Sajadi, Mohammad M MD*; Guan, Yongjun PhD*; DeVico, Anthony L PhD*; Seaman, Michael S PhD†; Hossain, Mian PhD*; Lewis, George K PhD*; Redfield, Robert R MD*
The immune system requires the presence of sufficient quantities of antigen to elicit an immune response. Despite numerous studies in HIV-infected patients, the relationship between HIV-1 antigenic load and HIV-1 neutralization remains unclear. Several lines of evidence suggest that the antibody response to the HIV-1 envelope lacks durability.1,2 Because of this, we hypothesized that a persistent but low amount of circulating HIV-1 virus would be necessary to produce and maintain a robust humoral immune response. This study was undertaken to examine the relationship between HIV-1 antigenic load (as reflected by HIV-1 plasma RNA copies/mL) and broad HIV-1 neutralizing antibody activity.
Characterization of HIV-1 antigenic load and neutralizing activity was performed in 120 HIV-infected patients. These included HIV-1-infected patients with a wide range of viremia, who belonged to one of the following 4 cohorts: (1) HIV-1 natural viral suppressors (NVS) defined as individuals with HIV-1 infection by both Western Blot and proviral DNA, and at least a 2-year period with <400 HIV-1 RNA copies per milliliter in the absence of highly active antiretroviral therapy (HAART)3,4; (2) low viral load (LVL) cohort consists of individuals with 500-20,000 HIV-1 RNA copies per milliliter in the absence of HAART; (3) medium/high viral load (MHVL) cohort consists of individuals with >20,000 HIV-1 RNA copies per milliliter in the absence of HAART; and (4) HAART cohort which consists of individuals on their first HAART regimen with suppressed viral loads for at least 1 year. Demographic characteristics of these cohorts are given in Table 1. Forty-eight NVS, 36 LVL, 18 MHVL, and 18 HAART patients were tested. This study has Instituional Review Board approval, and all individuals provided informed consent.
HIV-1 Neutralization Testing
HIV-1 neutralization testing was performed using a luciferase-based assay in TZM.bl cells as previously described.5 This assay measures the reduction in luciferase expression after a single round of virus infection. For plasma, starting at a 1:20 dilution, 3-fold serial dilutions of serum were performed in duplicate. Two hundred TCID50 of pseudovirus was added to each well (with plasma or IgG) and incubated for 1 hour at 37°C. TZM.bl cells were then added (1 × 104 per well) in 10% D-MEM medium containing DEAE-Dextran (Sigma, St. Louis, MO) at a final concentration of 11 μg per milliliter. after 48 hours (37°C), 150 uL of medium was added to 100 uL of Bright-Glo luciferase reagent (Promega, Madison, WI), and luminescence measured using a Victor 3 luminometer (Perkin Elmer, Waltham, MA). Stocks of Env-pseudotyped viruses were prepared by transfection of 293T/17 cells.5 All patient plasma was tested against 3 Tier 1 Clade B pseudoviruses (SF162.LS, BaL.26, SS1196.1), 12 Tier 2 Clade B pseudoviruses (6535.3, QH0692.42, SC422661.8, PVO.4, TRO.11, AC10.0.29, RHPA4259.7, THRO4156.18, REJO4541.67, TRJO4551.58, WITO4160.33, CAAN5342.A2), and MuLV control.
The presence of broadly neutralizing antibodies in plasma was confirmed with purified IgG tested against the above Tier 1 and 2 viruses. IgG was purified by Protein A (GE Healthcare, Piscataway, NJ) and tested in neutralization assays starting at 333 ug per milliliter or greater, with 3-fold serial dilutions. For patients with both plasma and IgG broadly neutralizing antibodies, heterologous clade testing was performed with the following 18 virus panel, depending on sample availability: Clade C (Du156.12, Du172.17, Du422.1, ZM53M.PB12, ZM135M.PL10a, ZM197M.PB7, ZM214M.PL15), Clade A (Q23.17, Q259.d2.17, Q461.e2, Q168.a2, 3415.v1.c1, 0439.v5.c1, 0260.v5.c1, 3365.v2.c20), CRF02_AG (257-31, 263-8, 211-9), and MuLV control.
