Journal of Clinical Immunology, Vol. 12, No. 2, 1992

Differential Isotype Expression and Binding Properties of T Cell-Reactive Antibodies in Human Immunodeficiency Virus (HIV) Infection FRANCO SILVESTRIS, 1'2 CLAUDIO AZZOLINI, 1 and FRANCO DAMMACCO t

Accepted: October 15, 1991

IgM-ALA in HIV infection.

Isotype and binding characteristics of T cell-reactive antilymphocyte antibodies (ALA) were investigated in 287 human immunodeficiency virus (HIV) + sera from patients with CDC II to IVC clinical disease. Using purified soluble T-lymphoblast (CEM cell line) membranes and an ELISA method, 29 HIV + sera showed significant reactions with this substrate and a selective expression of IgG-ALA was detected in 7 HIV ÷ sera. Subsequent microcytotoxicity assays, utilizing peripheral T lymphocytes and CEM ceils as targets, demonstrated no significant cytotoxic capability in such sera, whereas 12 of 17 HIV + serum samples with IgM-ALA ELISA reactivities showed a significant degree of killing in the Terasaki test. Further experiments of saturation of CD4 molecules on CEM extract by OKT4 monoclonal antibody (MoAb) induced a high inhibition of IgG-ALA binding to the T-cell membranes in only two IgG-ALA+ sera (No. 93, CDC III; No. 179, CDC II stage). Conversely, treatment of CEM membrane lysate with Leu3a MoAb, specific for the gp120 reactive domain of the HIV receptor, failed to prevent membrane binding in all seven of the IgG-ALA + sera. Following the adsorption of serum 93 on a T-cell membrane antigen affinity column, SDSPAGE analysis demonstrated that the predominant ALA material reacting with T-cell membranes was IgG with no detectable traces of IgM. These data provide evidence that ALA in HIV ÷ patients may be simultaneously or selectively expressed as IgG and/or IgM with different properties. While IgM-ALA show predominant cytotoxic activity, IgG-ALA may include anti-CD4 molecules. However, IgG binding to the C-terminal domain of native HIV receptor appears to occur at a lower rate than

KEY WORDS: HIV infection; CD4 receptor; anti-T-cell antibodies; T-helper defect.

INTRODUCTION Autoantibodies to structural proteins of T-cell membranes have been extensively described in sera from patients with human immunodeficiency virus (HIV) infection (1-7). Although different investigators have documented a partial specificity of antilymphocyte antibodies (ALA) to the HIV-1 receptor (8-10), their potential role in the immunopathogenesis of acquired immunodeficiency syndrome (AIDS) is still controversial. At present it is not clear whether ALA production in HIV + subjects is a generic effect of polyclonal B-cell activation (11, 12) or whether it has pathogenic relevance in aggravating the progressive decrement of CD4 ÷ cells during the course of disease (13, 14). Recent reports indicate that ALA binding to soluble, recombinant forms of CD4 molecule (rCD4) occurs in less than 20% of HIV-l-infected subjects and that their detection appears specific for HIV-1 infection (9, 15). Several hypotheses have been formulated to explain such anti-CD4 reactivities in these patients, and although these antibodies appear to be directed against a region of the viral receptor distinct from the virus-binding domain (10, I6), some workers have suggested that a putative conformational change in the C-terminal domains of CD4 molecules, induced by viral gpl20 binding, could lead to production of antibodies with antiCD4 specificity (17). Besides anti-CD4 reactivities, a panel of several lymphocyte-derived proteins has been described as molecular targets of ALA from HIV ÷ subjects (6, 8,

1Department of Biomedical Sciences and Human Oncotogy, University of Bad Medical School, Section of Internal Medicine and Clinical Oncology, 70124 Bad, Italy. 2To whom correspondence should be addressed at DIMO, Sez. Medicina Interna e Oncologia Clinica, Universifft di Bad, Piazza Giulio Cesare 11, 70124 Bad, Italy.

