Cross-Reactive T-Helper Responses in Patients Infected with Different Subtypes of Human Immunodeficiency Virus Type 1

Abstract

Immunization with a recombinant glycoprotein 160 envelope immunogen derived from a virus of genetic subtype B induced strong specific T-helper cell responses in asymptomatic human immunodeficiency virus (HIV) carriers infected with subtypes B to G. This indicates that the HIV-specific T-helper immunity, which is the basis for development of antibodies and cytotoxic T lymphocytes, can be improved by both homologous and heterologous antigens. It also suggests that a particular immunogen can be effective against many different HIV strains.

The extraordinary genetic diversity and rate of replication of human immunodeficiency virus type 1 (HIV-1) are major problems in the search for a prophylactic vaccine and in antiviral chemotherapy (5, 12). In many viral infections, immunity to the envelope and outer proteins is critical for viral clearance and control. The HIV infection is characterized by a continuous depletion of circulating CD4-positive cells, damaging the T-helper-cell function (14, 21, 23). This function is necessary for both memory antibody responses and clonal expansion of specific cytotoxic T lymphocytes (CTLs). T-cell recognition of envelope proteins and subsequent cellular activation have been detected in only a minority of HIV-infected individuals (3, 9, 20, 21). Following immunization with recombinant glycoprotein 160 (rgp160), HIV-specific cell-mediated lymphoproliferative responses and stabilization of CD4 values have been demonstrated in asymptomatic HIV carriers (9, 16, 17, 20, 22). While CD8⁺ cytotoxic activity towards target cells may be cross-reactive (7), very little is known about cross-reactivity of CD4⁺ T-helper-cell responses. Cross-reactive T-cell proliferative responses to more conserved epitopes of the virus, such as p24 (13), and restricted regions of the envelope have been described (4), but the relations between different subtypes regarding initial priming of reactivity have not been thoroughly investigated.

In early HIV disease, antiretroviral therapy has resulted in HIV-specific T-cell responses associated with long-term control of viremia (19). Highly active antiretroviral treatment (HAART) including multiple drugs targeting both the HIV-1 reverse transcriptase (RT) and protease generates a reduction in viral load and an increase in the number of circulating CD4⁺ T cells in HIV-1-infected individuals. However, in chronic infection, these CD4⁺ T cells do not seem to proliferate in response to HIV antigens (2, 9, 15). The exception is when very early antiretroviral treatment can be given (19). Consequently, the addition of HIV-specific immunization could be of benefit for HAART-treated patients. Our study aimed to investigate the efficacy of such immunizations. We investigated whether specific T-cell responses could be induced by immunization with rgp160 derived from a virus of genetic subtype B in patients also infected with other HIV-1 subtypes.

Sequencing was performed to determine the subtypes in 36 HIV-1-infected patients. The V3 region (400 bp) of the HIV-1 env gene was amplified directly from crude lysates of uncultured peripheral blood mononuclear cells (PBMCs) by PCR with nested primers and directly sequenced as described earlier (1, 10). The genetic subtype of the sequence fragments was compared to subtype reference sequences by using phylogenetic tree analysis (8). Twenty-seven of the patients clustered into subtype B, four into subtype C, two into D, and one each into subtypes E, F, and G. All patients had CD4 values above 200 cells/μl at the start of rgp160 immunization (mean, 439; standard deviation, 181). There was no correlation between CD4⁺ cell numbers and viral load or proliferation (data not shown). The participants received either multiple vaccinations (injections every 2 to 3 months) with rgp160 of subtype B (MicroGeneSys [now ProteinSciences], Meriden, Conn.) formulated in aluminium phosphate as adjuvant (29 patients) or placebo (aluminium phosphate) (seven patients) (9, 20). All patients were naive to protease inhibitors, but many were nucleoside analogue experienced. Cells from HIV-negative persons (over 100 patients, 400 samples), as well as from nonimmunized HIV-positive patients (over 100 patients, 400 samples), were used as controls in the T-cell proliferation assays (9).

T-cell responses were measured by using a proliferation assay (9). All assays were performed with fresh PBMCs. The antigens were as follows: rgp160 (the same as the vaccine) at a concentration of 1 μg/ml (0.1 μg/well), rgp120 at 10 and 1 μg/ml, and baculovirus control protein at 1 μg/ml. Envelope gp120 from virions of different subtypes (A, B, C, D, and E) was purified by affinity chromatography using Galantus nivalis agglutinin as described earlier (6). These native HIV envelope gp120 preparations were used at concentrations of 2 and 0.2 μg/ml. The recall antigen tetanus toxoid was also included as a control in all assays (tetanus toxoid 207-1, Seph 200, 910904; SBL Vaccin AB, Stockholm, Sweden) at a concentration of 10 μg/ml (1 μg/well). The mitogen phytohemagglutinin (PHA) (HA16; Glaxo Wellcome, Orion Diagnostica, Trosa, Sweden) at 1 μg/ml was used as a positive control, and complete medium was used to correct for spontaneous proliferation. The mean radioactivity (counts per minute) was calculated for all triplicates of antigens, mitogens, and the medium control. To obtain a value for the specific proliferation, a stimulation index (SI) was calculated by dividing the mean counts per minute for each antigen or mitogen by the mean counts per minute for medium or control baculovirus. An SI above 3 was defined as a specific response.

