Immunology Letters 51 (1996) 39-43

HIV- 1 -specific immunity in persistently seronegative risk for HIV infection

individuals at high

Albert0 Beretta”,“, Lucinda Furci”, Samuele Burastero”, Antonio Cosma”, Maria Elena Dinelli”, Lucia Lopalco”, Claudio De Santis”, Giuseppe Tambussi”,

Emily Carrow”, Sergio Sabbatam Ob, Mario Clerici”, Adrian0 Lazzarin”, Antonio G. Siccardi” “DIBIT and I@ctious Diseases Clinic, S. Raffaele Scientific Instirute, Via Olgertina 58, 20132 Milano. Ital)

bOspedak Maggiore di Bologna, Bologna. Italy ‘Catredra di Immunologia. Uniuersitb di Milano, Milano, Itall

Accepted I5 March 1996

Abstract

A growing number of reports indicates that certain groups of individuals who almost certainly have been exposed to human immunodeficiency virus (HIV), yet continue to exhibit no signs or symptoms of infection, often have subtle evidence of specific immunity. We studied such a high-risk (HR) cohort of persistently seronegative individuals with histories of long-term sexual exposure to an HIV-infected partner to look for evidence of both humoral and cellular immunity that might have been induced by exposure to the virus. Twenty-three heterosexual and four homosexual monogamous couples with discordant HIV status were included in the study. Twelve of the HR partners were studied for in vitro stimulation of peripheral blood mononuclear cells (PBMC) by HIV envelope-derived peptides. All 12 responded overwhelmingly to a peptide containing the fifth conserved region of gp120. By generating and cloning T cell lines specific for this peptide, we concluded that in these individuals the T cell response to the envelope is mainly focused on the carboxy-terminus region of gpl20 and is characterized by an oligoclonal expansion of CD4+ T cells expressing the same TCR. Eighteen HR partners and 37 HIV-l seropositive subjects were tested for the presence of anti-CD4 antibodies (anti-CD4 Abs) using a recombinant CD4-based enzyme-linked immunosorbent assay (ELISA). Anti-CD4 Abs were detected in eight of the HR partners (six confirmed by Western blot) and in nine of the HIV-1 seropositive subjects (eight confirmed by Western blot). Results from binding competition assays with a panel of monoclonal anti-CD4 Abs suggested that the anti-CD4 Abs detected in the HR partners are directed toward epitopes that are induced by gp120 binding. Twenty-seven of the HR partners were tested for the presence of antibodies that cross-react with HLA class I and gpl20 (anti-HLA Abs). Anti-HLA Abs were detected in 16 of the HR partner sera and in 4/94 sera from a control population of normal healthy blood donors. Taken together. the results suggest that in some individuals with a history of long-term exposure to HIV, specific immunity may develop in the absence of overt infection. The common trigger for these responses is gpl20.

Keywords: HIV; HLA; Protective immunity

1. Introduction

A paradoxical feature of human immunodeficency

virus type-l (HIV-l) infection is that the relatively

robust humoral and cellular immune responses induced

in the majority of patients are nevertheless insufficient to stem progression of the disease [l]. On the other hand. the fact that a subset of infected individuals

____ * Corresponding author. Present address: Centre IntCgrk de

Recherches Biocliniques SW le SIDA. HBpital Saint Joseph, 185 rue

Raymond Losserand, 75014 Paris, France. Tel.: + 33 1 44123174

(secrttariat 44123183): fax: +33 1 44123270.

remain asymptomatic for much longer periods than average indicates that in certain populations, at least some degree of protective immunity probably does develop. Moreover, it is now known that exposure to HIV does not always lead to overt infection and its conventional manifestations. Various groups of individ- uals who almost certainly have been exposed to HIV, such as long-term sexual partners of HIV-infected indi- viduals, injection drug users (IDU) and recipients of needle stick injuries, while continuing to exhibit none of the usual signs or symptoms of infection, often have subtle yet reproducible evidence of specific immunity when their lymphocytes are studied closely in vitro. For

0165-2478/96/$12.00 0 1996 Elsevier Science B.V. All rights reserved

PII SO1 65-2478(96)02553-9

40 A. Beretta et al. 1 Immunology Letters 51 (1996) 39-43

example, HIV-specific in vitro peripheral blood mononuclear cell (PBMC) proliferation and IL-2 pro- duction [2,7], as well as HLA class I-restricted cytotoxic T cells specific for HIV early proteins [8-lo] have been variously reported in these groups. Since HLA class l-restricted cells require viral integration to be trig- gered, this latter finding suggests that viral replication can take place within cells in the absence of a humoral response. It is possible then, that one or more protec- tive immune mechanisms become activated in certain people, thereby providing some control of HIV, per- haps sufficient to divert the infection towards a less virulent course, or possibly even to a latent state. It is also possible that such mechanisms are active in all infected individuals, but for some reason they operate less efficiently and thus are not effective in the long term.