107 0271-9142/92/0300-0107506.50/0© 1992PlenumPublishingCorporation

108

18), but because of the different technical approaches adopted (i.e., selection of cellular target; binding methods including microcytotoxicity, cytofluorimetry, immunoblotting, etc.), the identities of such proteins and their cellular location (cell membrane or intracellular) still remain to be defined. In previous studies (I-6, 13, 23) detection of ALA in HIV infection has frequently been carded out using as target heterologous T- and B-cell subsets or unfractionated peripheral blood lymphocytes. Indeed, using heterologous peripheral lymphocytes we have recently documented by fluorescence-activated cell sorter (FACS) analyses (19) that the majority of ALA with cytotoxic capability in HIV ÷ individuals are IgM isotype and that they do not react with CD4 and CD8 molecules. In the present study we have extended such observations with the purpose of investigating the biological properties of T cell-reactive IgG-ALA in this disease. In order to avoid heterogeneity of the cellular target, as induced by using peripheral lymphocytes at different steps in maturation, we employed purified cell membranes from clonotypic T and B lymphoblasts, namely, CEM and Daudi cell lines. Results presented here appear to differentiate the biological significance of IgM and IgG with antilymphocyte reactivities in HIV infection. While cytotoxic properties to T lymphoblasts were found predominantly in IgM-ALA, anti-CD4 specificities on CEM plasma membranes were selectively noted in IgG-ALA, detectable in a minority of patients. MATERIALS AND METHODS Patient Sera and Controls. The study included 287 HIV ÷ and 67 healthy control sera. The HIV ÷ sera were randomly selected from a larger number of HIV-infected subjects who met the criteria for CDC stages II to IVc of HIV infection. These subjects were periodically evaluated as regards the peripheral expression of CD3 ÷, CD4 ÷, and CD8 ÷ cells by cytofluorimetry. Serum samples were obtained after informed consent before initiation of any antiviral treatment and none of the patients studied received rCD4 therapy at any time. Control sera were derived from healthy blood donors who were negative at the screening test for the detection of anti-HIV glycoproteins antibodies. The HIV ÷ sera were collected during the last 3 years and stored frozen at -80°C until use. Preparation o f Cell Membrane Extracts. Cell membranes from T (CEM line) and B (Daudi line)

SILVESTRIS, AZZOLINI, AND DAMMACCO

clonotypic lymphoblasts were obtained by nonionic detergent lysis of cells. Briefly, cells at a density of 2 × 106/ml were incubated in lysis buffer [phosphatebuffered saline (PBS), pH 7.4, containing 0.5% Nonidet P-40, 2 mM phenylmethylsulfonyl fluoride and 1 trypsin inibitory U/ml aprotinin] for 60 min on ice with occasional vortexing. Clearance of nuclei from the lysates, evaluated by phase-contrast microscopy, was obtained by centrifugation at 400g, whereas the subsequent ultracentrifugation at 100,000g was performed in order to remove cytoskeletal components and unsolubilized membranes. Preservation of CD4 antigenic sites on these extracts was assessed in subsequent enzyme-linked immunosorbent assay (ELISA) using two mouse monoclonal antibodies (MoAb) to the HIV receptor (OKT4, Ortho, Raritan, NJ; Leu3a, Becton-Dickinson, Mountain View, CA). Aliquots of these membrane extracts were then kept frozen at -80°C until use. ELISA for ALA Detection. Purified plasma membranes from CEM and Daudi cell lines were used as antigens in immunoenzymatic techniques. These ELISA methods were optimized in order to assure equivalent binding capability of both cellular extracts to the plates with a low specific background. Membrane extracts were incubated overnight on 96 wells plates at identical concentrations of 10 ~g/ml in carbonate buffer, pH 9.6. After washing the plates with PBS-Tween solution and subsequent blocking of nonspecific antigenic sites by 5% ~/-globulin-free calf serum-PBS solution for 1 hr at room temperature, the plates were incubated for 3 hr with the HIV + sera diluted 1/300. Peroxidase-conjugated goat anti-human IgG (Fc fragment specific) antiserum (Jackson, Immunoresearch Laboratories, West Grove, PA) was then added to the wells at the dilution of 1/50,000. In order to detect IgM with ALA specificity, a parallel test was performed simultaneously and was developed with a peroxidaseconjugated goat anti-human IgM antiserum at a similar dilution (Jackson). After color development of the test with OPD (O-phenylenediamine), absorbance of the plates was evaluated at 492 nm on a micro-ELISA reader (Titertek, Flow Laboratories, Irvine, Scotland). Each serum sample was tested in triplicate and each test included the healthy controls group. HIV + sera with absorbance higher than the mean + 3 SD of normal controls' values were considered positive for the expression of IgG-ALA and/or IgM-ALA, respectively. Several experiments were directed at examination of anti-CD4 reactivity of IgG-ALA. Parallel Journal of Clinical Immunology, Vol. 12, No. 2, 1992

