Abstract
Anti-HIV-1 broadly neutralizing antibodies (bNAbs) have the dual
potential of mediating virus neutralization and antiviral effector
functions through their Fab and Fc domains, respectively. So far, bNAbs
with enhanced Fc effector functions in vitro have only been tested in
NHPs during chronic simian-HIV (SHIV) infection. Here, we investigate
the effects of administering in acute SHIV[AD8-EO] infection either
wild-type (WT) bNAbs or bNAbs carrying the S239D/I332E/A330L (DEL)
mutation, which increases binding to FcγRs. Emergence of virus in
plasma and lymph nodes (LNs) was delayed by bNAb treatment and occurred
earlier in monkeys given DEL bNAbs than in those given WT bNAbs,
consistent with faster clearance of DEL bNAbs from plasma. DEL
bNAb-treated monkeys had higher levels of circulating virus-specific
IFNγ single-producing CD8^+ CD69^+ T cells than the other groups. In
LNs, WT bNAbs were evenly distributed between follicular and
extrafollicular areas, but DEL bNAbs predominated in the latter. At
week 8 post-challenge, LN monocytes and NK cells from DEL bNAb-treated
monkeys upregulated proinflammatory signaling pathways and LN T cells
downregulated TNF signaling via NF-κB. Overall, bNAbs with increased
affinity to FcγRs shape innate and adaptive cellular immunity, which
may be important to consider in future strategies of passive bNAb
therapy.
Subject terms: HIV infections, Antibodies, Preclinical research,
Translational immunology
__________________________________________________________________
In this work, the authors study the immunological and virological
effects of administering either wild-type anti-HIV-1 broadly
neutralizing antibodies (bNAbs) or bNAbs with a mutation that increases
binding to Fc-gamma receptors (FcγRs) to rhesus macaques in the acute
phase of SHIV[AD8-EO] infection.
Introduction
To date, no effective and widely available approach to cure human
immunodeficiency virus type 1 (HIV-1) infection has been found. Daily
antiretroviral therapy (ART) is the current standard of care that
blocks new rounds of viral replication and can reduce plasma viral load
to undetectable levels while effectively limiting or eliminating
potential for transmission^[90]1,[91]2. However, HIV-1 persists despite
ART in infected cell populations, which contribute to viral
reservoirs^[92]3–[93]5 and lead to rapid rebound of plasma viremia once
treatment is discontinued^[94]6,[95]7. Thus, it is imperative to find
new therapeutic interventions for HIV-1 infection.
Over the past decade, potent anti-HIV-1 broadly neutralizing antibodies
(bNAbs) have been identified and gained considerable attention as tools
for HIV-1 prevention and treatment due to their ability to neutralize a
wide variety of HIV-1 strains with high potency. Anti-HIV-1 bNAbs
target different sites of the HIV-1 envelope (Env) glycoprotein such as
the CD4-binding site^[96]8–[97]11, the variable regions 1 and 2
(V1/V2)^[98]12–[99]14, the V3-glycan site^[100]15–[101]17, and the
membrane proximal external region^[102]18. Alone or in combination,
they have been tested in strategies for treatment of simian-HIV (SHIV)
infection in non-human primates (NHPs) and HIV-1 infection in humanized
mice and humans. bNAbs administered to NHPs or humans in the chronic
phase of infection were well tolerated and had a robust, but transient,
antiviral activity, as the decline in plasma viral load observed
shortly after their administration was followed by viral rebound to
pre-treatment levels^[103]19–[104]25. bNAb treatment initiated in the
acute phase of infection in infant and adult NHPs led to efficient
reduction and, in some cases even suppression, of viral
replication^[105]26–[106]29. In one of those studies, Nishimura et al.
^[107]29 treated SHIV[AD8-EO]-infected NHPs at days 3, 10, and 17
post-challenge with a combination of the wild-type (WT) bNAbs 3BNC117
and 10–1074, which target the CD4-binding site and V3-glycan site on
the HIV-1 Env, respectively^[108]10,[109]15. This treatment regimen
resulted in long-term suppressed plasma viremia in about half of the
treated animals in a CD8^+ T cell-dependent manner, as evidenced by
rapid reemergence of viremia in controller animals following CD8^+ T
cell depletion^[110]29. These findings led to the hypothesis that the
presence of bNAbs early in infection and before natural antibodies
occur may lead to the formation of immune complexes, which, following
cross-presentation to CD8^+ T cells, may boost antiviral immune
responses^[111]30 and lead to viral control.
The Fab domain of antiviral monoclonal antibodies (mAbs) can mediate
the neutralization of a virus, thereby blocking its entry into cells.
In addition, the Fc domain can recognize and clear virus-infected cells
by engaging host immune components in Fc-mediated effector functions,
such as antibody-dependent cellular cytotoxicity (ADCC), phagocytosis
(ADCP), and complement-dependent cytotoxicity^[112]31. Several studies
using Fc domain-engineered anti-HIV-1 bNAbs suggested that Fc-mediated
effector functions contribute to the in vivo antiviral activity of
bNAbs in humanized mice^[113]32–[114]35. Despite conflicting results
obtained from studies in NHPs^[115]36–[116]39, two recent studies
quantified the relative contribution of Fc-mediated vs. Fab-mediated
functions to the antiviral activity of anti-HIV-1 bNAbs and reported
that the former accounted for about 20 to 45% of the overall
activity^[117]40,[118]41. In these studies, anti-HIV-1 bNAbs with
enhanced or impaired Fc-mediated effector functions in vitro were
infused in HIV-1-infected humanized mice and chronically SHIV-infected
NHPs^[119]40,[120]41. However, no study thus far has investigated the
effects of administering anti-HIV-1 bNAbs with enhanced Fc effector
functions in the acute phase of SHIV or HIV infection.