In addition to T cells, several types of ‘unconven- tional’ antibody responses to infection, e.g. those di- rected against cellular rather than viral proteins, are potentially involved in protective immunity. For exam- ple, antibodies against cellular proteins have been asso- ciated with protection in macaques that had been vaccinated with uninfected human cells and challenged with simian immunodeficiency virus (SIV) grown in the same cells [ll]. Also in the SIV model, passively trans- ferred antibodies specific for MHC class II antigens can confer protection against viral challenge, as can anti- MHC antibodies induced by vaccination with purified HLA class I or class II molecules [12- 151. Another type of unconventional antibody response associated with MHC antigens has been reported in humans. These antibodies, which are cross-reactive for specific epitopes shared by both the HIV envelope and certain MHC class I molecules, have been found in HIV-infected individuals [16], in healthy volunteers vaccinated with recombinant envelope proteins [17] and, more recently, in individuals who are at high risk for HIV infection but remain seronegative [18]. Yet a third type of anti- bodies apparently induced by HIV infection includes those directed against the CD4 moiety (ret). These antibodies have been detected in approximately 20% of HIV-infected individuals at different stages of disease

v91. All of these findings seem to suggest that both cellu-

lar and humoral immunity in HIV infection do indeed have protective components, but they are probably rendered ineffective during the normal course of infec- tion by a more powerful array of competing responses. More study of the exposed but apparently resistant groups, and of long-term survivors, is needed to deter- mine exactly what triggers potentially protective im- mune mechanisms, and in what context they might be effective in containing infection.

We have been following for several years such a high-risk (HR) cohort of persistently seronegative indi-

viduals with histories of long-term sexual exposure to an HIV-infected partner. We summarize here our find- ings of HIV envelope-specific T cells as well as anticel- lular antibodies (anti-HLA and anti-CD4) in this unusual group.

2. Results and Discussion

We enrolled in the study 23 heterosexual and four homosexual monogamous couples with discordant HIV status. Inclusion in the study required that each couple had more than 2 years of regular unprotected sexual intercourse before the awareness of HIV-positive status in one partner (the index case), and no other risk factors during that period or afterward. The seroposi- tive subjects were monitored regularly for number of CD4-positive cells, serum p24 antigen levels (Coulter HIV-l p24 antigen; Coulter Corporation, Hialeah, FL) and clinical status. Seronegative subjects were tested initially for serum p24 antigen and HIV DNA by polymerase chain reaction (PCR), thereafter they were tested for HIV antibodies every 2 months. 20/27 cou- ples were typed at the HLA class I locus by standard serological techniques.

2.1. Oligoclonal T cell responses to selected envelope epitopes

HIV-1 T cell immunity has already been reported in several high risk groups [2-lo]. In this study, we have attempted to identify the subsets of T helper cells that are expanded following exposure to envelope antigens. For induction of specific T cells we used a panel of HIV-l gp120 peptides (Cl, V3, C4) representing three regions of the molecule previously reported to contain T cell epitopes [20]. In addition, we used a 26-residue peptide (C5) that contains the fifth conserved region of gp120. This region displays a structural homology to HLA class I heavy chains [21] and contains a stretch of 12 amino acids with a high degree of sequence homol- ogy to HLA class II [22]. Peptides were used to stimu- late in vitro PBMC from 12 seronegative high-risk (HR) partners and 10 low-risk subjects as controls. A donor was considered positive for the presence of envel- ope-specific T cells when a stimulation index greater than 2 was obtained upon a single round of stimulation with a given peptide.

