Simian-Human Immunodeficiency Virus SHIV.CH505-Infected Infant and Adult Rhesus Macaques Exhibit Similar Env-Specific Antibody Kinetics, despite Distinct T-Follicular Helper and Germinal Center B Cell Landscapes

Infant and adult RMs exhibit similar plasma viral load kinetics following SHIV.C.CH505 infection.

Twelve female, adult rhesus macaques (age range, 4 to 10 years) (Table 1) were infected intravenously with SHIV.C.CH505 at a dose of 3.4 × 10⁵ 50% tissue culture infective doses (TCID₅₀). All 12 adult monkeys became infected after the first inoculation. To mimic breast milk transmission, six infant rhesus macaques were infected orally starting at 4 weeks of age (Table 1) with a repeated-exposure regimen that has been successfully developed with simian immunodeficiency virus SIV_mac251 to simulate exposure to HIV by breastfeeding (24, 25). Once monkeys were infected, viremia peaked at 2 weeks in both age groups, and peak viremias were of similar magnitudes between the groups (infant range, 6.7 × 10⁵ to 3.2 × 10⁷ vRNA copies/ml of plasma; adult range, 3 × 10⁵ to 1.3 × 10⁷ vRNA copies/ml of plasma) (Fig. 1A). The range in peak viremia among both age groups was similar to those observed in acute SIV (25, 26) and HIV (27) infections. At 12 weeks postinfection (wpi), 11 of the 12 adult monkeys and 4 of the 6 infant monkeys maintained plasma viral loads above the limit of detection of the assay (15 vRNA copies/ml of plasma) (Fig. 1A). Both infant and adult monkeys exhibited a transient decrease in frequencies of CD4⁺ T cells at 3 wpi (Fig. 1B); however, these frequencies were maintained through 12 wpi. Infant monkeys had higher absolute CD4 T cell counts overall than did adult monkeys, with ranges similar to those reported for uninfected cohorts (24, 28). Changes in CD4⁺ T cell absolute counts were not observed over the course of SHIV infection in either age group (Fig. 1C); thus, the SHIV.C.CH505 virus exhibited a more attenuated phenotype in infant monkeys, in contrast to observations in the SIV_mac251 model (29–31). Since the same viremic pattern was observed in both age groups independent of the route of infection, this model provided an opportunity to define differences in the infant and adult immune responses to SHIV infection.

The kinetics, magnitude, and specificity of HIV Env-specific plasma IgG responses are similar in infant and adult RMs during acute SHIV.C.CH505 infection.

We measured HIV Env-specific antibody responses to evaluate whether the ontogeny of these responses would differ between the infant and adult monkeys during the acute phase of infection. In both groups, CH505 gp120-specific responses were detectable by 3 wpi, with no statistically significant difference in magnitude (median gp120 IgG levels in infants and adults of 1,297 ng/ml and 1,785 ng/ml, respectively [P = 0.301]) (Fig. 2A). These responses continued to increase through 12 wpi, with no significant difference in magnitude (median gp120 IgG levels in infants and adults of 1.3 × 10⁵ ng/ml and 2.3 × 10⁵ ng/ml, respectively [P = 0.494]) or kinetics between infants and adults (Fig. 2A). Overall, the kinetics and magnitude of the gp41-specific plasma IgG response were also not significantly different between both age groups (Fig. 2B). However, at 12 wpi, the infant monkeys exhibited a trend toward lower gp41-specific IgG responses than those of adults, although this difference was not significant after correction for multiple comparisons (median gp41 IgG levels in infants and adults at 12 wpi of 3,712 ng/ml and 23,305 ng/ml, respectively [P = 0.041; false discovery rate-adjusted P [FDR P] = 0.458]) (Fig. 2B).