In this study, we evaluated the virological and immunological effects
of administering early after SHIV[AD8-EO] infection either WT bNAbs or
bNAbs carrying the S239D/I332E/A330L (DEL) mutation, a mutation
introduced in the Fc domain that increases binding affinity to rhesus
FcγRIII and FcγRII and enhances Fc-mediated effector functions in
vitro^[121]40,[122]42. Notably, we extended our investigation to the
lymph nodes (LNs), where we studied in detail the presence and
localization of infused bNAbs as well as their effects at the cellular
level. WT or DEL bNAbs were administered at days 3, 10, and 17
post-intrarectal SHIV[AD8-EO] challenge, similar to the early WT bNAb
therapy regimen followed by Nishimura et al. ^[123]29. Treatment of
monkeys with DEL bNAbs allowed us to additionally test if early
administration of bNAbs with increased affinity to FcγRs could boost
antiviral effects. We used a combination of VRC07-523-LS and PGT121,
which target non-overlapping epitopes on the HIV-1 Env^[124]17,[125]43
and have similar pharmacokinetic (PK) profiles upon intravenous
infusion into NHPs^[126]20,[127]21,[128]43,[129]44. VRC07-523 is an
engineered, more potent variant of VRC01 that targets the CD4-binding
site of the HIV-1 Env^[130]9,[131]43, and the LS (M428L/N434S) mutation
introduced in its Fc domain extends its plasma half-life in
NHPs^[132]43,[133]45. PGT121 is a potent bNAb that targets
glycan-dependent V3 epitopes of the HIV-1 Env^[134]17. In monkeys
infected with SHIV[AD8-EO] and subjected to this early bNAb therapy
regimen, emergence of virus in plasma and LNs was delayed
proportionally to the duration of the infused bNAbs in circulation. DEL
bNAb-treated monkeys developed distinct circulating virus-specific
CD8^+ T cell responses and LN immune cell signatures that set them
apart from monkeys that received WT bNAbs or were left untreated.
Results
Early passive bNAb therapy suppressed acute-stage plasma viremia in
SHIV[AD8-EO]-infected rhesus macaques
To evaluate the therapeutic effect of administering a combination of
anti-HIV-1 bNAbs during acute infection, 30 Indian origin rhesus
macaques were inoculated intrarectally with 1000 median tissue culture
infective doses (TCID[50]) of SHIV[AD8-EO.] Monkeys were then either
left untreated (control group, n = 10) or intravenously infused at days
3, 10, and 17 post-challenge with 10 mg/kg each of VRC07-523-LS and
PGT121 (WT bNAb treatment, n = 10) or VRC07-523-LS/DEL and PGT121/DEL
(DEL bNAb treatment, n = 10) (Fig. [135]1A). VRC07-523-LS and PGT121
neutralization of the SHIV[AD8-EO] challenge stock was similar to that
of 3BNC117 and 10-1074, respectively, which were previously used by
Nishimura et al. in a similar therapeutic strategy^[136]29
(Table [137]1). In addition, the DEL mutation did not affect
neutralization of SHIV[AD8-EO] (Table [138]1).
Fig. 1. Study design and therapeutic efficacy of anti-HIV-1 bNAb
combinations.
[139]Fig. 1
[140]Open in a new tab
A Thirty rhesus macaques were inoculated intrarectally with
SHIV[AD8-EO]and either left untreated (n = 10) (left) or treated at
days 3, 10, and 17 post-challenge with a combination of either
VRC07-523-LS and PGT121 (WT bNAbs, n = 10) (middle) or VRC07-523-LS/DEL
and PGT121/DEL (DEL bNAbs, n = 10) (right). Each bNAb was infused
intravenously at 10 mg/kg. B Plasma viral load up to 72 weeks
post-challenge in untreated (left), WT bNAb-treated (middle), and DEL
bNAb-treated (right) monkeys. Gray and black lines represent individual
and median plasma viral loads, respectively. Vertical dotted lines
indicate the timings of bNAb infusions. C Time to first detectable
plasma virus leading to persistent infection (left) and to peak plasma
viremia (right) in untreated and bNAb-treated monkeys. N = 9, 8, and 7
untreated, WT bNAb-treated, and DEL bNAb-treated monkeys, respectively.
P < 0.0001 (left) and p = 0.0007 (right). D Peak plasma viral load
(left) and plasma viral load at week 36 post-challenge (right) in
untreated and bNAb-treated monkeys. N = 9, 8, and 7 (left) and n = 10,
10, and 10 (right) untreated, WT bNAb-treated, and DEL bNAb-treated
monkeys, respectively. E Plasma viral load throughout the first 72
weeks post-challenge in untreated and bNAb-treated monkeys as
determined by AUC analysis. N = 9, 8, and 7 untreated, WT bNAb-treated,
and DEL bNAb-treated monkeys, respectively. Individual data from the
initial 6 monkeys per group are indicated with gray solid lines (B) and
circles (C–E), whereas data pertaining to the later 4 monkeys per group
are denoted with gray dashed lines (B) and squares (C–E) (see text for
more details on the two monkey batches). Bar graphs show the mean and
individual datapoints (C–E). The Kruskal–Wallis test followed by Dunn’s
multiple comparison test was used to detect significant differences
between monkey groups. N indicates the number of biological replicates
(monkeys). Source data are provided in the Source Data file. AUC area
under the curve, IR intrarectal, IV intravenous; pc post-challenge,
TCID[50] median tissue culture infective doses, Tx treatment, UnTx
untreated, wk week.