Statistical Analysis and Definitions
Inhibitory dose 80 titer (ID80) was defined as the plasma dilution causing 80% reduction in relative luminescence units compared with controls. Broadly neutralizing was defined as plasma with ID80 neutralization titers at least twice the MuLV control against 8 or more of the 12 Tier-2 viruses.6 Statistical analysis was performed with Graph Pad Prism (San Francisco, CA) for the Mann-Whitney and Fisher exact tests. Logistic regression was done with STATA statistical software (College Station, TX). The P values <0.05 were considered statistically significant.
Identification of and Characteristics of Individuals With Broadly Neutralizing Antibodies
Twelve of the 120 individuals demonstrated broadly neutralizing plasma, with the ability to neutralize at least 75% of the Tier-2 viruses. After testing of purified IgG against the 15 Tier-1 and Tier-2 viruses, 10 individuals had plasma neutralization profiles that closely matched their IgG neutralization profiles. Table 2 provides a summary of the plasma Tier 2 testing of the 10 patients with broadly neutralizing antibodies. The other 2 individuals had IgG that had no neutralizing activity against Tier-2 viruses, and in retrospect, these 2 individuals had a positive MuLV control during ID50 plasma testing (but not ID80). One of these 2 individuals also had a contemporaneous serum sample tested, which did not show broadly neutralizing activity against HIV-1 (or activity against MuLV). Heterologous virus testing of IgG was done in 6 of 10 individuals with broadly neutralizing antibodies. Results demonstrated broad activity against Clade A, Clade C, and CR0F_AG viruses in all individuals tested. Table 3 provides a summary of the heterologous clade testing.
Overall, 2 of 48 NVS (4.2%), 8 of 34 LVL (22.2%), 0 of 18 MHVL (0%), and 0 of 18 HAART (0%) individuals had broadly neutralizing antibodies. These 10 carried a diagnosis of HIV-1 for a median of 13 years (range 1-22). Their median CD4 count was 589 cells per microliter (range 202-927) and HIV-1 viral load of 1611 copies per milliliter (range 110-8964).
Six of these 10 individuals with broadly neutralizing plasma continue to maintain normal CD4 counts without therapy. However, HAART was initiated in 4 individuals (3 in the LVL group and 1 in the NVS group). One patient, 14 years after diagnosis, had an increase in HIV-1 RNA from a baseline of 5000-10,000 copies per milliliter to close to 30,000 copies per milliliter with a CD4 loss from 507 to 288 cells per microliter in 6 months. Two patients, although maintaining HIV-1 RNA <5000 copies per milliliter, had slow but progressive CD4 loss to less than 300 cells per microliter requiring HAART therapy. The final patient, although maintaining HIV-1 RNA <200 copies per milliliter over many years, likewise had a slow CD4 loss over and HAART therapy was initiated at a CD4 count of 489 (CD4% of 18) 19 years after diagnosis of HIV-1.
Correlation of the Presence of Broadly Neutralizing Antibodies With Viral Load
The presence of broadly neutralizing antibodies was analyzed according to cohort and HIV-1 viral load. Although all of the patients with broadly neutralizing antibodies were in the NVS and LVL groups, there was no correlation between presence of broadly neutralizing plasma and patient cohort (data not shown). However, there was a significant correlation between plasma viral load and the presence of broadly neutralizing antibodies in those with HIV-1 RNA between 102 and 104 copies per milliliter compared with those <102 and those with >104 HIV-1 RNA copies per milliliter (P = 0.0003 and 0.0245, respectively) with the 2-tailed Fisher exact test. In addition, individuals with HIV-1 RNA between 102 and 104 copies per milliliter had a higher number of Tier-2 viruses neutralized compared with the <102 or >104 copies per milliliter groups. This reached significance when compared with the <102 group (P ≤ 0.0001), and there was a trend toward significance in the >104 group (P = 0.076) with the 2-tailed Mann-Whitney test. The correlation of broadly neutralizing antibodies with viral load, and the correlation between the total number of Tier-2 viruses neutralized for the 102-104 compared with the <102 group was significant whether or not the HAART-treated patients were included in the analyses (P = 0.0018 and P = 0.0002, respectively) (Fig. 1).