To determine the specificity of the plasma Env-specific IgG responses between age groups, we used a binding antibody multiplex assay (BAMA) to assess binding to various HIV Env linear and conformational epitopes at week 4 (Fig. 3A) and week 12 (Fig. 3B) postinfection (Table 2). For both groups, anti-V3 and -C5 binding responses were dominant and increased from week 4 to week 12 (Fig. 3A and B). Although infant monkeys exhibited a slightly higher antibody specificity for the CD4 binding site at 12 wpi, these differences were not statistically significant (P = 0.335) (Fig. 3C). While anti-V2 responses were rarely detected by BAMAs (Fig. 3A and B), linear peptide microarray analysis demonstrated that these responses were primarily CH505 specific (Fig. 3D). We also used a BAMA to assess cross-clade gp120 and gp140 breadth at 4 and 12 wpi (Fig. 4A and B). At 12 wpi, antibodies from infants and adults recognized all nine gp120 and gp140 antigens tested, demonstrating breadth acquisition (Fig. 4B). Overall, the median gp120- and gp140-specific IgG responses across all clades trended higher at week 12 in adult monkeys, consistent with a higher gp41 plasma IgG binding response (Fig. 4B).

Functional T cell responses in infant and adult monkeys during the acute phase of SHIV.C.CH505 infection.

We assessed functional T cell responses by measuring the frequency of interferon gamma (IFN-γ)-, interleukin-2 (IL-2)-, IL-17-, or tumor necrosis factor alpha (TNF-α)-producing CD4⁺ and CD8⁺ T cells. Due to limited sample availability, only peripheral blood mononuclear cells (PBMCs) collected from infant and adult monkeys at 8 wpi were analyzed (Fig. 5). Overall, the majority of CD4⁺ and CD8⁺ T cells from infants had little to no cytokine production compared to those from adults (Fig. 5A and B), similar to what has been observed in SIV and HIV infections (26, 32). When assessing the mean proportions of cytokine-producing T cells between both age groups, adult CD4⁺ T cells predominantly produced IL-2, IL-17, or TNF-α, while infant CD4⁺ T cells produced either IFN-γ, IL-2, or IL-17 (Fig. 5C). In contrast, CD8⁺ T cells from both groups predominantly produced either IL-17 or TNF-α (Fig. 5D).

SHIV.C.CH505-infected infant monkeys have high proportions of T-follicular helper cells in the lymph node compared to those of adults.

Frequencies of T-follicular helper (Tfh) cells have been reported to be increased in HIV-infected children compared to adults (33). In order to investigate Tfh cell responses during the acute phase of SHIV infection, we evaluated the proportions of CH505-specific and total CD4⁺ CXCR5^hi PD1^hi (Fig. 6A to D) Tfh cells in the lymph node of infant and adult monkeys at 12 wpi. The frequencies of total CD4⁺ T cells were similar between the two age groups, and the proportion of CH505-specific Tfh cells was not significantly different between the two groups (P = 0.592), yet infants had significantly higher proportions of total CD4⁺ CXCR5^hi PD1^hi Tfh cells (P = 0.024; FDR P = 0.036) (Fig. 6A to D). Additionally, we compared the frequencies of the Tfh subsets based on the surface expression of CXCR3 and CCR6, as follows: Tfh1 (CXCR3⁺ CCR6⁻), Tfh2 (CXCR3⁻ CCR6⁻), and Tfh17 (CXCR3⁻ CCR6⁺) (Fig. 6E and F). In all, adult monkeys exhibited a significantly higher proportion of Tfh1 cells in the lymph node (P = 0.013; FDR P = 0.021) (Fig. 6E), while infants had a significantly higher proportion of Tfh17 cells (P = 0.001; FDR P = 0.003) (Fig. 6G) at 12 wpi. No significant difference was observed in Tfh2 proportions between the two age groups (P = 0.066) (Fig. 6F). Thus, SHIV-infected infant macaques exhibited a distinct Tfh landscape compared to adult macaques.

Proportions of CH505-specific memory B cells are similar between infant and adult monkeys.