Table 1.
Neutralization potency of anti-HIV-1 bNAbs against the SHIV[AD8-EO]
challenge stock
bNAb IC[50] (µg/mL) IC[80] (µg/mL)
VRC07-523-LS^a 0.723 1.570
VRC07-523-LS/DEL^a 0.731 1.680
PGT121^a 0.167 0.375
PGT121/DEL^a 0.168 0.413
3BNC117^b 0.517 1.000
10-1074^b 0.210 0.389
[141]Open in a new tab
^aDilution points for each bNAb were run in duplicate. bNAbs were
tested against SHIV[AD8-EO] in at least 5 experiments, with no more
than 2-fold variability between them; representative IC[50] and IC[80]
values are shown.
^bData for 3BNC117 and 10–1074 were obtained from Pegu et al., Cell
Host & Microbe, 2019.
Peripheral blood samples were collected from these animals at multiple
timepoints throughout 17 months to monitor plasma viremia. Viremia was
detected in untreated monkeys within, on average, the first 2 weeks
post-challenge (Fig. [142]1, B and C left). In contrast, in treated
monkeys, viremia was overall suppressed for the first 10 weeks
post-challenge upon WT bNAb treatment and 6 weeks post-challenge upon
DEL bNAb treatment (Fig. [143]1, B, C left). Thus, both the time to
first measurable plasma virus leading to persistent infection and the
time to reach peak viremia were significantly longer in the WT
bNAb-treated group than in the untreated group (p < 0.001 for both
comparisons) (Fig. [144]1C). However, the peak and set-point viral
loads (Fig. [145]1D) and the overall plasma viremia as measured by area
under the curve (AUC) (Fig. [146]1E) did not vary significantly with
any of the bNAb treatments. Taken together, these data show that triple
administration of bNAbs targeting two distinct and non-overlapping
viral epitopes in the acute phase of SHIV[AD8-EO] infection delays
onset of plasma viremia without affecting the overall plasma viral
burden.
One untreated monkey, two WT bNAb-treated monkeys, and three DEL
bNAb-treated monkeys did not develop plasma viremia for more than 30
weeks post-challenge, as evaluated by standard (Fig. [147]1, B and D
right) and ultrasensitive (Supplementary Table [148]1) SIV Gag RNA
quantitative RT-PCR (qRT-PCR) with detection limits of 15 and 1
copies/mL, respectively. To ascertain the infection status of these
monkeys, we depleted their CD8^+ T cells through intravenous infusion
of the anti-CD8β mAb CD8b255R1, which specifically targets macaque
CD8^+ T cells (Supplementary Fig. [149]1A). Circulating CD8^+ T cells
were selectively depleted during the 36 weeks that followed anti-CD8β
mAb infusion (Supplementary Fig. [150]1, B, C). In contrast, the levels
of other T cell subsets, B cells, monocytes, and NK cells quickly
returned to baseline after the first 6 h of anti-CD8β mAb infusion
(Supplementary Fig. [151]1, B–D). Despite efficient depletion of CD8^+
T cells, no burst in plasma viremia occurred up to 21 days post-mAb
infusion (Supplementary Table [152]2). We also did not detect
cell-associated SIV Gag DNA or RNA in peripheral blood mononuclear
cells (PBMCs) from the DEL bNAb-treated, anti-CD8β mAb-infused monkeys
up to 21 days post-mAb infusion (Supplementary Table [153]3).
Therefore, we concluded that these monkeys never became infected and
that the SHIV[AD8-EO] stock used in this study, when given
intrarectally at the dose of 1000 TCID[50], does not infect all
monkeys. Unless otherwise indicated, these uninfected monkeys were
excluded from further analyses.
Of note, this study was performed in two separate batches, the first
with 6 monkeys per group (depicted in full gray lines in Fig. [154]1B
and circles in Fig. [155]1, C–E) and the second with 4 monkeys per
group (depicted in dashed lines in Fig. [156]1B and squares in
Fig. [157]1, C–E). Plasma viremia did not segregate the animals into
the two different batches (Fig. [158]1, B–E), thus viremic monkeys from
both batches were grouped together for all subsequent analyses.
bNAbs persisted in plasma throughout the period of infusions and
progressively declined thereafter as strong anti-drug antibody (ADA)
responses developed
Given the different patterns of virus emergence between WT and DEL
bNAb-treated monkeys, we hypothesized that the circulating titers of WT
and DEL bNAbs might differ. To test this hypothesis, we measured
longitudinally by enzyme-linked immunosorbent assay (ELISA) the plasma
concentration of the infused bNAbs in all animals (both infected and
uninfected) up to 14 weeks post-challenge. Right after the first
infusion, all bNAbs reached plasma levels that well exceeded their
IC[80] values for in vitro neutralization against SHIV[AD8-EO]
(Fig. [159]2A and Supplementary Fig. [160]2A). bNAb levels increased
after each infusion timepoint, with VRC07-523-LS and PGT121 reaching
their average peak concentrations of 113 and 139 μg/mL, respectively,
at week 3 post-challenge (i.e., 4 days after the third infusion), and
VRC07-523-LS/DEL and PGT121/DEL reaching their average peak
concentrations of 15 and 56 μg/mL, respectively, at week 2
post-challenge (i.e., 4 days after the second infusion). Overall levels
of the WT bNAbs as measured by AUC were significantly higher than of
their corresponding DEL mutants (p < 0.001 for both VRC07-523-LS vs.