The initial results confirmed earlier findings that ap- parently uninfected HR individuals frequently manifest specific T cell immunity to HIV-l. In our hands, all HR donors tested responded to at least one gp120 peptide, while only two of the 10 control PBMC proliferated weakly to the C5 peptide. A particularly striking fea- ture in the HR PBMC was the preponderance of prolif- erative responses to the C5 peptide. It induced

A. Beretta et al. 1 Immunology Letters 51 (1996) 39-4-Z 41

proliferation in PBMC from IO/l2 HR donors, whereas peptides Cl, V3 and C4 stimulated 219, l/9 and O/9 respectively. In addition, in all donor PBMC the mag- nitude of the proliferative response to the C5 peptide was at least 3-times greater than the response to the other peptides.

Primary T cell lines were then generated from the PBMC cultures of five HR donors and two controls using the C5 and Cl peptides. C5 peptide-specific pri- mary T cell lines were cloned and the pattern of TCR gene expression and of cytokine production by each single clone were analyzed. All clones obtained were CD4+ /CD8 ~. Upon stimulation with the C5 peptide, 50% of the clones produced THl cytokines (IFN-7 and IL-2), 25% produced TH2 cytokines (IL-4 and IL-IO), and the remaining clones produced both types of cy- tokines (THO). The expression of given VP genes was tested by FACS analysis with specific monoclonal anti- bodies, and in some cases confirmed by PCR amplifica- tion and sequencing. In these cases, the Vcr genes were also analyzed. Interestingly, all clones independently generated from a single individual expressed the same V/j’ clonotype identified by the N-region sequence. When available, the Vn sequences also indicated the TCR identity of the clones. In contrast to the CS-spe- cific clones, TCR gene expression by clones specific for the other gp120 peptides was not restricted to a single clonotype. These results indicate that the proliferative responses of the HR donor cells to the C5 peptide, reflected an in vivo expansion of cells bearing identical receptors.

Three T cell epitopes recognized by the majority of the clones were identified by epitope mapping with a panel of overlapping 12-residue peptides covering the C5 region.

Thus, it appears that in HR partners the T cell response to the envelope is mainly focused on the carboxy-terminus region of gp120 and is characterized by an oligoclonal expansion of CD4’ T cells express- ing the same TCR.

3 7 Anti-CD4 antibodies _._.

Eighteen HR partners and 37 HIV-1 seropositive subjects were tested for the presence of anti-CD4 Abs using a recombinant CDCbased ELISA. Anti-CD4 Abs were detected in S/IS HR partners and 9/37 HIV-l seropositive subjects. Anti-CD4 Abs were confirmed by Western blot in six of the HR partners and in eight of the HIV-1 seropositive subjects. In an attempt to define the epitope specificities of the anti-CD4 Ab positive sera, a flow cytometry binding interference assay was used to test the ability of sera or IgG fractions to inhibit the binding of three anti-CD4 monoclonal anti- bodies. Sera from both HR partners and HIV-I seropositive subjects inhibited mab55 (specific for a

conformational CD4 epitope induced by gp120 bind- ing) to an equal extent. In contrast, only the HIV-l seropositive sera inhibited binding by mab94 (specific for the gp120 binding region of CD4). Binding of OKT4 mab (specific for a CD4 epitope not involved in gp120 binding) was inhibited by sera from both groups but to a lesser extent than the other mabs. These results would suggest that the anti-CD4 antibodies detected in the HR partners are directed toward an epitope that is induced by gp120 but is distant from the binding site.

2.3. Anti-HLA untibodies

A consistent fraction of the total antibody response to the envelope is directed toward epitopes of the fifth conserved region of gp120 (C5) [23]. We have previ- ously shown that the anti-C5 antibodies generated in HIV-infected individuals are cross-reactive to the z-he- lix of HLA class I heavy chains [16]. The molecular mimicry was revealed by mapping with monoclonal antibody M38 [21]. The cross-reactive anti-HLA Abs have been found in individuals with full-blown AIDS but not in HIV-positive asymptomatic patients [24]. Vaccination of healthy volunteers with recombinant envelope also reportedly induced anti-HLA class I anti- bodies 1173. During our early studies on HIV-positive patients, we noted similar reactivities in samples from HIV-negative injection drug users (IDU). A study on a second cohort of seronegative IDU revealed the pres- ence of anti-HLA Abs in 33% of the individuals tested. Furthermore, all anti-HLA Ab-positive IDU also ex- hibited T cells specific for envelope peptides suggesting that the presence of anti-HLA Abs was directly linked to exposure to envelope antigens [18].

