Characteristics of the earliest cross-neutralizing antibody response to HIV-1

Abstract

Recent cross-sectional analyses of HIV-1+ plasmas have indicated that broadly cross-reactive neutralizing antibody responses are developed by 10%-30% of HIV-1+ subjects. The timing of the initial development of such anti-viral responses is unknown. It is also unknown whether the emergence of these responses coincides with the appearance of antibody specificities to a single or multiple regions of the viral envelope glycoprotein (Env). Here we analyzed the cross-neutralizing antibody responses in longitudinal plasmas collected soon after and up to seven years after HIV-1 infection. We find that anti-HIV-1 cross-neutralizing antibody responses first become evident on average at 2.5 years and, in rare cases, as early as 1 year following infection. If cross-neutralizing antibody responses do not develop during the first 2-3 years of infection, they most likely will not do so subsequently. Our results indicate a potential link between the development of cross-neutralizing antibody responses and specific activation markers on T cells, and with plasma viremia levels. The earliest cross-neutralizing antibody response targets a limited number of Env regions, primarily the CD4-binding site and epitopes that are not present on monomeric Env, but on the virion-associated trimeric Env form. In contrast, the neutralizing activities of plasmas from subjects that did not develop cross-neutralizing antibody responses target epitopes on monomeric gp120 other than the CD4-BS. Our study provides information that is not only relevant to better understanding the interaction of the human immune system with HIV but may guide the development of effective immunization protocols. Since antibodies to complex epitopes that are present on the virion-associated envelope spike appear to be key components of earliest cross-neutralizing activities of HIV-1+ plasmas, then emphasis should be made to elicit similar antibodies by vaccination.

Figure 1. Cross-neutralizing activities in plasmas from the Vanderbilt Cohort.

The cross-neutralizing activities of plasmas from the indicated subjects (PID) were evaluated against the indicated clade B, C and A viruses. The values are the plasma titers at which 50% neutralization (IC50) was recorded. For clarity this information is color-coded: (blue) IC50<1∶100; (orange) 1∶100≤IC50≥1∶250; (red) IC50>1∶250. With the exception of SF162.LS (tier 1 virus), all other viruses are tier 2 , , . (-): less than 50% neutralization was recorded; YPI: years post-infection; ‘breadth’: the percent of isolates neutralized by a plasma sample, out of the total number of isolates tested, irrespective of the potency of neutralization . Each experiment was performed at least two independent times.

Figure 2. Cross-neutralizing activities in plasmas from the MGH Acute HIV infection Cohort.

The cross-neutralizing activities of plasmas from the indicated subjects (PID) were evaluated against the indicated clade B, C and A viruses. The values are the plasma titers at which 50% neutralization (IC50) was recorded. For clarity this information is color-coded: (blue) IC50<1∶100; (orange) 1∶100≤IC50≥1∶250; (red) IC50>1∶250. With the exception of SF162.LS (tier 1 virus), all other viruses are tier 2 , , . (-): less than 50% neutralization was recorded; YPI: years post-infection; ‘breadth’: the percent of isolates neutralized by a plasma sample, out of the total number of isolates tested, irrespective of the potency of neutralization . Each experiment was performed at least two independent times.

Figure 3. Evolution of the potency and breadth of cross-neutralizing antibody responses.

The pie charts represent the evolution of breadth and potency of serum neutralizing activities in the MGH Acute HIV Infection Cohort. Subjects who developed cross-neutralizing activities within the first 2 years of infection, but who were not followed longitudinally post that period, were not included in our calculations in the subsequent years. Samples from 17 subjects were available during the 1^st year of infection, samples from 16 during the 2^nd year, 11 during the 3^rd, and 7 during the 4^th year.

Figure 4. Immune activation markers and cross-neutralizing antibody responses.

The frequencies of (A) CD8+, and (B) CD4+ T cells expressing the indicated markers in subjects who developed cross-neutralizing antibodies (at least 75% breadth) (AC049, AC053, AC128, AC131 and AC180) and those who did not (no breadth) (AC093, AC110, AC167, AC183, AC194, AC212) are shown. These frequencies were determined at the earliest time point when cross-neutralizing antibody responses were evident: for AC049 at 2.62 year post-infection (ypi), for AC053 at 3.29 ypi, for AC128 at 1.41 ypi, for AC131 at 1.52 ypi, and for AC180 at 2.19 ypi. Similar time points of infection were used for those subjects who did not develop cross-neutralizing antibody responses.

Figure 5. Contribution of anti-gp120 antibodies to the overall neutralizing activity of HIV+ plasmas.

(A) Log10 decrease in neutralizing activity caused by the elimination of anti-gp120 antibodies, and (B) Log10 decrease in neutralizing activity of plasmas in the presence of D368R. The values are the average from 2–3 independent experiments in most cases. Light blue: no effect or less than 0.5 Log10 decrease; Yellow: decrease between 0.5 and 0.9 Log10; Red: over 0.9 Log10 decrease. >1.00: indicates that the depletion of anti-gp120 antibodies from plasma resulted in complete loss of the neutralizing activity. (--) the experiment was not performed because that particular plasma did not neutralize that particular virus; nd: experiment was not performed; YPI: years post infection.

Figure 6. Neutralizing activities of anti-gp120 antibody-depleted plasmas.

(A and B) Neutralizing activities of 2 plasmas against the indicated primary isolates prior to and following the removal of anti-gp120 antibodies, are shown. (A) Plasma AC053 against the TRO.11, JRFL, and YU2 viruses. (B) Plasma AC131 against TRO.11, JRFL, and Zm214M viruses. TRO.11 (red squares), JRFL (blue triangles), YU2 and ZM214M (green circles). Patient ID, breadth and years post infection are shown. Undepl: Undepleted plasma. Depleted: plasma depleted from anti-gp120 antibodies. Closed symbols and solid lines – undepleted plasmas; open symbols and dashed lines – gp120-depleted plasmas. (C) Plasmas from subjects AC098 (squares), AC115 (inverted triangles), and AC212 (circles) did not display significant breadth and neutralized only SF162.LS (Figure 2). Their anti-SF162.LS neutralizing activities were determined before and following depletion of the anti-gp120 antibodies. Closed symbols and solid lines: prior to depletion of anti-gp120 antibodies; open symbols and dashed lines: following the depletion of anti-gp120 antibodies. Each experiment was performed at least three independent times.

Figure 7. Contribution of anti-CD4-BS antibodies in the overall neutralizing activities of plasmas collected longitudinally.

The values indicate the Log10 decrease in neutralizing activity in the presence of the D368R construct. The values are the average from 2–3 independent experiments in most cases. The color-coding is the same as in Figure 5. (--) the experiment was not performed because that particular plasma did not neutralize that particular virus; YPI: years post infection. Cross-neutralizing antibodies that bind the CD4-BS do not recognize the D368R mutant , thus, the majority of the neutralizing activities that remain in the plasmas that have been incubated with D368R most likely are due to cross-neutralizing antibodies that bind the CD4-BS. There are some exceptions, which are discussed in the Results section.