We evaluated systemic memory B cells in infant and adult monkeys at 0, 6, and 12 wpi (Fig. 7A and B). Additionally, we compared the frequencies of total B cells and germinal center (GC) B cells in lymph nodes at 12 wpi between both age groups (Fig. 7C to E). Due to a lack of sample availability, memory B cell populations at week 0 in the systemic compartment were evaluated in only 3 monkeys from each age group. Changes from the baseline in total memory B cells (CD14⁻ CD16⁻ CD20⁺ IgD⁻ CD27⁺) and CH505 gp120-specific memory B cells were not significantly different between age groups at 6 and 12 wpi (P = 1; 6 and 12 wpi for both parameters) (Fig. 7A and B). Similarly, the frequencies of total B cells (CD3⁻ CD20⁺) in lymph nodes were not significantly different (P = 0.143) (Fig. 7D) at 12 wpi. However, infants exhibited a significantly higher frequency of GC B cells, defined as CD20⁺ Bcl6⁺ Ki67⁺ (Fig. 7C), than did adults (P = 0.03; FDR P = 0.05) (Fig. 7E), consistent with high proportions of Tfh cells in lymph nodes at 12 wpi (Fig. 6D).

SHIV.C.CH505-infected infant and adult monkeys develop comparable virus tier 2 autologous plasma neutralization responses.

Since the kinetics, magnitude, and breadth of plasma HIV Env-specific IgG responses were similar between infant and adult monkeys, we next evaluated the HIV neutralization activity of these plasma IgG responses. We found that both age groups developed similar neutralization activity at 12 wpi against the tier 1 clade-matched isolates MW965 (50% infective dose [ID₅₀] ranges of 135 to 31,736 for infants and 324 to 4,949 for adults [P = 0.384]) (Fig. 8A) and CH505 w4.3 (ID₅₀ ranges of 45 to 950 for infants and 45 to 750 for adults [P = 0.605]) (Fig. 8B). Furthermore, neutralization activity against the autologous tier 2 neutralization-sensitive challenge virus CH505 T/F was not observed in the majority of monkeys from both age groups until 12 wpi (Fig. 8C). Eight out of 12 adults (66.7%) and 3 out of 6 infants (50%) developed autologous virus-neutralizing responses by 12 wpi, with no significant difference in potency (ID₅₀ ranges of 45 to 161 for infants and 45 to 380 for adults [P = 0.963]) (Fig. 8C).

ADCC activity of plasma antibodies develops similarly in adult and infant acutely SHIV-infected monkeys.

ADCC responses directed against CH505 gp120-coated target cells were measured using plasma samples collected from infant and adult RMs at 0, 6, and 10 wpi. ADCC activity was detectable in both age groups by week 6 of infection and was maintained through week 10, with no significant difference in the magnitude or kinetics of the response between the two groups (6 wpi, P = 0.494; 10 wpi, P = 0.591) (Fig. 9A). Additionally, similar kinetics were observed in the development of CH505-specific ADCC Ab titers in adults and infants, with no significant difference in magnitude (6 wpi, P = 0.801; 10 wpi, P = 0.807) (Fig. 9B).

Viral load at 12 wpi is significantly correlated with development of autologous virus neutralization in infant and adult monkeys.

We calculated Spearman correlation coefficients to determine if a subset of the immunological responses assessed could predict the development of autologous neutralization in infant and adult monkeys at 12 wpi. CH505-specific gp120 IgG responses (rho = 0.69; P = 0.002; FDR P = 0.07) (Fig. 10B), ADCC antibody titers at 10 wpi (rho = 0.63; P = 0.01; FDR P = 0.16) (Fig. 9C), and CH505-specific Tfh cell frequencies at 12 wpi (rho = 0.49; P = 0.05; FDR P = 0.19) (Fig. 10D) were correlated with the development of autologous neutralization. However, these results were not statistically significant after adjustment for multiple comparisons. Yet the plasma viral load at 12 wpi was correlated with autologous neutralization after adjustment for multiple comparisons (rho = 0.76; P < 0.001; FDR P = 0.03) (Fig. 10A). A summary of all the immune parameters assessed can be found in Fig. 11 and Table S1 in the supplemental material.

ADCC activity has been described to be important for a reduced risk of infection and/or viral control (7, 30, 34, 35) and has also been associated with levels of Env-specific IgG responses (34–37). Therefore, we calculated Spearman correlation coefficients to determine if a subset of the immunological parameters assessed was associated with the development of antibodies capable of mediating ADCC. Overall, none of the immune responses evaluated appeared to predict ADCC activity. A summary of all the immune parameters assessed can be found in Fig. 12 and Table S2.

Animal care and sample collection.