T CELL=REACTIVE ANTIBODIES tN HIV INFECTION

ELISA tests were performed using seven IgG ALA + sera in identical assays including previous exhaustive saturation of CD4 antigenicity by two anti-CD4 MoAbs, namely, OKT4 and Leu3a, reactive to different epitopes of CD4 molecule (20, 21). The assays were performed by incubating the IgGALA ÷ sera on separate plates coated with CEM membranes and, in parallel, CEM membranes preincubated with OKT4 and Leu3a, respectively, diluted at 5 I~g/ml for 90 rain at room temperature. The test was then completed as described. Microcytotoxicity Assay. A basic Terasaki microcytotoxicity test (22) was applied to evaluate the lymphocytotoxic capability of the IgG-ALA + and IgM-ALA ÷ sera. The assay included as target peripheral T cells derived by nylon-wool column adsorption from two healthy donors in parallel with CEM lymphoblasts separately used in control tests. The various steps of these Terasaki methods were performed as previously described (19). Positive cytotoxicity tests were assigned to sera inducing a percentage higher than 50% target cells retaining the supravital Eosin Y dye. In several instances periodic evaluations of cytotoxicity properties in selected sera were also performed by the basic Terasaki technique. Cytotoxicity assays were performed by one experienced observer who was uninformed on the previous ELISA test results. Purification of ALA from HIV ÷ Sera. In one instance we attempted to purify ALA by adsorbing the relative serum on affinity Sepharose 4B column coupled with 7 mg of CEM membrane lysate. The material adsorbed was eluted by acidic buffer (0. I M glycine-HCl, pH 2.1) and the fractions were promptly adjusted to neutral pH by the addition of 1 ml of 1 M Tris buffer, pH 8.3. The content of the elution material was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) according to the standard description using a 12.5% polyacrylamide gradient gel (23). Statistical Analysis. Significance of positive detection of IgG and IgM-ALA was evaluated using the Student t test as compared to the group of normal controls. RESULTS

Data related to the detection of ALA by ELISA assays using lymphoblast plasma membranes as antigen are shown in Fig. 1. A total of 29 HIV ÷ sera (10.1% of the entire HIV ÷ group) was found to show significant reactivJournal of Clinical Immunology, Vol. 12, No. 2, 1992