VRC07-523-LS/DEL and PGT121 vs. PGT121/DEL) (Fig. [161]2B). In
addition, while the WT bNAbs had similar PK profiles, the levels of
PGT121/DEL were significantly higher than those of VRC07-523-LS/DEL
(p < 0.001) (Fig. [162]2B). Shortly after the third infusion, bNAb
levels started declining over time (Fig. [163]2A and Supplementary
Fig. [164]2A). VRC07-523-LS and PGT121 reached undetectable levels
within 5–12 and 6–11 weeks post-challenge, respectively, with the
exception of one monkey in which both bNAbs were still detectable 14
weeks after challenge (Supplementary Fig. [165]2A). VRC07-523-LS/DEL
and PGT121/DEL remained in circulation for a shorter time, reaching
undetectable levels within 2–5 and 4–9 weeks post-challenge,
respectively (Supplementary Fig. [166]2A).
Fig. 2. PK of infused bNAbs and ADA responses.
[167]Fig. 2
[168]Open in a new tab
A Concentration of VRC07-523-LS (dark pink), PGT121 (dark blue),
VRC07-523-LS/DEL (light pink), and PGT121/DEL (light blue), as measured
by ELISA in the plasma of monkeys that received either the WT or DEL
bNAbs at days 3, 10, and 17 post-SHIV[AD8-EO] challenge. Vertical
dotted lines indicate the timings of bNAb infusions and horizontal
dotted lines indicate the in vitro neutralization IC[80] values against
SHIV[AD8-EO] averaged for VRC07-523-LS and VRC07-523-LS/DEL (pink), and
PGT121 and PGT121/DEL (blue). N = 10 for each bNAb and each timepoint
except for VRC07-523-LS and PGT121 at week 0 and VRC07-523-LS/DEL and
PGT121/DEL at weeks 10–14: n = 4; PGT121 at week 9: n = 9. B Plasma
concentration of each infused bNAb up to 14 weeks post-challenge as
determined by AUC analysis. N = 10 for each bNAb. Each P < 0.0001,
two-sided p-values. C ADA response against VRC07-523-LS (dark pink),
PGT121 (dark blue), VRC07-523-LS/DEL (light pink), and PGT121/DEL
(light blue), as measured by ELISA in the plasma of monkeys that
received either the WT or DEL bNAbs at days 3, 10, and 17
post-SHIV[AD8-EO] challenge. Vertical dotted lines indicate the timings
of bNAb infusions. N = 10 for each bNAb and each timepoint except for
α-VRC07-523-LS at weeks 0 and 1, α-PGT121 at week 0, and
α-VRC07-523-LS/DEL at week 6: n = 9; α-VRC07-523-LS/DEL at week 4:
n = 8; α-VRC07-523-LS at days 10 and 17 and α-PGT121 at days 3, 10, and
17: n = 4, and α-VRC07-523-LS at day 3: n = 3. D ADA response against
each infused bNAb up to 14 weeks post-challenge as determined by AUC
analysis. N = 10 for each bNAb. P = 0.0089 for α-VRC07-523-LS vs.
α-VRC07-523-LS/DEL and p = 0.0029 for α-PGT121 vs. α-PGT121/DEL,
two-sided p-values. Graphs show the mean±SEM (A, C) and the mean and
individual datapoints (B, D). The Mann-Whitney test was used to detect
significant differences between the WT bNAbs, the DEL bNAbs, or the WT
and DEL mutant of the same bNAb. N indicates the number of biological
replicates (monkeys). Source data are provided in the Source Data file.
ADA anti-drug antibody; AUC area under the curve.
Clearance of bNAbs from the plasma occurred concomitantly with the
development of strong ADA responses (Fig. [169]2C and Supplementary
Fig. [170]2B). ADA responses against the DEL bNAbs were significantly
stronger than those against their WT counterparts (p = 0.009 for
VRC07-523-LS vs. VRC07-523-LS/DEL and p = 0.003 for PGT121 vs.
PGT121/DEL) (Fig. [171]2D), consistent with the faster clearance of DEL
bNAbs from circulation. As expected, the overall magnitude of ADA
responses inversely correlated with the duration of bNAbs in plasma
(Spearman’s r = −0.8273 and p < 0.0001 for VRC07-523-LS and
VRC07-523-LS/DEL, and Spearman’s r = −0.9195 and p < 0.0001 for PGT121
and PGT121/DEL) (Supplementary Fig. [172]2C), which in turn strongly
correlated with the time to onset of persistent infection (Spearman’s
r = 0.8991 and p < 0.0001 for VRC07-523-LS and VRC07-523-LS/DEL, and
Spearman’s r = 0.9158 and p < 0.0001 for PGT121 and PGT121/DEL)
(Supplementary Fig. [173]2D).