In addition, individuals with HIV-1 RNA between 102 and 104 copies per milliliter had a higher mean ID80 neutralization titer for each of the 15 viruses tested compared with the <102 or >104copies per milliliter groups. This reached significance for 46.7% of the individual viruses tested when analyzed by single patient neutralization using the 2-tailed Mann-Whitney test. The above is seen in Table 4.
Using logistic regression, there was no relation between presence of broadly neutralizing antibody and age, HIV risk factor, duration of HIV infection, HCV status, or CD4 count (data not shown). There was, however, an association with male sex (P = 0.016), as 10 of 11 persons with broadly neutralizing plasma were male.
The data suggest that in HIV-1-infected individuals, the ability to produce broadly neutralizing antibodies correlates with plasma viremia. The LVL group, which was defined as having viral loads between 500 and 20,000 copies per milliliter, had the most individuals with HIV-1 broadly neutralizing plasma. In addition, there were 2 individuals in the NVS cohort who had broadly neutralizing activity. Importantly, both of these NVS individuals had low-level viremia detectable by routine HIV-1 RNA testing, as opposed to the majority of NVS who consistently have undetectable viremia and do not have broadly neutralizing antibodies. Because of this observation, the data were analyzed according to HIV-1 RNA copies, and the presence of broadly neutralizing antibodies correlated with a viral load of 102 and 104 copies per milliliter. We did not observe any persons with broadly neutralizing antibodies who had <102 or >104 HIV-1 RNA copies per milliliter. This suggests that an optimum amount of antigen production is needed for this response, which is less likely to be seen in patients with higher or lower HIV-1 RNA copies.
Patients with <102 HIV-1 RNA copies per milliliter, whether due to natural control or HAART, demonstrated diminished ID80 neutralization values, an observation made in prior studies.2,7 Likewise, there were no individuals with broadly neutralizing activity in this group. Overall, the low neutralization titers in the NVS and those on HAART suggests that a decrease in antigen load led to a decrease in antibody titer, a situation that has been described in patients after starting HAART.2 The lack of continual high titers in these situations may mirror the poor long-term immunogenicity of gp120 seen in the VaxGen vaccine trial.1 Importantly, a low-plasma neutralization titer in an NVS patient does not necessarily mean that they never had a robust humoral response to HIV. As we have shown in the NVS, a patient can harbor a large number of memory B cells that can be stimulated to produce epitope-specific HIV antibodies that cannot be found in circulation.8
Several previous studies in LTNPs noted increasing neutralization breadth and titers to heterologous viruses over time.9-11 Importantly, not all studies have shown this.12,13 LTNPs, which were defined by CD4 count only, have wide ranging viral loads, and the combined LVL and NVS groups in this study can be viewed as constituting a cohort similar to LTNPs (though not all would meet the CD4 or time requirements). More recent studies have focused on cohorts of patients, such as ours, that have a definition of HIV suppression based on viral load and not CD4 count.2,7,14,15 In this study, 12% of the combined NVS/LVL groups had broadly neutralizing plasma, which is comparable to the 12% reported for a combined cohort of elite and viremic controllers (similar to our definition of NVS and LVL).14 The disparate results in this study between those NVS who have greater or less than 102 HIV-1 RNA copies per milliliter may account for the discrepancies in prior LTNP studies.
At the high levels of plasma viremia, a diminished humoral response was observed. The ID80 titers in those with >104 HIV-1 RNA copies per milliliter for the Tier 1 and Tier 2 viruses were lower than those with 102-104 HIV-1 RNA copies per milliliter but comparable with those with those with <102 HIV-1 RNA copies per milliliter. Also, there were no persons in this group who demonstrated broadly neutralizing plasma. These data are consistent with the observation of low antibody titers and neutralization breadth seen in the later stages of HIV,10,11,17 which could be due to progressive B cell dysfunction caused by HIV-1.
