Immunology Letters, 30 (1991) 31-36

Elsevier IMLET 01651

Autoantibodies in male homosexuals and HIV infection S h m u e l A r g o v ~, A m i S c h a t t n e r 2, R i m o n a Burstein 2, Zeev T. H a n d z e l 2, Yehuda S h o e n f e l d 1. a n d Zvi B e n t w i c h 2 Unit for Autoimmune Diseases, Ben-Gurion University, Health Science Center, Beer-Sheba, and 2Ruth Ben-Ari Institute of Clinical Immunology, Kaplan Hospital (Affiliated to the Hebrew University Medical School), Rehovot, Israel

(Received20 January 1991;revision received 15 May 1991; accepted 21 May 1991)

1. Summary We have used ELISA to study the frequency of autoantibodies to several antigens in the serum of 17 male homosexuals (MHS) negative for HIV ( H I V - ) , 11 asymptomatic HIV seropositive MHS (HIV ÷) and patients with ARC (N = 15) or AIDS (N = 13), and compared them to 20 matched healthy heterosexual controls. Serum antibody binding to histones, cardiolipin, ss-A, ss-B and Sm was found to be significantly higher in each of the MHS groups studied as compared to controls ( P < 0.001), and was also increased in the HIV ÷ patients vs. the H I V - group ( P < 0.05), In contrast, no increase in autoantibodies to ss-DNA, ds-DNA, poly(I), poly(G) or RNP were found in any of the groups tested. These results enlarge the spectrum of autoimmunity previously reported in HIV infection and identify a similar pattern to a lesser degree, already present in H I V - MHS, suggesting a role for HIV or concomitant virus infections in its pathogenesis. 2. Introduction Viruses have long been considered as possible inducers of autoimmunity and autoimmune diseases [1]. Autoimmunity has also been suggested to have an important role in the extremely varied manifesta-

tions and in the pathogenesis of infection by the human immunodeficiency virus type 1 (HIV-1) [2-4]. One of the prime reasons for this suggestion was the demonstration of multiple autoantibodies (autoAb) in HIV-seropositive (HIV ÷) patients. However, some investigators did not find an increased incidence of autoAb in these patients [5-7]. Furthermore, the data regarding the presence of autoAb in healthy H I V - male homosexuals (MHS), and their association with the progression of disease through the stages of asymptomatic infection, AIDS-related complex (ARC) and AIDS, is often unclear. This study examines the presence of 10 different autoantibodies in the serum of controls and MHS in the different stages of HIV infection, and discusses their possible origin and significance. 3. Materials and Methods 3.1. P a t i e n t s

Fifty-six male homosexual patients (MHC) of a large cohort followed by us for several years [8-10], were compared with 20 matched heterosexual controis. The MHS group included 17 individuals with negative HIV serology and 45 patients who were HIV ÷ by ELISA and Western blots [11], of which 11 were asymptomatic, 15 had ARC and 13 had AIDS, by CDC criteria [12].

Key words: AIDS; HIV; Homosexual; Antinuclear antibody;

Autoantibody; Autoimmunity Correspondence to: Zvi Bentwich,M.D., Ruth Ben Ari Institute of Clinical Immunology,KaplanHospital, Rehovot,Israel76100. * Present address: Sheba Medical Center, Tel Hashomer, Israel.

3.2. A n t i g e n s ss-DNA and ds-DNA were prepared from calf thymus DNA (Worthington Biochemicals, Freehold, N J) as described [13]. Poly(I), poly(G), histones, and

0165-2478 / 91 / $ 3.50 © 1991 ElsevierSciencePublishers B.V.All rights reserved.

31

cardiolipin were purchased from Sigma Chemical Co., while Sm*, RNP (ribonucleoprotein), Ro/SSA and La/SSB (Sj6gren's syndrome A and B) were obtained from BioHytech Co., Ramat-Gan, Israel (Ro and La represent the initials of the patients whose serum was used to identify the reaction in agar diffusion). 3.3. ELISA Serum antibody levels to each of the above antigens were determined by ELISA as previously described [14-18]. Briefly, tested sera diluted 1/400 were added to antigen-coated polystyrene plates (Dynatech Laboratories, Alexandria, VA), incubated for 1 h at 23 °C, and washed. Alkaline phosphatase (AP)-conjugated anti-human Ig ( I g G + I g M + I g A ; Sigma Chemical Co.) was then added, incubated for 2 h at 23 °C, washed, and pnitrophenyl phosphate added to measure bound AP conjugate. The absorbance was read at 405 nM in a Dynatech micro ELISA reader. The mean _+ S.D. A for each antigen was calculated for each group of patients, and the values which reflect the binding of serum antibodies were compared to controls using the non-parametric Mann-Whitney U-test [19]. 4. Results