109

ity against CEM membrane proteins (Fig. 1A). However, the 29 ALA + sera, derived from patients at nonidentical CDC-stage disease, appeared to express a different isotype reactivity. Only 7 sera (2.43%) showed high levels of IgG-ALA to the CEM membranes with no significant amounts of IgM-ALA; conversely, 17 sera (5.92%) were positive for IgM-ALA with low levels of IgG-ALA. A minor subgroup of five subjects (1.74%) was positive for both IgG and IgM with anti-CEM membrane reactivity. The results of the assay using Daudi cell membranes as antigen are illustrated in Fig. lB. As shown, a reduced number of HIV + sera (13/287 = 4.5%) were found to show significant reactions with this membrane extract, although absorbance of these sera appeared, in most instances, to be lower than the corresponding antiCEM reactivities. Among the anti-CEM IgG-ALA + sera, only serum 76 showed a low, although significant, difference between the anti-CEM and the anti-Daudi reactivities, whereas this difference was highly significant in the other six sera (P < 0.01 in all instances), thus confirming their predominant T-cell membrane specificity. Moreover, anti-CEM reactivities of selected IgM and IgG-ALA + sera were also tested in a parallel set of sequential ELISA assays in order to verify the effect of the dilution on binding specificity as well as the reproducibility of results. Data are expressed in Fig. 2. As can be seen, the intensity of the binding to CEM extract was significantly lower in the sera tested at a dilution higher than 1/400. The reproducibility of these tests was also assessed by repeating three times each test. However, slight variations of the OD values included in the SD of single samples (Fig. 2) were detected, thus suggesting a suitable reproducibility of our immunoenzymatic methods. The next set of experiments addressed the question whether, in ALA + subjects, predominant expression of IgG- or IgM-ALA, detected by ELISA, paralleled any cytotoxic activity in vitro. Accordingly, IgG-ALA + and IgM-ALA + sera were separately tested in Terasaki microcytotoxicity assays using peripheral blood T lymphocytes in parallel with CEM lymphoblasts. Results are shown in Fig. 3, which indicates a different pattern of reactivity in the IgG-ALA + sera compared to the group of IgM-ALA + subjects. A lack of cytotoxic activity was recorded in all the IgG-ALA + sera using both peripheral T lymphocytes and CEM cells, whereas a variable degree of killing was recorded in most IgM-ALA ÷ sera. This parallelism of positive results

110

SILVESTRIS, AZZOLINI, AND DAMMACCO

(A)

1.4

(B)

1.4

ga

~

1.2

1.2

94

1.0

1.0 215

r~

0.8

I

--

~

t--

6O Z 8,G.,F,OANOE

E

_d s

u.

O

40

p. z IJJ

(J

C¢ uJ

m.

20 o

i

T-|ymp

M

IgG-ALA + SERA

T-|ymphocytee

CEM

IgM-ALA + SERA

-lymphocyte8 NORMAL

DONOR

CEM SERA

Fig. 3. Comparative pattern of cytotoxic activity of 7 IgG-ALA + and 17 IgM-ALA+ sera from HIV + patients detected by microcytotoxicity Terasaki test using as cellular targets peripheral T cells from normal donors and CEM lymphoblasts. Joining lines indicate the variability of killing related to single samples. The assays included 18 normal donor sera as control.

Journal of Clinical Immunology, Vol. 12, No. 2, 1992

112

SILVESTRIS, AZZOLINI, AND DAMMACCO

Table I. Quantitative Variations of Peripheral CD4 Cells in 11 IgM-ALA ÷ and 5 IgG-ALA + HIV-Infected Subjects During an Observation Period of 18 Monthsa

Patient No.

Cytotoxicity/CD4 cell count (% +- SD) Observation period (months)

CDC stage

0

II III II IVc III III IVc II IVc III II

--/34 --/41.3 ND/47.4 +-/21.6 ++/20.5 +-/39.4 ++/16.3 +-/37.5 ++/23.4 ++/23.4 +-/51.3

II III III II IVc

--/39.9 --/20.7 +-/26.0 ND/32.3 --/12.4

IgM-ALA 19 39 52 58 65 118 119 171 180 201 207 IgG-ALA 76 93 176 179 215

4

8

12

18

- 3 +- 4 _ 3 -+ 1.3 -+ 4 +- 5 + 3 - 3 +- 5 -+ 4 - 4

ND/38.2 + 5 ND/43.5 -+ 8 +-/45.2 ± 3 ND ND/22.8 ~ 0.2 ND/43.2 - 4 ND/21.4 - 3 ND/45 - 2 ++/11.5 -+ 6 ND ND

ND ND +-/46 + 3 ND/12.1 --- 4 ND +-/38.5 --- 5 ++/19.8 +- 4 +-/46.3 +- 3 ND ++/31.8 - 4.3 ND/46.8 +- 4

- - / 3 0 . 3 -+ 2 ND/38.4 -+ 3 -/40.3 +_ 5 +-/9 +_ 3.8 ++/17.5 + 3 ND ++/12.4 - 6 ND ++/13.5 -+ 5 ND/27.3 -+ 6.4 +-/49.9 -+ 3