We tested 27 HR partners for anti-HLA Abs using a flow cytometry assay that measures serum inhibition of M38 binding to cell surface HLA class I. Anti-HLA Abs were detected in 16/27 (59%) HR partner sera and in 4/94 (4.2%) sera from a control population of normal healthy blood donors. The specificity of HLA Abs-pos- itive sera for HLA class I was further confirmed by immunoprecipitation followed by Western blot with a HLA class-I-specific mab. HPLC fractionation of eight HLA Abs-positive sera indicated that the binding activ- ity resides in the IgG fraction. Immunoprecipitation of HLA class I by anti-HLA Ab-positive sera followed by isoelectric focusing revealed two bands corresponding to HLA-Cwl and HLA-A2. HLA C and HLA-A2 had been previoulsy identified as the HLA class I alleles that express the gp120 homologous epitopes recognized by M38 [21]. As these specificities are not what would be expected if the antibodies had been induced by the HLA antigens expressed by their partner’s cells, one possibility is that these antibodies which share specific- ity for HIV were instead induced by exposure to envel- ope antigens.

42 A. Beretta et al. I Immunology Letters 51 (1996) 39-43

3. Conclusions

The HR partners in our study displayed three appar- ently distinct types of immune reactivities: anti-CD4 Abs; anti-HLA class I Abs; and helper T cells specific for envelope peptides. While all of these reactivities may be present in seropositive individuals, it has been difficult to sort out their roles from the other abundant immune responses to infection. Their presence in indi- viduals that otherwise appear to be uninfected, despite long-term exposure to the virus, indicates that they may be protective. Further, they may reflect a type of immu- nity that is triggered only under certain circumstances, such as chronic exposure to low doses of viral antigens or alternatively, exposure to alloantigens. In the cohort we studied, for example, repetitive exposure to infected or uninfected cells from a seropositive partner might theoretically induce a kind of personalized protective immunity in the seronegative partner. So far we have not found any evidence for this: the anti-HLA Abs detected in the seronegative partners were not specific for the seropositive partner’s HLA but rather for the HLA class I alleles that bear the gp120 cross-reactive epitopes.

We identified instead a pattern of anti-cellular im- mune responses that all appear to be induced by expo- sure to gp120. The specificity of the anti-CD4 antibodies found in this cohort imply that a unique conformational epitope of CD4 is apparently induced by its interaction with gp120. Regions of gp120 itself may also become exposed due to CD4 binding, uncov- ering epitopes normally hidden in the native oligomeric form of the molecule. Using conventional serological tests based on native gp 120, antibodies generated against these epitopes would normally go undetected, and in fact, we were first able to detect them by virtue of their shared specificity for HLA class I. These anti- bodies recognize common sequences within the cr-heli- cal regions of HLA class I heavy chains and the C5 region of gp120 [21], one of several instances of molec- ular mimicry seen with HIV. Further, the unusually strong T cell response against the same region of the envelope in the same individuals may be another conse- quence of the molecular mimicry between HLA anti- gens and gp120. T cells expressing receptors for allo-MHC antigens may be recruited to the site of infection by antigen-presenting cells that present viral peptides. Due to the high frequency of alloreactive T cell precursors present in an unprimed repertoire, the extent of the response may be sufficient to clear HIV-in- fected cells or to provide help for other protective mechanisms.

Most of the seronegative partners have now been followed for at least 2 years. As they were instructed at the beginning of the study to practice safe sex, it can be assumed that most of them are no longer being exposed

to the virus by their partner. Nevertheless, the specific immune responses have persisted throughout the study period, raising the possibility that the seronegative part- ners actually harbor the virus, perhaps in a latent state that continuously exposes them to low levels of anti- gens.

These findings provide encouraging evidence that some components of the immune response to HIV are potentially beneficial to the host. We need now to identify the conditions for the generation and mainte- nance of these responses in order to exploit them for the development of effective vaccines.

References

[I] Fauci, A.S. (1993) Science 262, 101 l-1018. [2] Ranki, A., Mattinen, S., Yarchoan, R., Broder, S., Ghrayeb, J.,

Lahdevirta, J. and Krohn, K. (1989) AIDS 3, 63-69. [3] Borkowski, V., Krasinski. K., Moore, T. and Papangelou, V.