Adult female rhesus macaques ranged from 4 to 10 years of age, and infant rhesus macaques ranged from 6 weeks of age (Table 1). All macaques were of Indian origin and from the type D retrovirus-free, SIV-free, and simian T cell leukemia virus type 1 (STLV-1)-free colony of the California National Primate Research Center (CNPRC) (Davis, CA). Animals were maintained in accordance with American Association for Accreditation of Laboratory Animal Care standards and the Guide for the Care and Use of Laboratory Animals (59). For sample collections, animals were sedated with ketamine HCl (Parke-Davis) injected at 10 mg/kg of body weight. EDTA-anticoagulated blood was collected via peripheral venipuncture. Plasma was separated from whole blood by centrifugation, and PBMCs were isolated by density gradient centrifugation using Ficoll-Paque (Sigma) or lymphocyte separation medium (MP Biomedicals). All protocols were reviewed and approved by the University of California, Davis, Institutional Animal Care and Use Committee (IACUC) prior to the initiation of the study.

SHIV challenge of infant and adult monkeys.

The generation of SHIV.C.CH505.375H.dCT was previously described (20). The SHIV.C.CH505.375H.dCT challenge stock (provided by George M. Shaw, University of Pennsylvania) was prepared by infecting primary activated Indian rhesus macaque CD4⁺ T cells, and 7 to 14 days later, culture supernatants were pooled, as previously described (20). Virus titers were determined in TZM-bl cells, yielding 6.8 × 10⁶ TCID₅₀/ml.

Twelve adult monkeys were challenged intravenously with SHIV.C.CH505 at a dose of 3.4 × 10⁵ TCID₅₀ (Table 1). Six infants were challenged orally beginning at 4 weeks of age. Initially, infants were exposed to SHIV.C.CH505 three times per day for 5 days at a dose of 8.5 × 10⁴ TCID₅₀/ml in an isotonic sucrose solution and bottle-fed, in order to simulate oral acquisition via breastfeeding. After 1 week, only one infant became infected, and the remaining five infants were challenged 3 weeks later under sedation at a dose of 6.8 × 10⁵ TCID₅₀/ml until infected. After 3 weeks, one infant remained uninfected and thus was subsequently challenged with an increasing dose (3.4 × 10⁶ TCID₅₀/ml) until infected (Table 1).

Viral RNA load quantification.

Plasma RNA load was quantified using a well-established quantitative reverse transcriptase PCR (RT-PCR) assay targeting SIVgag RNA, as previously described (20). RNA was isolated from plasma samples using the QIAsymphony Virus/Bacteria Midi kit on the QIAsymphony SP automated sample preparation platform (Qiagen, Hilden, Germany). RNA was extracted manually if plasma volumes were limited. Data reported are the numbers of SIV RNA copy equivalents per milliliter of plasma, with a limit of detection of 15 copies/ml.

Lymphocyte counts.

Absolute lymphocyte counts in blood were calculated using the PBMC counts obtained by automated complete blood counts, multiplied by the lymphocyte percentages.

Enzyme-linked immunosorbent assay, recombinant protein, and soluble CD4 blocking.

Env binding IgG was assessed in plasma in a 384-well-plate format. The plates were coated overnight with HIV CH505 gp120 (30 ng/well) or MN gp41 (3 μg/ml) and then blocked with the assay diluent (phosphate-buffered saline [PBS] containing 4% whey, 15% normal goat serum, and 0.5% Tween 20). Serially diluted plasma samples were then added to the plates and incubated for 1 h, followed by detection with a horseradish peroxidase (HRP)-conjugated antibody, polyclonal goat anti-monkey IgG (Rockland Immunochemicals). The plates were developed by using the 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS-2) peroxidase substrate system (KPL). The monoclonal antibody b12R1 was used to develop standard curves, and the concentration of IgG antibody was calculated relative to the standard using a 5-parameter fit curve (SoftMax Pro 7). For monoclonal antibodies, the 50% effective concentration (EC₅₀) was calculated as the concentration of antibody which resulted in a 50% reduction in the optical density (OD) from the maximum value.