ND e.c. +-/41.5 - 0.3 +-/11.3 + 4 ++/19.8 --- 5 +-/30.4 - 4 e.c. +-/40.3 +-- 5 ND ++/12.9 - 2.3 + - / 3 9 -- 7

-+ 5 -+ 2 -+ 1 - 4 _ 5

- - / 2 7 . 4 -+ 5 ND +-/29.3 -+ 2 --/30.1 - 3 ND/10.5 - 1

ND ND ND/20.2 _+ 3 --/36 __ 5 ND

ND/30.5 - 2 ND +-/21.5 _+ 3 - - / 1 8 . 4 -+ 2 e.c.

e.c. --/29 -+ 5 +-/23.4 - 5 - - / 2 3 . 5 -+ 4 e.c.

aCDC clinical stage of each patient is referred to time 0. Cytotoxic properties of the sera were arbitrarily indicated as -+ depending on their percentual killing capability on heterologous lymphocytes in Terasaki assay. - - , 50%; + - , 50-70%; + +, 70% killing, ND, not done; e.c., evaluation in course.

in the anti-CEM adsorbed material from serum 93 represented possible evidence that IgG-ALA with apparent specificity to the OKT4-reactive domain of HIV receptor may selectively arise in some patients with HIV infection. DISCUSSION Results provided by this report represent an extension of our previous observations concerning serological expression and cellular targets of ALA in HIV-infected subjects (19, 24). By using purified T-cell membranes, derived from CEM clonotypic lymphoblasts, our data appear to suggest a different biological significance of IgG and IgM displaying antilymphocyte reactivity which may be concurrently or separately produced by HIV + subjects. IgM-ALA are more likely to represent cytotoxic antibodies without CD4 specific reactivity (19) and are the most suspected autoantibodies to play a pathogenic role in increasing the lymphopenia associated with the HIV infection (2, 4, 5, 8, 13). To this regard our data provide evidence that, in a few instances (sera 65, 119, and 201), high titers of IgM-ALA might correlate with a progressive decrease in the percentage of CD4 ÷ cells (Table I). Conversely, IgG-ALA seem to lack cytotoxic properties and their selective occurrence in HIV + sub-

jects is much lower than that of IgM-ALA. In addition, the putative specificity of IgO-ALA for CD4 epitopes occurs rarely and does not seem to involve the same domain of HIV receptor in all instances. Whether anti-CEM IgM and IgG ALA belong to the broad spectrum of the antilymphocyte specificities detected in other reports remains to be established. Previous studies indicate that ALA production in HIV + subjects is not to be considered merely a result of the enhanced B-cell activation (12, 25). Alternatively, it could represent an unusual autoimmune phenomenon in a relatively restricted number of patients who are believed at higher risk to develop AIDS since CD4-reactive ALA could interact to increase the helper T-cell defect (13, 14). However, the significance, molecular targets, and pathogenic relevance of ALA in HIV infection are still not clear. Data derived from our study suggest that in the broad pool of ALA, cytotoxic IgM may participate directly in the impairment of CD4 + cells by cytolytic effects. Despite their inability to react with the HIV receptor (19), our preliminary data (26) indicate a 42.7-kDa structure, detected on CEM lysate by immunoblotting techniques, as the preferential molecular target of such IgM-ALA. Further studies are in progress to identify this membrane Journal of Clinical Immunology, Vol. 12, No. 2, 1992

T CELL-REACTIVE ANTIBODIES IN HIV INFECTION

113

Reactivity against: CEM membrane CEM membrane saturated by Leu3a !