Individuals infected with different subtypes of HIV-1 had a very low or no T-cell response to HIV antigens before immunization (Fig. 1). After immunization with rgp160 of subtype B, all HIV-infected individuals improved their capacity to respond immunologically to rgp160 of subtype B. This occurred in patients infected with subtypes C, D, E, F, and G in addition to B (Fig. 1). The responses were maintained at high levels during the whole period studied (Fig. 1b). Nonimmunized patients or individuals receiving placebo rarely responded. Reactivity to non-gp160 components of the immunogen was low (0 to 15% of the HIV-specific reactivity [not shown]). The increases in T-cell proliferative responses were HIV specific as determined by reactivity to rgp160 and/or gp120, but responses did not increase or they increased to a minor degree (<15%) to recall or control antigens (9).

FIG. 1.

(a) Magnitudes of specific T-cell responses to rgp160 before and after immunization at all time points (788 assays in total) for 36 patients infected with HIV-1 subtypes B to G (the median and 25th and 75th percentiles, as well as 10th and 90th percentiles, are shown). Single values are shown prevaccination for subtype C and postvaccination for subtype D. (b) Specific T-cell responses over time to rgp160 of subtype B in patients infected with subtype B, C, D, E, F, or G. The patients were immunized either with rgp160 of subtype B or placebo (alum). All time points were not evaluated for all patients, but there is at least one pre- and one postvaccination assay for each subtype. Mean responses from both nonimmunized (n = 400) and noninfected individuals (n = 400) represent control values.

We also evaluated the T-cell reactivity to native gp120 derived from virions of isolates from subtypes A to E as well as a recombinant preparation of gp120. Nine patients who themselves were infected with HIV-1 of subtype B developed significant reactivities to the native gp120 antigens (Table 1). Reactivity to the B envelope antigen developed in all nine individuals after immunization with gp160 of subtype B. One patient infected with subtype C and immunized with subtype B reacted to the native gp120 of subtypes B, D, and E. Placebo recipients did not react with any of the envelope antigens (Table 1). Thus, 7 out of 10 tested immunized individuals responded to native gp120 from at least one other subtype in addition to subtype B, whereas a placebo recipient did not respond to any gp120 antigen (Table 1). The nine vaccinated individuals in Table 1 were evaluated at only one time point for these specific native antigens. A responder was defined as an individual responding with a SI above 3, as described above. Due to the fact that the study was blinded at the time the assay was performed, there were nine vaccinated individuals and only one placebo recipient. Specific responses to HIV antigens are very low or nonexistent before immunizations, as shown by placebo and preimmunization values. For the placebo recipient in this particular case the subtype is not known, but for the placebo patients that were investigated for rgp160 or rgp120, the subtypes were B to G.

TABLE 1.

HIV-specific T-cell responses to native preparations of gp120 derived from different genetic subtypes

Activating antigen	No. of responders/no. tested
	rgp160-immunized infected patient(s)		Placebo recipient(s)
	Subtype B	Subtype C
Native gp120
Subtype A	3/9	0/1	0/1
Subtype B	7/9	1/1	0/1
Subtype C	5/9	0/1	0/1
Subtype D	3/9	1/1	0/1
Subtype E	2/9	1/1	0/1
rgp120 of subtype B	8/9	1/1	0/1
rgp160 of subtype B	25/25	1/1	0/6

These data indicate that it is possible to induce cross-subtype HIV-specific T-cell proliferative responses in patients already infected with another HIV-1 subtype. Immunized individuals infected with any one of the subtypes B, C, D, E, F, and G responded to antigen of subtype B. Most patients also responded to at least one more subtype antigen in addition to B. Such responses were low or nonexistent before immunization.

For protection against a primary HIV infection, we envisage that both a primary local humoral and a cell-mediated immunity in mucosa are needed. Following the primary infection, a strong specific T-helper response and a systemic cytotoxic T-cell response appear to be associated with a better prognosis (18, 19). It is desirable to induce this second barrier, which should include cellular responses at all levels, such as T-helper cells, natural killer cells, antibody-dependent cellular cytotoxicity, and CTLs. However, it is only the CTL response that persists naturally for any demonstrable period.

HAART can potently and durably reduce HIV-1 replication in vivo, but there are now indications that even prolonged treatment will not result in total eradication of replication- competent virus in PBMCs. In order to be able to eradicate the virus, probably new therapeutic approaches must be considered in combination with HAART. These approaches may need to include both an activation of resting CD4 T cells and stimulation of the HIV-1-specific immunity (11). Therapeutic vaccination with rgp160 yields strong HIV-specific T-helper-cell responses and has a modest positive effect on CD4 counts but does not have any effect on viral load (9, 20). HAART alone does not seem to affect the HIV-specific T-helper-cell responses but reduces viral load. This suggests that a combination of HAART and immunization would improve the total effect. The finding that cross activation of the T-helper response can also occur in patients infected with subtypes other than that of the immunogen, gives hope that HAART can be combined with immunogens that do not have to represent all locally thriving strains or subtypes of HIV.