To determine whether the virus emerged due to the decline in bNAb
levels or to the development of mutations conferring resistance to the
infused bNAbs, we sequenced and analyzed the diversity of SHIV[AD8-EO]
Env at the time of peak viremia. An average of 1160 full-length
single-genome Env sequences per animal (range 286–2329) were recovered
via high-throughput single-genome sequencing (HT-SGS)^[174]46. About
half of the animals available for this analysis retained the WT virus
sequence at peak viremia (Supplementary Fig. [175]2E) and its frequency
did not vary significantly across the monkey groups (Supplementary
Fig. [176]2F). Haplotypes that, in some monkeys, completely overtook
the WT sequence included mutations located outside the CD4-binding
sites and the V3 loop of the virus Env (Supplementary Fig. [177]2E and
Supplementary Table [178]4) and were unlikely to represent escape
mutations to the infused bNAbs.
Taken together, these data show that bNAbs infused early after
SHIV[AD8-EO] infection persisted in circulation throughout the time of
infusions without inducing selective pressure on the virus, which only
emerged as strong ADA responses developed and bNAb levels went down.
All infused bNAbs potently neutralized the challenge virus SHIV[AD8-EO] ex
vivo
To ensure that the infused bNAbs in circulation retained their capacity
to neutralize SHIV[AD8-EO], we measured ex vivo the neutralizing
activity of plasma from both infected and uninfected monkeys against
the challenge virus SHIV[AD8-EO]. We used virus SIVmac251.30.SG3 as
negative control and viruses 00836-2.5 and X2088.c9, each of them
sensitive to only one of the two infused bNAbs (Supplementary
Table [179]5), to individually assess the neutralizing capacity of each
bNAb. Plasma samples from treated animals showed high neutralizing
titers (ID[50] and ID[80] values) against SHIV[AD8-EO], 00836-2.5, and
X2088.c9 throughout the first 3 weeks post-challenge (Supplementary
Fig. [180]3, A and B), consistent with the PK profile of these bNAbs
(Fig. [181]2A and Supplementary Fig. [182]2A). Interestingly, despite
the lower plasma concentrations of DEL bNAbs (Fig. [183]2B), plasma
from these monkeys neutralized both SHIV[AD8-EO] and X2088.c9 to a
similar extent as plasma from WT bNAb-treated monkeys, and only
00836-2.5 to a significantly lower extent (p = 0.03 for both ID[50] and
ID[80] AUC) (Supplementary Fig. [184]3C). The predicted plasma
neutralizing titers, derived from the measured plasma bNAb
concentrations and the in vitro neutralizing activity of each bNAb,
very closely matched the experimental neutralizing titers
(Supplementary Fig. [185]3A, B). Together, these data showed that each
of the infused bNAbs retained their capacity to neutralize the
challenge virus despite the existence of ADA responses.
We also tested the plasma samples from infected monkeys for endogenous
neutralization breadth against a wide panel of HIV-1 Env pseudoviruses
from clades A, B, and C (Supplementary Tables [186]6 and [187]7).
Plasma samples were tested at weeks 64 and 116 post-challenge, long
after the infused bNAbs had been cleared from the circulation. Both
untreated and treated monkeys developed heterologous clade B
neutralization as well as some levels of cross-clade neutralization
(Supplementary Tables [188]6 and [189]7). However, the neutralization
scores, calculated to summarize the plasma neutralizing capacity
against all tested viruses, were not higher upon bNAb treatment
(Supplementary Tables [190]6 and [191]7 and Supplementary
Fig. [192]3D). Thus, early bNAb administration did not improve
long-term neutralizing responses in treated monkeys.
Early bNAb treatment delayed the emergence of SHIV[AD8-EO] RNA in LNs
We next investigated the effect of early bNAb treatment in LN viremia.
LN sections were stained using the RNAscope in situ hybridization
technology (Fig. [193]3A), and the levels of SHIV[AD8-EO] RNA^+ cells
and SHIV[AD8-EO] RNA^+ virions were quantified and normalized to the LN
areas for accurate comparisons of monkey groups. The total LN areas
analyzed did not differ considerably among groups (Supplementary
Fig. [194]4A). At week 2 post-challenge, SHIV[AD8-EO] RNA^+ cells and
virions were detected in the LNs of the two untreated monkeys available
for this analysis, whereas their levels were either null or very low in
monkeys that received either WT or DEL bNAbs (Fig. [195]3B). In
contrast, at week 8 post-challenge, SHIV[AD8-EO] RNA^+ cells and
virions were detected in DEL bNAb-treated monkeys in addition to
untreated monkeys but were still either undetectable or very low in
monkeys that received the WT bNAbs (p = 0.04 for WT vs. DEL
bNAb-treated groups for both SHIV[AD8-EO] RNA^+ cells and virions)
(Fig. [196]3B). This virus detection pattern in LNs generally matched
the emergence of virus in plasma, which occurred at approximately weeks
2, 6, and 10 post-challenge in untreated, DEL bNAb-treated, and WT
bNAb-treated monkeys, respectively (Fig. [197]1B, C left). In fact,
plasma viral loads strongly correlated with the levels of LN
SHIV[AD8-EO] RNA^+ cells (Spearman’s r = 0.8760, p < 0.0001) and
virions (Spearman’s r = 0.7632, p < 0.0001) at weeks 2 and 8
post-challenge (Supplementary Fig. [198]4B, C).
Fig. 3. Detection of SHIV[AD8-EO] RNA in LN sections.