1.4

CEM membrane saturated by OKT4 1,2

E 0.8 c ~r

.. 0.6

d

o o.4 0.2

0

i

(76)

(82)

(93)

(176)

(179)

(215)

(223)

IgG-ALA sera Fig. 4. Reactivity of seven IgG-ALA + sera from HIV + subjects, obtained in ELISA tests, against intact CEM membrane extract and after different saturation of its CD4 antigenic sites by Leu3a and OKT4 MoAbs. Relative inhibition of the IgG binding to OKT4-reactive domains of the HIV receptor was detected in sera Nos. 93 and 179, whereas the other sera reacted with similar OD values to the original and to the CD4-saturated CEM membranes. The presence of unaltered Leu3a- and OKT4-reactive domains of CD4 molecule on CEM extract was evaluated before its saturation by parallel ELISA by the mentioned monoclonals. Confined areas indicate the range of OD reactivity related to each MoAb.

antigen, since an actin-like 43-kDa molecule, described on the surface of neutrophils (27), appears to react with antineutrophil antibodies in autoimmune neutropenia, sometimes associated with HIV infection (28). We observed that the selective occurrence of IgG-ALA in HIV + subjects was present in a small group of patients (2.43% of the entire population). In only two instances were we able to demonstrate IgG-ALA relative specificity for a domain of the viral receptor, located on the C-terminal immunoglobulin-like epitopes and recognized by the OKT4 MoAb. Similar data have been provided by other investigators who used fragments of rCD4 protein (9, 17) or, alternatively, insertion mutants of rCD4 (I0). In these studies a higher percentage of positively reacting HIV ÷ patients at different clinical stages with anti-CD4 specificity was reported (20%). However, these antibodies were not found to bind native CD4 structures on T cells (9). Our findings appear to suggest that autoantibodies with binding specificities similar to those exhibited by Journal of Clinical Immunology, Vol. 12, No. 2, 1992

OKT4 MoAb to native CD4 molecules rarely arise in HIV ÷ subjects and that these CD4-reactive ALA include predominantly IgG antibodies. Conformational changes of the CD4 C-terminal domain, theoretically induced by viral gpl20, have been proposed as inducers of anti-CD4-specific antibodies (17), but more data are required to support this hypothesis. A further comment is related to the significance of these anti-CD4 IgG as regards the defective helper function in HIV infection. Although antiCD4 antibodies have in the past been associated with a demonstrable helper defect in several HIV ÷ patients with CDC stages II, III, and IV disease (14), no evidence is actually available on the possible mechanism(s) affecting the helper function in these subjects. Other investigators have documented that antibodies to the homologous region of the HIV envelope protein gp41 cross-react with the t3 chain of human HLA class II antigens (29) and, moreover, with human interleukin-2 (30). These observations suggest that the helper activity of

114

SILVESTRIS, AZZOLINI,AND DAMMACCO

ACKNOWLEDGMENTS This work was supported in part by "Third AIDS Research Project" (1990), Italian Ministry of Health, ISS, Rome, Grant 5206 029. The authors thank Giuseppe P. Albergo for technical assistance.

REFERENCES

Fig. 5. SDS-PAGE pattern of ALA from serum 93 adsorbed on a CEM membrane affinity column compared to molecular weight (MW) standard markers (lane 1) and to purified human IgGk and IgMk monoclonal components (lanes 3 and 4). Electrophoretic separation of CEM-adsorbed ALA revealed a prevalent content of polyclonal IgG with no detectable traces of IgM. Additional, although weak, bands including albumin and another high MW protein were also present.

CD4 ÷ cells might be damaged by such autoantibodies independently of concomitant anti-CD4 antibodies. On the other hand, since the N-terminal domains of CD4 act as receptors for the gpl20 (31), it is unlikely that CD4-reactive IgG may interfere with the HIV binding or may arise from an antiidiotype response related to the gpl20 (16). The exact role of anti-CD4 IgG in HIV infection is still unknown and our present results as well as those of other studies are unable to assess their potential suppressive effect on helper cells or the antiviral function of these antibodies. In this context, a hypothetical modulatory effect on CD4 + cells, mediated by such molecules, cannot be excluded and further investigation is actually in progress in our laboratory to elucidate the possible influence of purified anti-CD4 IgG on biologic responses in vitro of noninfected CD4 ÷ lymphocytes.