Recently, some groups have noted increasing neutralization breadth with higher viral loads.16,18-20 These studies have had fewer patients than this study, and most of them have tested fewer viruses than ours. However, it is likely that some of these patients do develop broadly neutralizing antibodies, which are lost as HIV disease progresses. It should be noted that 1 of the 10 patients with broadly neutralizing antibodies in this study (who also eventually received HAART) experienced an increase in HIV-1 viral load to close to 30,000 copies per milliliter. In this individual, we retested the plasma from the time point just before starting HAART and broadly neutralizing antibody response with a similar neutralization pattern to what was seen from the time point 18 months prior. Thus, the development of broadly neutralizing antibodies in this individual was related to the 13 years his immune system was exposed to HIV at lower viral loads (rather than the six months with higher viral loads), and that without HAART, these antibodies could have disappeared with further HIV-mediated immune dysfunction and destruction.
Besides the role of the optimum amount of antigen associated with broadly neutralizing plasma highlighted in this study, duration of infection has also been noted to be important.9,10,16 In this study, the patients with broadly neutralizing antibodies also had been infected with HIV for a relatively long time (a median of 11 years since HIV diagnosis), but we were unable to show an association with time since diagnosis. Furthermore, this cannot be considered the sole determinant, as the NVS and LVL had comparable length of HIV infection, but a difference in HIV-1 RNA copies and ability to neutralize various strains of HIV-1. We were unable to find other correlations with broadly neutralizing plasma other than the viral load and male sex. Regarding the later, 90% (9 of 10) of the patients with broadly neutralizing plasma were male; however, the significance of this association remains unknown.
The study's strengths include the total number of patients tested (120 patients), the high number of Tier 1 and Tier 2 viruses tested (at least 15 viruses per patient). In addition, although we used a definition of broadly neutralizing that has been used previously,6 these were verified with both IgG and heterologous clade testing. Finally, this study was focused on individuals with LVLs (82 NVS/LVL), where the majority of patients with broad neutralization were found. However, our findings need to be considered within the inherent limitations of a cross-sectional study. In terms of generalizability, it is important to note that the study subjects are predominantly African American (>95%), Clade B infected, and have relatively high rates of IDU, thus the findings need to be confirmed in other cohorts of patients. In addition, although we have demonstrated a correlation between viral load and broadly neutralizing antibodies, this study was not designed to find causality or a putative mechanism (such as low antigen stimulation leading to low neutralization levels or high viral loads causing B-cell dysfunction). Finally, correlations about the potential protective effects of neutralizing antibodies cannot be done in this type of study. More prospective studies starting in acute HIV infection, such as the one by Piantadosi et al,18 should be able to answer this question.
In conclusion, these results indicate that low but persistent HIV antigen expression correlates with high levels of broad HIV-1 neutralizing antibody activity. The presence of broadly neutralizing activity was associated with 102 and 104 copies of HIV-1 RNA. At higher levels of plasma viremia, neutralization levels were diminished. Conversely, at lower levels, either due to antiretrovirals or natural suppression by other mechanisms, there seems to be insufficient antigen stimulation to maintain high neutralization levels. These findings may have important implications in furthering the understanding of the humoral response to HIV infection.
We would like to thank members of the NVS Study and Becky Boyce, RN, the study coordinator.
1. Pitisuttithum P, Berman PW, Phonrat B, et al. Phase I/II study of a candidate vaccine designed against the B and E subtypes of HIV-1. J Acquir Immune Defic Syndr
. 2004;37:1160-1165. Erratum in: J Acquir Immune Defic Syndr
2. Bailey JR, Lassen KG, Yang HC, et al. Neutralizing antibodies do not mediate suppression of human immunodeficiency virus type 1 in elite suppressors or selection of plasma virus variants in patients on highly active antiretroviral therapy. J Virol
3. Sajadi MM, Heredia A, Le N, et al. HIV-1 natural viral suppressors: control of viral replication in the absence of therapy. AIDS
4. Sajadi MM, Constantine NT, Mann DL, et al. Epidemiologic characteristics and natural history of HIV-1 natural viral suppressors. J Acquir Immune Defic Syndr
. 2009;50:403-408. PMID: 19214118.