[41

PI

PI

171

PI

[91

PO1

[ill u 21

I131

[I41

(1990) AIDS Res. Hum. Retrovir. 6, 6733678. Clerici, M., Berzofsky, J.A., Shearer, G.M. and Tackett, C.O. (1991) J. Infect. Dis. 164, 178182. Clerici, M., Giorgi, J.V., Chou, CC., Gudeman, V.K., Zack, J.A., Gupta, P,. Ho, H.N., Nishanian, P.G., Berzofsky, J.A. and Shearer, G.M. (1992) J. Infect. Dis. 165, 1012-1019. Clerici, M., Sison, A.V., Berzofsky, J.A., Rakusan, T.A., Brandt. C.D, Ellaurie, M., Villa, M., Colie, C., Venzon. D.J. and Sever, J.L. (1993) AIDS 7, 142771433. Clerici. M., Levin, J.M., Kessler, H.A., Harris, A., Berzofsky, J.A., Landay, A.L. and Shearer, G.M. (1994) JAMA 27 1,42-46. Chenyer, R., Langlade-Demoyen, P., Marescot, M.R. and Plata, F. (1992) Eur. J. Immunol. 22, 2211-2217. Rowland-Jones, S.L., Nixon, D.F., Aldhous, MC.. Gotch, F., Ariyoshi, K., Hallam, N., Kroll, J.S., Froebel. K. and McMichael, A. (1993) Lancet 341, 860-863. Langlade-Demoyen, P., Ngo-Giang-Huong, N., Ferachal. F. and Oksenhendler, E. (1994) J. Clin. Invest. 93. 1297-1297. Stott. E.J. (1991) Nature 353, 393-394. Chan, W.L., Rodgers, A., Hancock, R.D.. TatIs, F., Kitchin, P., Farrar, G. and Liew, F.Y. (1992) J. Exp. Med. 176, 1203-1207. Chan, W.L.. Rodgers, A., Grief, C., Almond, N.. Ellis, S., Flanagan, B., Silvera. P., Bootman, J., Stott, J., Kent, K. and Bomford. R. (1995) AIDS 9, 223-228. Stott, E.J., Almond, N., West, W., Kent, K., Cranage, M.P. and Rud, E. (1995) in: 9’ Colloque des Cent Gardes (M. Girard and B. Dodet, Eds.). Fondation Marcel Merieux p. 2199224.

1151 Arthur, L.O., Bess, J.W. Jr, Urban, R.G., Strominger, J.L., Morton, W.R., Mann, D.L., Henderson, L.E. and Benveniste, R. (1995) J. Virol. 69, 311773124.

[16] De Santis, C., Lopalco, L., Robbioni, P.. Longhi, R., Rappocci- 010. G., Siccardi, A.G. and Berettd, A. (1994) AIDS Res. Hum, Retrovir. 10, 157- 162.

[l7] De Santis, C., Robbioni, P., Longhi. R., Lopalco, L., Siccardi, A.G.. Beretta, A. and Roberts. N.J. (1993) J. Infect. Dis. 168. 139661403.

118) Beretta, A.. Weiss S.H., Rappocciolo, G. et al. (1996) J. Inf. Diseases 173, 472-476.

[l9] Muller, C., Kukel, S. and Bauer, R. (1993) Immunology 79, 2488254.

1201 Clerici, M., Stocks, N.I., Zazajac, R.A. et al. (1989) Nature 339. 383385.

A. Berettu et al. 1 Immunology Letters 51 (1996) 39-43 43

[21] Lopalco. L., De Santis, C., Meneveri, R., Longhi, R., Ginelli, E., Grassi. F., Siccardi, A.G. and Beretta, A. (1993) Eur. J. lm- munol. 23. 2016-2021.

[22] Habeshaw, J.A., Hounsell, E. and Dalgleish, A. (1992) Immunol. Today 13, 207-210.

[23] Palker, T.J.. Matthews, T.J., Clark, M.E., Cianciolo, G.J., Ran-

dall, R.R., Langlois, A.J.. White, G.C., Bijan, S., Snyderman, R., Bolognesi, D.P. and Barton, F.H. (1987) Proc. Natl. Acad. Sci. USA 84. 2479.

[24] Borghi. O., De Santis, C., Barcellini. W., Lopalco, L., Fain, C., Lazzarin, A.. Siccardi, A.G., Meroni, P.L., Zanussi, C. and Beretta, A. (1993) J. AIDS 6. I114~1119.