1. Pruzanski W, Jacobs H, Lorne P, Laing MD: Lymphocytotoxic antibodies against peripheral blood B and T lymphocytes in homosexuals with AIDS and ARC. AIDS Res 1:211-220, 1984 2. Williams RC Jr, Masur H, Spira TJ: Lymphocyte reactive antibodies in acquired immune deficiency syndrome. J Clin Immunol 4:118-123, 1984 3. Tomar RH, John PA, Henning AK, Kloster B: Cellular targets of anti-lymphocyte antibodies in AIDS and LAS. Clin Immunol Immunopathol 37:37-47, 1985 4. Dorsett B, Cronin W, Chuma V, Ioachim HL: Antilymphocyte antibodies in a patients with the acquired immune deficiency syndrome. Am J Med 78:621-626, 1985 5. Ozturk GE, Kohler PF, Horsburgh CR Jr, Kirkpatrick CH: The significance of anti-lymphocyte antibodies in patients with acquired immunodeficiency syndrome (AIDS) and their sexual partners. J Clin Immunol 7:130-139, 1988 6. Warren RQ, Johnson EA, Donnelly RP, Lavia MF, Tsang KY: Specificity of anti-lymphocyte antibodies in sera from patients with AIDS related complex (ARC) and healthy homosexuals. Clin Exp Immunol 73:168-173, 1988 7. Daniel V, Schimpf K, Opelz G: Lymphocyte auto- and allo-antibodies in HIV positive hemophilia patients. Clin Exp Immunol 75:178-t83, 1989 8. Strickler RB, McHugh TM, Moody DJ, Morrow WJW, Stites DP, Shuman MA, Levy JA: An AIDS-related cytotoxic autoantibody reacts with a specific antigen on stimulated CD4 T cells. Nature 327:710-713, 1987 9. Chams V, Jouault T, Fenouillet E, Gluckman JC, Klatzmann D: Detection of anti-CD4 autoantibodies in the sera of HIV-infected patients using recombinant soluble CD4 molecules. AIDS 2:353-361, 1988 10. Kowalski M, Ardman B, Basiripour L, Lu Y, Blohm D, Haseltine W, Sodroski J: Antibodies to CIM in individuals infected with human immunodeficiency virus type 1. Proc Natl Acad Sci USA 86:3346-3350, 1989 11. Pahwa S, Pahwa R, Good RA, Gallo RC, Saxinger C: Stimulatory and inhibitory influences of human immunodeficiency virus on normal B lymphocytes. Proc Natl Acad Sci USA 83:9124-9132, t986 12. Yarchoan R, Redfield RR, Broder S: Mechanisms of B cell activation in patients with the acquired immuno deficiency syndrome and related disorders: Contribution of antibodyproducing B cells, of Epstein-Barr virus-infected B cells and of Ig production induced by human T cell lymphotropic virus, type III/lymphadenopathy-associated virus. J Clin Invest 78:439-447, 1986 13. Dorsett BH, Cronin W, Ioachim HL: Presence and prognostic significance of antilymphocyte antibodies in symptomatic and asymptomatic human immunodeficiency virus infection. Arch Intern Med 150:1025-1028, 1990

