Comprehensive analysis of unique cases with extraordinary control over HIV replication

Abstract

True long-term nonprogressors (LTNPs)/elite controllers (ECs) maintain durable control over HIV replication without antiretroviral therapy. Herein we describe 4 unique persons who were distinct from conventional LTNPs/ECs in that they had extraordinarily low HIV burdens and comparatively weak immune responses. As a group, typical LTNPs/ECs have unequivocally reactive HIV-1 Western blots, viral loads below the lower threshold of clinical assays, low levels of persistent viral reservoirs, an over-representation of protective HLA alleles, and robust HIV-specific CD8(+) T-cell responses. The 4 unique cases were distinguished from typical LTNPs/ECs based on weakly reactive Western blots, undetectable plasma viremia by a single copy assay, extremely low to undetectable HIV DNA levels, and difficult to isolate replication-competent virus. All 4 had at least one protective HLA allele and CD8(+) T-cell responses that were disproportionately high for the low antigen levels but comparatively lower than those of typical LTNPs/ECs. These unique persons exhibit extraordinary suppression over HIV replication, therefore, higher-level control than has been demonstrated in previous studies of LTNPs/ECs. Additional insight into the full spectrum of immune-mediated suppression over HIV replication may enhance our understanding of the associated mechanisms, which should inform the design of efficacious HIV vaccines and immunotherapies.

Figure 1

HIV-1 Western blots of the cases were weakly reactive in contrast to the unequivocally positive Western blots of typical LTNPs/ECs. (A) HIV-1 Western blots with controls performed with serial, including standard (1:100; boxes), dilutions of patient sera from each of the cases C1 to C4 to diluent. (B) HIV-1 Western blots with controls performed with serial, including standard (1:100; box), dilutions of patient serum to diluent from one representative LTNP/EC (left), and at the standard dilution for a subset of typical LTNPs/ECs (right).

Figure 2

Replication-competent HIV could be recovered after 120-day culture from the activated GALT-derived CD4⁺ T cells of a typical LTNP/EC, but not from those of the cases. (A-B) Positively selected CD4⁺ T cells from PBMCs (A; n = 7) or gut biopsy (B; n = 6) samples of the cases (C1, ×; C2, †; C3, ★, C4,◊) and controls were stimulated with medium containing soluble anti-CD3/anti-CD28 monoclonal antibodies and recombinant IL-2 (40 IU/mL) for 3 days and subsequently cultured in IL-2-containing medium for 120 days. Analysis of intracellular HIV p24 expression and overgrowth of CD8⁺ T cells occurred every 48 to 72 hours. CD8 depletion was performed if CD8⁺ cells represented more than 1% of the T cells in culture.

Figure 3

Proviral nef sequence analysis of the cases. (A) nef sequences obtained from C1 to C3 are aligned with the clade B consensus sequence. Dots indicate identity with the consensus sequence; dashes, gaps; and asterisks, stop codons. Sequence variations in C1 and C2 occurred in regions of common length polymorphisms. Premature stop codons occurred in all 5 sequences obtained from C3 and were not due largely to G to A hypermutation. (B) A phylogenetic tree of nef sequences from 3 cases (C1, ×; C2, †; C3, ★) and other chronically infected patients from another study (M.S., personal communication), including conventional LTNPs/ECs (black circles), viremic progressors (solid black upward turning arrow heads), and Rx<50 (solid black downward turning arrow heads) was constructed for each patient by classic, maximum-likelihood and Bayesian methods. Other clade subtype nef sequences are included for comparison.

