Effect of B-cell depletion on coreceptor switching in R5 simian-human immunodeficiency virus infection of rhesus macaques

Abstract

We recently described a coreceptor switch in rapid progressor (RP) R5 simian-human immunodeficiency virus SF162P3N (SHIV(SF162P3N))-infected rhesus macaques that had high virus replication and undetectable or weak and transient antiviral antibody response (S. H. Ho et al., J. Virol. 81:8621-8633, 2007; S. H. Ho, N. Trunova, A. Gettie, J. Blanchard, and C. Cheng-Mayer, J. Virol. 82:5653-5656, 2008; and W. Ren et al., J. Virol. 84:340-351, 2010). The lack of antibody selective pressure, together with the observation that the emerging X4 variants were neutralization sensitive, suggested that the absence or weakening of the virus-specific humoral immune response could be an environmental factor fostering coreceptor switching in vivo. To test this possibility, we treated four macaques with 50 mg/kg of body weight of the anti-CD20 antibody rituximab every 2 to 3 weeks starting from the week prior to intravenous infection with SHIV(SF162P3N) for a total of six infusions. Rituximab treatment successfully depleted peripheral and lymphoid CD20(+) cells for up to 25 weeks according to flow cytometry and immunohistochemical staining, with partial to full recovery in two of the four treated monkeys thereafter. Three of the four treated macaques failed to mount a detectable anti-SHIV antibody response, while the response was delayed in the remaining animal. The three seronegative macaques progressed to disease, but in none of them could the presence of X4 variants be demonstrated by V3 sequence and tropism analyses. Furthermore, viruses did not evolve early in these diseased macaques to be more soluble CD4 sensitive. These results demonstrate that the absence or diminution of humoral immune responses by itself is insufficient to drive the R5-to-X4 switch and the neutralization susceptibility of the evolving viruses.

FIG. 1.

Effect of rituximab treatment on CD20⁺ B cells in R5 SHIV_SF162P3N-infected macaques. (A) B-cell concentration in peripheral blood and inguinal lymph node (Ing LN) over time. An arrow indicates the time of anti-CD20 MAb infusions, and a dagger indicates the time of death due to AIDS. (B) Immunohistochemistry for CD20 and CD35 in mesenteric lymph nodes. Sections of LN samples collected at the chronic stage (12 wpi; biopsy specimen), at the terminal disease stage for macaques EK23 (18 wpi), DV60 (32 wpi), and DJ11 (43 wpi), and at the end of the study period for EN03 (68 wpi) were stained for CD20- and CD35-expressing cells as described in the text, with the staining of corresponding sections from an R5 SHIV_SF162P3N-infected RP (DG08) in parallel for comparison.

FIG. 2.

Virologic and immunologic measurements in B-cell-depleted macaques. (A) Viral load and absolute CD4⁺ T-cell counts. Arrows indicate the time of rituximab infusions, and a dagger indicates the time of euthanasia due to AIDS. (B) Percentages of CD4⁺ T cells in the iliac (Ili), mesenteric (Mes), and colonic (Col) lymph nodes and lamina propria lymphocyte (LPL) from the jejunum during primary infection (3 to 6 wpi) and at the time of necropsy. Baseline values generated from two or three uninfected macaques are shown for reference. N.A., not available.

FIG. 3.

Coreceptor usage of viruses recovered from PBMCs of EK23, DV60, and DJ11 at end-stage disease. CXCR4 and CCR5 usage was determined by blocking entry into TZM-bl cells with 1 μM CXCR4-(AMD3100)- or CCR5-(TAK779)-specific inhibitor (A) and the infection of U87.CD4.CCR5 and U87.CD4.CXCR4 cells (B). For TZM-bl cells, error bars indicate standard errors of data in triplicate wells, with the inoculating SHIV_SF162P3N virus (P3N) serving as a control for CCR5 usage. For the infection of indicator cell lines, values represent p27 antigen production at day 7 postinfection. Results shown are representative of at least two independent experiments.

FIG. 4.

Comparison of V3 loop sequence of viruses in the inoculum SHIV_SF162P3N and in PBMC and lymph nodes of EK23, DV60, and DJ11 at the time of death. Gaps are indicated as dots, dashes denote similarity in sequence, and the net-positive charge of this region was calculated and shown. Positions 11 and 25 within the V3 loop are designated, and the N-terminal glycan site is boxed. The numbers in parentheses represent the number of clones with the indicated V3 loop sequence/total number sequenced. PBMC, peripheral blood mononuclear cells; Ax, axillary lymph node; Col, colonic lymph node; Mes, mesenteric lymph node.

FIG. 5.

sCD4 sensitivity of acute, chronic, and end-stage disease viruses in B-cell-depleted macaques. The susceptibility of replication-competent viruses recovered from EK23, DV60, and DJ11 at 2 (acute) and 12 wpi (chronic) and the time of necropsy to sCD4 (CD4-IgG2) were determined and compared to that of the SHIV_SF162P3N inoculum (P3N). Numbers above bars indicate the fold increase in sCD4 sensitivity compared to the level of the acute virus. Data are the means and standard deviations from at least two independent neutralization experiments.

FIG. 6.

Relationship between plasma RNA levels, survival, and antiviral antibody responses in SHIV_SF162P3N-infected macaques. Comparison of virus replication (A), plasma vRNA levels (AUCs) through 28 wpi (B), and survival rate (C) between R5 SHIV_SF162P3N-infected macaques that are rapid progressors (RPs; n = 5), chronic progressors (CP; n = 6), and B-cell-depleted animals (n = 4). The mean plasma vRNA levels through 28 wpi (the longest time of survival for the RPs) were transformed into AUCs and compared using Wilcox rank sum analysis, and the survival rate comparison was based on Kaplan-Meier analysis.