[199]Fig. 3
[200]Open in a new tab
A Representative RNAscope staining of LN sections from monkeys that
were left untreated or were treated with bNAbs early after SHIV[AD8-EO]
challenge. Staining was performed in one LN section per timepoint per
animal. Big images (scale bars: 1000 µm) show SHIV[AD8-EO] RNA (green),
CD20 (yellow), and cellular nucleic acids (SYTO, blue) in untreated
(left), WT bNAb-treated (middle), and DEL bNAb-treated (right) monkeys
at weeks 2 (top) and 8 (bottom) post-challenge. Small images show a LN
SHIV[AD8-EO] RNA^+ cell and virions in the DEL bNAb-treated monkey at
week 8 post-challenge. Top image (scale bar: 4 µm) shows green spheres
denoting the SHIV[AD8-EO] RNA^+ cell and virions; top middle image
(scale bar: 4 µm) shows the real signal for the SHIV[AD8-EO] RNA^+
cell; and the middle bottom and bottom images (scale bars: 2 µm) show
the spheres and real signal for virions, respectively. Viral events are
shown in association with cellular nucleic acids (blue) alone or in
combination with either CD20 (yellow) or CD3 (red) and CD4 (turquoise).
B, C RNAscope quantification of SHIV[AD8-EO] RNA^+ cells (left) and
virions (right) in whole LN sections and in follicular (circles) and
extrafollicular (triangles) areas from untreated (n = 2), WT
bNAb-treated (n = 5), and DEL bNAb-treated (n = 3) monkeys. P = 0.0357
for WT vs. DEL group at week 8 (B, left and right) and p = 0.0238 for
week 2 vs. 8 in WT group (B, right), two-sided p-values. D Percentage
of follicular SHIV[AD8-EO] RNA^+ cells (squares) and virions
(triangles) in untreated (n = 2) and DEL bNAb-treated (n = 3) monkeys.
Bar graphs show the mean and individual datapoints (B–D). The
Mann-Whitney test was used to detect significant differences in viral
parameters between WT and DEL bNAb-treated monkeys at each timepoint
(B), between timepoints for each bNAb-treated group in whole LN
sections (B) and in follicular or extrafollicular areas (C), and
between follicular and extrafollicular areas at each timepoint in DEL
bNAb-treated monkeys (C). N indicates the number of biological
replicates (monkeys). Source data are provided in the Source Data file.
Tx treatment, UnTx untreated, vRNA viral RNA.
To determine the localization of SHIV[AD8-EO] RNA in the LNs of
untreated and DEL bNAb-treated monkeys, SHIV[AD8-EO] RNA^+ cells and
virions were quantified in the follicular and extrafollicular areas,
which were defined by the presence and absence of CD20 expression,
respectively (Fig. [201]3A). At week 2 post-challenge, viral events
were similarly distributed across the LN follicular and extrafollicular
areas in untreated monkeys (average percentage of follicular
SHIV[AD8-EO] RNA^+ cells and virions: 56.0% and 52.4%, respectively),
whereas at week 8 post-challenge, they tended to predominantly
concentrate in the follicles in both untreated and DEL bNAb-treated
monkeys (average percentage of follicular events: 75.5% and 65.6% for
SHIV[AD8-EO] RNA^+ cells and 65.2% and 74.5% for SHIV[AD8-EO] RNA^+
virions in untreated and DEL bNAb-treated monkeys, respectively)
(Fig. [202]3C, D). Similar to the total LN area, the follicular and
extrafollicular areas that were analyzed did not significantly differ
across monkey groups (Supplementary Fig. [203]4D).
Altogether, these data indicate that, like in plasma, early bNAb
treatment delayed the emergence of virus in LNs, and that the location
of SHIV[AD8-EO] RNA^+ cells and virions in the LNs may vary during the
course of the infection.
Infused WT and DEL bNAbs were differently located in the LNs
Next, we investigated if infused bNAbs were also present in the LNs of
treated monkeys (both infected and uninfected). LN sections from weeks
2 and 8 post-challenge (i.e., 4 days after the second bNAb infusion and
approximately 5 weeks after all 3 infusions, respectively) were stained
for confocal microscopy (Fig. [204]4A). Antibodies that recognize
either the kappa or lambda light chain of human antibodies were used to
identify VRC07-523-LS and VRC07-523-LS/DEL, which express kappa light
chains^[205]9,[206]43, and PGT121 and PGT121/DEL, which express lambda
light chains^[207]47 (Fig. [208]4A). Importantly, these antibodies did
not cross-react to rhesus antibodies. Like in the RNAscope analysis,
the numbers of bNAb^+ cells were normalized to the LN areas, which did
not vary considerably across monkey groups (Supplementary
Fig. [209]4E). WT and DEL bNAbs were detected in LNs at weeks 2 and 8
post-challenge (Fig. [210]4B). In some monkeys, the levels of DEL
bNAb^+ cells were higher than those of WT bNAb^+ cells, consistent with
the higher affinity of DEL bNAbs to some rhesus FcγRs^[211]40, and that
difference reached statistical significance at week 8 post-challenge
(p = 0.009) (Fig. [212]4B). Interestingly, bNAb^+ cell levels were
comparable between weeks 2 and 8 post-challenge in each bNAb-treated
group (Fig. [213]4B), in contrast to plasma bNAb levels, which were
much lower at week 8 than week 2 post-challenge (Fig. [214]2A). In a
separate analysis where we segregated cells bound with VRC07-523-LS
from those bound with PGT121 (and the same for the DEL mutants), we
found that the two bNAbs infused into monkeys were always bound to
similar numbers of cells in the LNs (Fig. [215]4C).
Fig. 4. Detection of infused bNAbs in LN sections.