For CD4-blocking enzyme-linked immunosorbent assays (ELISAs), 384-well plates (Corning Life Sciences) were coated with C.1086 gp120 at 30 ng/well. Following the same steps as the ones stated above, plates were blocked with assay diluent, and serially diluted monoclonal antibody and plasma were added and incubated for 1 h. Soluble CD4 (sCD4) (NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH [human soluble CD4 recombinant protein from Progenics]) was then added at a concentration of 0.64 μg/ml. sCD4 binding was detected using biotinylated human anti-CD4 (Thermo Fisher Scientific) followed by HRP-conjugated streptavidin. Percent sCD4 binding inhibition was calculated as follows: 100 − (average of serum duplicate OD/average of negative-control OD) × 100. A reduction of absorbance by >50% by Abs present in plasma indicated blocking of sCD4 binding to C.1086 gp120. The monoclonal antibody VRC01, which was used as a standard in this assay, has CD4 binding-site specificity and exhibits 80% blocking of sCD4 binding.

Binding antibody multiplex assay.

HIV-1 epitope specificity and breadth were determined using a BAMA, as previously described (11). HIV-1 antigens were conjugated to polystyrene beads (Bio-Rad) as previously described (18), and binding of IgG to the bead-conjugated HIV-1 antigens was then measured in plasma samples from the infant and adult monkey cohorts. The positive control was purified IgG from pooled plasma of HIV Env-vaccinated rhesus macaques (RIVIG) (60). The conjugated beads were incubated on filter plates (Millipore) for approximately 30 min before plasma samples were added. The plasma samples were diluted in assay diluent (1% dry milk–5% goat serum–0.05% Tween 20 in 1× phosphate-buffered saline [pH 7.4]) at a 1:500 dilution. Beads and diluted samples were incubated for 30 min, and IgG binding was then detected using phycoerythrin (PE)-conjugated mouse anti-monkey IgG (Southern Biotech) at 4 μg/ml. The beads were washed and acquired on a Bio-Plex 200 instrument (Bio-Rad), and IgG binding was expressed as the mean fluorescence intensity (MFI). To assess the assay background, the MFI of binding to wells that did not contain beads or a sample (blank wells) and nonspecific binding of the samples to unconjugated blank beads were evaluated during assay analysis. High background detection for plasma samples was noted, and assays were repeated if necessary. An HIV envelope-specific antibody response was considered positive if it was above the lower limit of detection (MFI value of 100). To check for consistency between assays, the EC₅₀ and maximum MFI values of the positive control (RIVIG) were tracked by Levy-Jennings charts. The antigens conjugated to the polystyrene beads are as follows: C.1086 gp140, C.1086 gp120, A1.Con_env03 gp140, A233 gp120, B.Con_env03 gp140, Con6 gp120, ConC gp120, MN gp120, linear V2.B, V3.C, C5.2.C, C1, conformational V1V2, ConC V3, MN V3, and C.1086 V1V2 (see Table S1 in the supplemental material).

Linear peptide microarray mapping and data analysis.

Solid-phase peptide microarray epitope mapping was performed as previously described (61), with minor modifications. Briefly, JPT Peptide Technologies GmbH (Germany) prepares arrays slides by printing a library designed by B. Korber, Los Alamos National Laboratory, onto epoxy glass slides (PolyAn GmbH, Germany). The library contains 15-mer peptides overlapping by 12 residues, covering consensus Env (gp160) of clades A, B, C, and D; group M; CRF1; and CRF2 and vaccine strains (gp120) 1.A244, 1.TH023, MN, C.1086, C.TV1, and C.ZM651. To assess CH505-specific responses, a peptide library containing 59 CH505 strains (gp120, gp145, gp160, and SOSIP [sequences provided by Barton Haynes, Duke University]) was also designed (22). Sera were diluted 1:50 and applied to the peptide array, followed by washing and detection using goat anti-human IgG-Alexa Fluor 647 (AF647). Array slides were scanned at a wavelength of 635 nm with an InnoScan 710 AL scanner (Innopsys, France) using the XDR mode. Scan images were analyzed using MagPix 8.0 software to obtain binding intensity values for all peptides. Microarray data were then processing using the R package pepStat (62) to obtain a binding signal for each peptide, which is defined as log₂(intensity of the sample at 12 wpi/intensity of the matched baseline sample). The magnitude of binding to each identified epitope is defined as the highest binding signal by a single peptide within the epitope region.