The mean _ S.D. A of the binding of sera from our group of 20 heterosexual controls to each antigen is shown in Table 1. These individual A numbers (different for each antigen) were generally closely TABLE 1 The binding of sera of the healthy control group to the autoantigens tested (mean absorbance units -+ S.D.). Mean _+ S.D. of controls Histone Cardiolipin ss-A ss-B Sm RNP Poly(1) Poly(G) ss-DNA ds-DNA

32

112+ 38 166 _+ 75 300_+ 144 316_+ 160 266 +_ 113 508 _+ 184 533 + 133 192+ 77 305 + 136 245 + 157

clustered. For each of the antigens tested, the binding of 2 - 4 cases was above the mean _ I S.D. of the controls (10-20%, median 15%); however, only rarely were A levels higher than mean + 2 S.D. encountered in the control group (median 5 %) (Table 2 and Figs. 1 and 2). In contrast, HIV-seronegative MHS already showed a marked increase in serum autoAb against histones, cardiolipin, ss-A ss-B; in the HIV ÷ cases, as many as half of the patients had a significant increase in these autoAb ( > mean of controls + 2 S.D.; Table 2). Serum antibodies to Sm were also similarly increased, but their incidence was lower than the autoAb to the antigens in the first group. The differences between the HIV + group and controls and between the HIV ÷ and H I V groups were both statistically significant ( P < 0.001 and p < 0.05, respectively). Interestingly, other commonly occurring autoAb to ss-DNA, ds-DNA, RNP, poly(I) and poly(G) were not usually demonstrated in either the H I V - or the HIV ÷ MHS. As can be seen from Table 3, the progression of HIV infection from the asymptomatic stage to ARC or AIDS did not considerably affect autoAb production, though a trend for fewer positive cases in AIDS was noted. 5. Discussion

This study demonstrates that patients with AIDS and ARC, as well as asymptomatic HIV ÷ MHS, have significantly elevated levels of several autoAb in their serum. Notable are the levels and prevalence of the antibodies to histones, cardiolipin, ss-A (Ro), ssB (La) and Sm, the titers of which were higher in the HIV-infected M H S than in normal controls ( P < 0.001). However, significantly elevated levels of these antibodies (except to Sm) were also demonstrated in HIV-seronegative MHS ( P < 0.05), though their prevalence in this group was clearly smaller. On the other hand, all four groups of MHS showed normal levels of antibodies to the other five autoantigens tested including ss-DNA, ds-DNA, poly(I), poly(G) and RNP. A number of investigators have noted an increased prevalence of many types of autoAb in ARC and AIDS patients. These include antibodies to lymphocytes, platelets, neutrophils, erythrocytes, sperm, immunoglobulins, as well as to varied nuclear and cytoplasmic antigens [20-23]. In contrast, others have not found increased autoAb in these patients

TABLE 2 Numbers (percents in parentheses) o f male homosexuals negative or positive for HIV-1 at different stages o f infection, and of heterosexual controls, with increased serum antibodies to a panel o f autoantigens a. Histone

Cardiolipin-ss-A

Controls (N = 20) 1 (5%) 0 H I V - MHS (N = 17) 4 (24%) 2 (12O7o) All HIV + MHS c (N = 39) 20 (51%) 19 (49070) pd < 0.05 < 0.05

SS-B

1 (5%) 4 (24%) 18 (46°70) < 0.05

Sm

RNP

2 (10%) 2 (10%) 4 (24°70) 3 (18070) 20 (51°7o) 7 (18070) < 0.05 < 0.05

Poly(I)/poly(G) b ss-DNA/ds-DNA b

1 (5%) 0-1 (0-5%) 2 (12070) 3 (18%) 2 (5070) 5 (13070) NS NS

2 (10%) 1 (6%) 1 (2.5070) NS

a Defined as levels of antibody binding which exceed the mean of the control group by more than 2 S.D. b The numbers for each of these 2 antigens were almost identical, c All HIV + MHS including asymptomatic seropositive cases, ARC and AIDS, as separately detailed below, d Difference between all HIV + and H I V - MHS; the significance o f the difference between controls and each of the HIV + MHS groups was