5. Li M, Gao F, Mascola JR, et al. Human immunodeficiency virus type 1 env clones from acute and early subtype B infections for standardized assessments of vaccine-elicited neutralizing antibodies. J Virol
6. Scheid JF, Mouquet H, Feldhahn N, et al. Broad diversity of neutralizing antibodies isolated from memory b cells in HIV-infected individuals. Nature
7. Pereyra F, Addo MM, Kaufmann DE, et al. Genetic and immunologic heterogeneity among persons who control HIV infection in the absence of therapy. J Infect Dis
8. Guan Y, Sajadi MM, Kamin-Lewis R, et al. Humoral immunity against conserved epitopes of the HIV-1 envelope protein archived in memory B cells in natural viral suppressors: discordance with plasma antibodies. Proc Natl Acad Sci U S A
. 2009;106:3952-3957. PMID: 19225108.
9. Pilgrim AK, Pantaleo G, Cohen OJ, et al. Neutralizing antibody responses to human immunodeficiency virus type 1 in primary infection and long-term-nonprogressive infection. J Infect Dis
10. Cecilia D, Kleeberger C, Muñoz A, et al. A longitudinal study of neutralizing antibodies and disease progression in HIV-1-infected subjects. J Infect Dis
11. Zhang YJ, Fracasso C, Fiore JR, et al. Augmented serum neutralizing activity against primary human immunodeficiency virus type 1 (HIV-1) isolates in two groups of HIV-1-infected long-term nonprogressors. J Infect Dis
12. Barker E, Mackewicz CE, Reyes-Terán G, et al. Virological and immunological features of long-term human immunodeficiency virus-infected individuals who have remained asymptomatic compared with those who have progressed to acquired immunodeficiency syndrome. Blood
13. Harrer T, Harrer E, Kalams SA, et al. Strong cytotoxic T cell and weak neutralizing antibody responses in a subset of persons with stable nonprogressing HIV type 1 infection. AIDS Res Hum Retroviruses
14. Scheid JF, Mouquet H, Feldhahn N, et al. Broad diversity of neutralizing antibodies isolated from memory b cells in HIV-infected individuals. Nature
15. Doria-Rose NA, Klein RM, Manion MM, et al. Frequency and phenotype of human immunodeficiency virus envelope-specific B cells from patients with broadly cross-neutralizing antibodies. J Virol
. 2009;83:188-199. E-pub October 15, 2008.
16. Sather DN, Armann J, Ching LK, et al. Factors associated with the development of cross-reactive neutralizing antibodies during human immunodeficiency virus type 1 infection. J Virol
. 2009;83:757-769. E-pub November 5, 2008. Erratum in: J Virol
. 2009;83:4713-4715. PMID: 18987148.
17. Wong MT, Warren RQ, Anderson SA, et al. Longitudinal analysis of the humoral immune response to human immunodeficiency virus type 1 (HIV-1) gp160 epitopes in rapidly progressing and nonprogressing HIV-1-infected subjects. J Infect Dis
18. Piantadosi A, Panteleeff D, Blish CA, et al. Breadth of neutralizing antibody response to human immunodeficiency virus type 1 is affected by factors early in infection but does not influence disease progression. J Virol
. 2009;83:10269-10274. E-pub July 29, 2009. PMID: 19640996.
19. Deeks SG, Schweighardt B, Wrin T, et al. Neutralizing antibody responses against autologous and heterologous viruses in acute versus chronic human immunodeficiency virus (HIV) infection: evidence for a constraint on the ability of HIV to completely evade neutralizing antibody responses. J Virol
20. Rodriguez SK, Sarr AD, MacNeil A, et al. Comparison of heterologous neutralizing antibody responses of human immunodeficiency virus type 1 (HIV-1)- and HIV-2-infected Senegalese patients: distinct patterns of breadth and magnitude distinguish HIV-1 and HIV-2 infections. J Virol
. 2007;81:5331-5338. E-pub February 14, 2007. PMID: 17301136.
© 2011 Lippincott Williams & Wilkins, Inc.