Neutralization assays.

Neutralization by antibodies in plasma of MW965.LucR.T2A.ecto/293T IMC (clade C, tier 1), CH505 w4.3 HIV-1 pseudovirus (clade C, tier 1a), and autologous CH505 T/F (clade C, tier 2) HIV-1 pseudovirus was measured in TZM-bl cells (NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH [from John Kappes]) via a reduction in luciferase reporter gene expression after a single round of infection as previously described (63–65). Prior to screening, plasma was heat inactivated at 56°C for 30 min. Luminescence was measured using a Victor X3 multilabel plate reader, at 1 s per well (PerkinElmer). The ID₅₀ was calculated as the dilution that resulted in a 50% reduction in relative luminescence units (RLU) compared to virus control wells. The monoclonal antibody b12R1 was used as a positive control for MW965 assays, and VRC01 was used a positive control for all other assays.

ADCC.

The ADCC-GranToxiLux (GTL) assay was used to measure plasma ADCC activity as previously described (66). Briefly, CEM.NKR_CCR5 target cells (NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH [from Alexandra Trkola]) (67) were coated with recombinant CH505 gp120. Cryopreserved human PBMCs from an HIV-1-seronegative donor with the heterozygous 158 F/V genotype for Fcγ receptor IIIa were used as the source of effector cells (68, 69). Adult and infant plasma samples were tested after 4-fold serial dilution starting at 1:100. ADCC was measured as percent granzyme B (GzB) activity, defined as the frequency of target cells positive for proteolytically active GzB out of the total viable target cell population. Final results are expressed after subtraction of the background GzB activity observed in wells containing target and effector cells in the absence of plasma. ADCC endpoint titers were determined by interpolating the last positive dilution of plasma (>8% GzB activity).

Antigen-specific T cell responses.

HIV gp120-specific T cell responses were measured as described previously (70). Briefly, thawed cells were enumerated, and 2 × 10⁶ cells in suspension were cultivated in complete RPMI (RPM1 1640 with glutamine [Gibco] plus 10% heat-inactivated fetal bovine serum [FBS] [Serum Source International, Charlotte, NC] and 1× penicillin-streptomycin [Sigma-Aldrich, St. Louis, MO]) and stimulated with (i) no stimulant, (ii) 0.5× cell stimulation cocktail (eBioscience), or (iii) 10 μg of a consensus C HIV gp120 peptide pool (NIH AIDS Reagent Program) for 6 h, with 1× brefeldin A (eBioscience) present after the first hour. Cell preparations were then stained with a viability dye, and the following antibodies were used: CD3 allophycocyanin (APC)-Cy7 (clone SP34-2), CD4 PE-CF594 (L200), CD8 BV786 (RPA-T8), CD45RA V450 (5H9), CCR7 PE-Cy7 (3D12), IFN-γ AF700 (B27), IL-2 peridinin chlorophyll protein (PerCP)-Cy5.5 (MQ1-17H12), IL-17 PE (eBio64CAP17; eBioscience), and TNF-α APC (Mab11) (all antibodies were manufactured by BD Biosciences unless otherwise indicated). A minimum of 300,000 paraformaldehyde-fixed lymphocyte events were acquired on a BD LSR Fortessa instrument using FACSDiva v8.0 software. Data were analyzed using FlowJo software v10.5.3 (TreeStar, Ashland, OR).

CH505 envelope-specific memory B cell phenotyping.