Journal of Clinical Immunology, Vol. 12, No. 2, 1992

T CELL-REACTIVE ANTIBODIES IN HIV INFECTION

14. Weimer R, Daniel V, Zimmermann R, Schimp K, Opelz G: Autoantibodies against CD4 cells are associated with helper defects in human immunodeficiency virus-infected patients. Blood 77:133-140, 1991 15. Thiriart C, Goudsmith J, Schellekens P, Barin F, Zagury D, De Wilde M, Bruck C: Antibodies to soluble CD4 in HIV-I infected individuals. AIDS 2:345-351, 1988 16. Wilks D, Walker LC, Habeshaw JA, Youle M, Gazzard B, Dagleish AG: Anti-CD4 autoantibodies and screening for anti-idiotypic antibodies to anti-CD4 monoclonal antibodies in HIV-seropositive people. AIDS 4:113-118, 1990 17. Sekigawa I, Groopmen JE, Allan JD, Ikeucbi K, Biberfield G, Takatsuki K, Byrn RA: Characterization of autoantibodies to the CD4 molecule in human immunodeficiency virus infection. Clin Immunol Immunopathol 58:145-153, 1991 18. Ardman B, Sikorski MA, Settles M, Stauton DE: Human immunodeficiency virus type 1-infected individuals make autoantibodies that bind to the CD43 on normal thymic lymphocytes. J Exp Med 172:1151-1158, 1990 19. Silvestris F, Edwards BS, Sadeghi OM, Frassanito MA, Williams RC Jr, Dammacco F: Isotype, distribution and target analysis of lymphocyte reactive antibodies in patients with human immunodeficiency virus infection. Clin Immunol Immunopathol 53:329-340, 1989 20. Reinherz EL, Kung PC, Goldstein G, Schlossman SF: Further characterization of human inducer T cell subset defined by monoclonal antibody, J Immunol 123:2894-2903, 1979 2t. Evans RL, Wall DW, Platsoucas CD, Siegal FP, Fikrig SM, Testa CM, Good RA: Thymus dependent membrane antigens in man: Inhibition of cell-mediated lympholysis by monoclonal antibodies to the TH2 antigen. Proc Natl Acad Sci USA 78:544-553, 1981 22. Terasaki PI, McClelland JD: Microdroptet assay of human serum cytotoxins. Nature (London) 204:998-1000, t964

Journal of Clinical Immunology, Vol. 12, No. 2, 1992

115

23. Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 277:680-683, 1970 24. Silvestris F, Frassanito MA, Dammacco F: Antilymphocyte antibodies (ALA) and human immunodeficiency virus (HIV)-induced infection: Clinical and pathogenic study. FASEB J 2:A470, 1988 25. Pollack MS, Callaway C, Leblanc D, Robinson HM, Mastrota F, Dupont B, Koziner B: Lymphocytotoxic antibodies to non-HLA antigens in the sera of patients with acquired immunodeficiency syndrome (AIDS). In Non-HLA Antigens in Health, Aging and Malignancy, E Cohen, DP Singal (eds). New York, A. Liss, 1983, pp 209-214 26. Silvestfis F, Dammacco F: Anticorpi anti-linfocitari nell'infezione da HIV: caratterizzazione della reattivita molecolare. In AIDS e Sindromi Correlate, F De Lorenzo, M Piazza (eds). Bologna, Italy, Monduzzi, 1990, pp 173-179 27. Hartman KR, Mallet MK, Nath J, Wright DJ: Antibodies to actin in autoimmune neutropenia. Blood 75:736-743, 1990 28. Murphy MF, Metcalfe P, Waters AH, Linch DC, CheingsonPopov R, Came C, Weller ID: Immune neutropenia in homosexual men. Lancet 1:217-222, 1985 29. Golding H, Shearer GM, Hillman K, Lucas P, Manischewitz J, Zajac RA, Clerici M, Gress RE, Boswell RN, Golding B: Common epitope in human immunodeficiency virus (HIV)Igp41 and HLA class II elicits immunosuppressive autoantibodies capable of contributing to immune dysfunction in HIV I-infected individuals. J Clin Invest 83:1430-1441, 1989 30. Bost KL, Hahn BH, Saag MS, Shaw GM, Weigent DA, Blalock JE: Individuals infected with HIV possess antibodies against IL-2. Immunology 65:611-622, 1988 31. Arthos J, Deen KC, Chaikin MA, Fornwald JA, Sathe G, Sattentau QJ, Clapham PR, Weiss RA, McDougal JS, Pietropaolo C, Axel R, Truneh A, Maddon PJ, Sweet RW: Identification of residues in human CIM critical for the binding of HIV. Cell 57:469-481, 1989