[216]Fig. 4
[217]Open in a new tab
A Representative confocal microscopy staining to detect infused bNAbs
in LN sections from monkeys that were treated with bNAbs early after
SHIV[AD8-EO] challenge. bNAb staining was performed in one LN section
per timepoint per animal. Inset (scale bar: 500 µm) shows CD20 (yellow)
and cellular nucleic acids (JOPRO, blue) in the whole LN section; big
image (scale bar: 50 µm) shows human Ig kappa light chain^+ cells
(pink), human Ig lambda light chain^+ cells (turquoise), and cellular
nucleic acids (blue); and small images (scale bars: 5 µm) show cellular
nucleic acids (blue) with only human Ig kappa light chain^+ cells
(pink) (top left), only human Ig lambda light chain^+ cells (turquoise)
(top right), or an overlay of both (bottom). The big image and small
image on the bottom right show pink and turquoise Surface objects
assigned by Imaris for easier visualization of human Ig kappa light
chain^+ and lambda light chain^+ cells. All other images show real
staining signals. Quantification of total WT bNAb^+ and total DEL
bNAb^+ cells (B) or VRC07-523-LS^+, PGT121^+, VRC07-523-LS/DEL^+, and
PGT121/DEL^+ cells (C) in whole LN sections at weeks 2 (circles) and 8
(squares) post-challenge. N = 6 per group per timepoint. Each
p = 0.0087, two-sided p-values. D Percentage of follicular
VRC07-523-LS^+, PGT121^+, VRC07-523-LS/DEL^+, and PGT121/DEL^+ cells in
LN sections at weeks 2 (circles) and 8 (squares) post-challenge. N = 6
per group per timepoint. P = 0.0260, two-sided p-value. Bar graphs show
the mean and individual datapoints (B–D). The Mann-Whitney test was
used to detect significant differences in bNAb^+ cell levels at each
timepoint or between timepoints for each type of bNAb (WT or DEL) (B).
The Mann–Whitney test was also used to detect significant differences
between the levels of VRC07-523-LS^+ and PGT121^+ cells,
VRC07-523-LS/DEL^+ and PGT121/DEL^+ cells, VRC07-523-LS^+ and
VRC07-523-LS/DEL^+ cells, or PGT121^+ and PGT121/DEL^+ cells at each
timepoint, or between timepoints for cells bound to each bNAb (C, D). N
indicates the number of biological replicates (monkeys). Source data
are provided in the Source Data file. kappa, human Ig kappa light
chain; lambda, human Ig lambda light chain.
We were also interested in understanding where bNAbs were located in
the LNs; thus, we quantified bNAb^+ cells in the LN follicular and
extrafollicular areas. While DEL bNAbs bound to more cells in the LNs
(Fig. [218]4A), they tended to localize more in the extrafollicular
areas than the WT bNAbs, which were more evenly distributed between the
follicular and extrafollicular areas at any of the evaluated timepoints
(average percentage of follicular VRC07-523-LS^+ cells and PGT121^+
cells: 44.5% and 44.6% at week 2 post-challenge, respectively, and
40.9% and 47.3% at week 8 post-challenge, respectively; average
percentage of follicular VRC07-523-LS/DEL^+ cells and PGT121/DEL^+
cells: 27.5% and 27.0% at week 2 post-challenge, respectively, and
27.9% and 25.4% at week 8 post-challenge, respectively; p = 0.03 for
follicular levels of PGT121^+ cells vs. PGT121/DEL^+ cells at week 8
post-challenge) (Fig. [219]4D).
Taken together, these data showed that WT and DEL bNAbs infused early
after SHIV[AD8-EO] infection were detected in LNs as early as 2 weeks
after challenge and were maintained throughout week 8. Furthermore, WT
and DEL bNAbs had a distinct localization pattern in the LNs, whereby
WT bNAbs were evenly distributed between follicular and extrafollicular
areas while DEL bNAbs tended to mostly concentrate in the
extrafollicular areas.
DEL bNAb-treated monkeys developed distinct SIV Gag-specific CD8^+ T cell
responses in peripheral blood
The distinct detection patterns of virus and bNAbs between the monkey
groups prompted us to investigate if and how early bNAb infusions after
SHIV[AD8-EO] infection altered the immune cell population dynamics in
peripheral blood and LNs. Major cell populations (T cells, B cells,
monocytes, and NK cells) and subsets of T cells (CD4^+ and CD8^+ T
cells, LN follicular CD8^+ (fCD8^+) T cells, and germinal center (GC) T
follicular helper (Tfh) cells) were assessed by flow cytometry
(Supplementary Fig. [220]5A). Both in peripheral blood and LNs, the
levels of all major cell populations remained relatively stable over
time, with no apparent differences across monkey groups (Supplementary
Fig. [221]5B). However, as expected, CD4^+ T cells decayed (and CD8^+ T
cells reciprocally increased) later in WT bNAb-treated monkeys than in
untreated and DEL bNAb-treated monkeys (Supplementary Fig. [222]5C),
consistent with the longer time it took for plasma viremia to emerge in
the former animals (Fig. [223]1C). Finally, although levels of fCD8^+ T
cells and GC Tfh cells appeared to increase in both WT and DEL
bNAb-treated monkeys at week 20 post-challenge, they were not
significantly different from untreated monkeys (Supplementary
Fig. [224]5D). Taken together, these data indicate that SHIV[AD8-EO]
infection followed by early bNAb treatment affected the dynamics of
CD4^+ T cells but not of other immune cell subsets in either peripheral
blood or LNs.