For phenotyping of CH505 Env-specific memory B cells, a suspension of 10⁶ PBMCs was blocked with 6.25 μg/ml anti-human CD4 antibody (BD Biosciences) at 4°C for 15 min. After incubation, PBMCs were washed twice with PBS and pelleted at 1,500 rpm for 5 min. PBMCs were then incubated at 4°C with a Live/Dead fixable aqua dead cell stain kit (Thermo Fisher Scientific) for 30 min. Following incubation and washing with PBS, PBMCs were then stained with a cocktail of fluorescently conjugated antibodies for surface markers, including CD20 fluorescein isothiocyanate (FITC) (clone L27), CD3 PerCP-Cy5.5 (SP34-2), IgM PE-Cy5 (G20-127), CD16 PE-Cy7 (3G8), CD8 PE-Texas Red (3B5), IgD PE (polyclonal), CD14 BV570 (M5E2), CD27 APC-Cy7 (O323), and custom-conjugated BV421–HIV-1 gp120 (C.CH505 T/F) and AF647–HIV-1 gp120 (C.CH505 T/F), prepared as described previously (71). The stained PBMCs were acquired on an LSRII flow cytometer (BD Biosciences) using BD FACSDiva software and analyzed as described above. The following gating strategy was applied: lymphocytes were gated on singlets, and live cells were selected to gate on CD3⁻ cells (T cells), CD14⁻ cells (monocytes/macrophages), and CD20⁺ cells (B cells). B cells were further gated on CD27⁺ memory B cells. Only B cells positive for both BV421-gp120 and AF647-gp120 were considered CH505-Env specific.

Lymph node Tfh phenotyping and activation-induced marker assay.

The activation-induced marker (AIM) assay was based on previous work (72). Briefly, cryopreserved rhesus macaque lymph node cells (12 wpi) were thawed, rested for 3 h at 37°C with 5% CO₂, resuspended in AIM V medium (Gibco), and transferred at 10⁶ cells per well to a 24-well plate. Cells were cultured for 18 h at 37°C with 5% CO₂, with no exogenous stimulation (negative control) or with gp140 stimulation (5 μg/ml CH505 T/F gp140 protein and 0.5 μg/ml of a 15-mer peptide pool with an 11-residue overlap spanning CH505 T/F gp140). As a positive control, cells were stimulated with 0.5 μg/ml staphylococcal enterotoxin B (SEB) (Sigma). Duplicates for each condition were performed when cell numbers permitted. Following stimulation, cells were labeled with fluorescently labeled antibodies to the following surface antigens: PD1 BV421 (clone EH12.2H7), CD8a BV570 (RPA-T8), CD25 BV650 (BC96), CD4 BV711 (OKT4), CD69 PE-Cy5 (FN50), CD45RA FITC (MEM-56), CD137 AF647 (4B4-1), and CD185 PE-Cy7 (MU5UBEE) (BioLegend) as well as CD3 APC-Cy7 (clone SP34-2), CD20 FITC (L27), CD196 PerCP-Cy5.5 (11A9), OX40 PE (L106), and CD183 PE-CF594 (1C6/CXCR3) (BD Biosciences). For Tfh phenotyping, the following gating strategy was applied (Fig. 6A): lymphocytes were gated on singlets, and live cells were selected to gate on CD4⁺ CXCR5⁺ Foxp3⁻ cells, followed by gating on CCR6 and CXCR3. Cell viability was measured using Live/Dead fixable aqua stain (eBioscience). Flow cytometry data were acquired on an LSRII instrument running FACSDiva software (BD Biosciences) and analyzed as described above.

Statistical methods.

Immune assay measurements at various time points postinfection and the change in immune assay measurements from the baseline were compared between SHIV-infected infant and adult monkeys using Wilcoxon rank sum tests with exact P values. Spearman’s rank correlation coefficients were estimated for the cohort as a whole as well as by adult monkeys and infants separately. All correlations were tested with exact P values to assess whether any were significantly different from zero. To adjust for multiple comparisons, the Benjamini-Hochberg (BH) procedure was used to control the false discovery rate (FDR). Separate adjustments to control the FDR at an α value of 0.05 were performed for comparisons between the infant and adult monkeys using (i) the prespecified primary endpoints for a total of 26 tests (Table S3), (ii) the prespecified secondary endpoints for a total of 43 tests (Table S4), (iii) correlations between prespecified parameters and CH505 T/F neutralization at 12 wpi for a total of 60 tests (Table S1), and (iv) correlations between prespecified parameters and ADCC minus percent GzB activity at 10 weeks postinfection for a total of 15 tests (Table S2). Both the unadjusted P values (raw P) and FDR-adjusted P values (FDR P) are reported in the supplemental material. All statistical tests were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).