyclonal B cell activation in HIV infection.

(This work was supported by a grant from Comissão Nacional de Luta Contra a Sida)