Figure 4

CD4⁺ T cells of the cases were fully receptive to HIV super-infection, excluding passive intrinsic resistance to HIV. (A) Purified CD4⁺ T cells derived from the cases (C1, ×; C2, †; C3, ★; C4, ◊), HIV-negative controls (n = 10), typical LTNPs/ECs (n = 11), viremic progressors (n = 10), and Rx<50 (n = 17) were stimulated with medium containing soluble anti-CD3/anti-CD28 monoclonal antibodies and recombinant IL-2 (40 IU/mL) for 3 days, coincubated with magnetized HIV_SF162 in the presence of a super-magnet for 15 minutes, and cultured for 36 hours before analysis for intracellular HIV p24 expression. HIV infection is expressed as the percentage of CD3⁺ lymphocytes expressing p24. Horizontal lines represent median values. Comparisons were made using the Wilcoxon 2-sample test. None of the comparisons was significant. (B) CD4⁺ T cells from a subset of 3 cases (C1, ×; C2, †; C4, ◊), 3 randomly selected LTNPs/ECs, and 3 HIV-negative controls were negatively selected with magnetic beads and subjected to redepletion of CD8⁺ cells. Purity was confirmed to exceed 98% by flow cytometry. After 4-hour incubation with HIV_SF162 at a TCID₅₀ of 2000, cells were washed twice, resuspended in 10% human AB medium not containing human IL-2 at a concentration of 5 × 10⁵ cells/mL, and incubated in 24-well plates at 37°C for 4 days before assessment by flow cytometry for HIV Gag p24 expression and CD8⁺ T-cell overgrowth. HIV infection is expressed as the percentage of CD3⁺ lymphocytes expressing p24.

Figure 5

HIV-specific antibody response profiles clearly distinguished cases from other chronically infected patients, including typical LTNPs/ECs. (A-E) Antibody response profiles specific for the HIV-1 gene products gp41 (A), p24 (B), matrix (MA) protein (C), reverse transcriptase (RT; D), and integrase (E), as determined by the LIPS assay and reported in log₁₀ luminometer units (LU), are shown for the cases (C1, ×; C2, †; C3, ★; and C4, ◊), HIV-negative controls (n = 7), typical LTNPs/ECs (n = 12), and viremic progressors (Prog; n = 11). Horizontal lines represent median values. Comparisons were made using the Wilcoxon 2-sample test. Only P values referring to comparisons between the responses of the cases and other groups are shown.

Figure 6

Low frequencies of HIV-specific CD8⁺ T cells in the cases exhibited significant proliferative and cytotoxic responses after 6-day stimulation that were less than those of typical LTNPs/ECs. (A) Frequencies of IFN-γ–secreting CD8⁺ T cells measured in response to peptide pools spanning the HIV-1 Nef, Gag, and Pol gene products are shown for the cases (C1, ×; C2, †; C3, ★; C4, ◊), typical LTNPs/ECs (n = 21), viremic progressors (Prog; n = 12), and Rx<50 (n = 14). (B) Frequencies of IFN-γ–secreting CD8⁺ T cells measured in response to autologous HIV_SF162-infected CD4⁺ T-cell targets are shown for the 4 cases, typical LTNPs/ECs (n = 27), viremic progressors (Prog; n = 13), and Rx<50 (n = 50). (C) Proliferation as assessed by CFSE dilution over 6 days in response to autologous HIV_SF162-infected CD4⁺ T-cell targets is shown for the 4 cases, HIV-negative controls (n = 11), typical LTNPs/ECs (n = 24), viremic progressors (Prog; n = 11), and Rx<50 (n = 55). (D-E) Total cytotoxic responses measured by GrB target cell activity (D) or ICE (E) with day 6 CD8⁺ T cells are shown for the 4 cases, HIV-negative controls (n = 11), typical LTNPs/ECs (n = 18), viremic progressors (Prog; n = 17), and Rx<50 (n = 17). Horizontal lines represent median values. Comparisons were made using the Wilcoxon 2-sample test. Only P values referring to comparisons between the responses of the cases and other groups are shown. (F) Using day 6 CD8⁺ T cells, GrB target cell activity correlates directly with ICE (n = 67). Statistical analysis was performed using the Spearman correlation.