Next, we evaluated the effect of bNAb therapy initiated during acute
SHIV[AD8-EO] infection on virus-specific CD8^+ T cell responses. These
responses were evaluated by flow cytometry after 6-hour in vitro
stimulation of PBMCs and LN cells with SIV Gag peptide pool and defined
as CD8^+ T cells expressing CD107a as a marker of degranulation or
co-expressing the early activation marker CD69 with IFNγ, MIP-1β, or
TNF (Supplementary Fig. [225]6A). In peripheral blood, SIV Gag-specific
CD8^+ T cell responses were initially detected in most of the untreated
monkeys at week 4 post-challenge, followed by DEL bNAb-treated monkeys
at week 6, and finally WT bNAb-treated monkeys at week 16 (Fig. [226]5A
left). Thus, during the first 8 weeks post-challenge, CD8^+ T cell
responses in WT bNAb-treated monkeys were significantly lower than in
untreated monkeys (p = 0.05, p = 0.05, and p = 0.005 at weeks 4, 6, and
8 post-challenge, respectively) and DEL bNAb-treated monkeys (p = 0.05
and p = 0.04 at weeks 6 and 8 post-challenge, respectively)
(Fig. [227]5A left). A similar pattern was observed in LNs, where most
of the SIV Gag-specific CD8^+ T cell responses at week 8 post-challenge
came from untreated and DEL bNAb-treated monkeys, followed by WT
bNAb-treated monkeys at week 20 post-challenge (Fig. [228]5B right).
This detection pattern of virus-specific CD8^+ T cell responses in
peripheral blood and LNs was consistent with the detection of virus in
the plasma and LNs of these animals (Figs. [229]1C and [230]3B),
showing that virus-specific CD8^+ T cell responses tracked with viral
antigen levels. In LNs, SIV Gag-specific fCD8^+ T cell responses were
still very low in DEL bNAb-treated monkeys at week 8 post-challenge
(p = 0.04 for untreated vs. DEL bNAb-treated monkeys) and were mostly
detected in both DEL and WT bNAb-treated monkeys at week 20
post-challenge (Supplementary Fig. [231]6B). Within each monkey group,
the magnitude of SIV Gag-specific CD8^+ T cell responses was comparable
between peripheral blood and LNs, except for WT bNAb-treated monkeys,
whose responses in LNs were significantly higher than in peripheral
blood at week 20 post-challenge (p = 0.003) (Supplementary
Fig. [232]6C). In addition, the magnitude of the responses did not vary
significantly among monkey groups at week 20 post-challenge when
responses were measurable in all three monkey groups (Fig. [233]5A).
Fig. 5. SIV Gag-specific CD8^+ T cell responses.
[234]Fig. 5
[235]Open in a new tab
A Frequency of SIV Gag-specific CD8^+ T cells in PBMCs (left) and LN
cells (right) from monkeys that were left untreated or were treated
with bNAbs early after SHIV[AD8-EO] challenge. N = 9 untreated monkeys
except for LN cells pre: n = 7, week 2: n = 6, and week 20: n = 4.
N = 8 and 7 WT and DEL bNAb-treated monkeys, respectively. P = 0.0474,
0.0480, and 0.0053 for untreated vs. WT group at week 4, 6, and 8,
respectively, and p = 0.0485 and 0.0408 for WT vs. DEL group at weeks 6
and 8, respectively (left) (B, C) Polyfunctional profile of SIV
Gag-specific CD8^+ T cell responses in PBMCs (top) and LN cells
(bottom) at week 20 post-challenge. N = 9 and 4 untreated monkeys for
PBMCs and LN cells, respectively; n = 8 and 7 WT and DEL bNAb-treated
monkeys, respectively. B P = 0.04 for WT vs. DEL group in PBMCs;
p < 0.001 for PBMCs vs. LN cells in untreated and DEL groups, and
p = 0.002 for PBMCs vs. LN cells in WT group; two-sided p-values. C
PBMCs, IFNγ^+MIP-1β^+TNF^-CD107a^- CD8^+ T cells: p = 0.0099 and 0.0014
for untreated vs. DEL and WT vs. DEL group, respectively. PBMCs,
IFNγ^+MIP-1β^-TNF^-CD107a^- CD8^+ T cells: p = 0.0209 and 0.0240 for
untreated vs. DEL and WT vs. DEL group, respectively. LN cells,
IFNγ^-MIP-1β^+TNF^+CD107a^+ CD8^+ T cells: p = 0.0333 for untreated vs.
WT group. Graphs show the mean (A–C) and individual datapoints. The
Kruskal–Wallis test followed by Dunn’s multiple comparison test was
used to detect significant differences between monkey groups at each
timepoint (A) and for each functional readout combination (C), and
between timepoints for each monkey group (A, colored statistics).
Statistics in gray, orange, and green, denote, respectively,
differences from week 2 for untreated monkeys, week 4 for WT
bNAb-treated monkeys, and pre-challenge for DEL bNAb-treated monkeys in
PBMCs (A, left), and from week 20 in LN cells (A, right). The
permutation test was used to compare pie charts (B). N indicates the
number of biological replicates (monkeys). Source data are provided in
the Source Data file. LN lymph node, PBMCs peripheral blood mononuclear
cells, pre pre-challenge, Tx treatment, UnTx untreated, wk week.
To investigate whether the quality of the SIV Gag-specific CD8^+ T cell
responses at week 20 post-challenge was affected by early bNAb
treatment, we conduced polyfunctionality analysis using the Simple
Presentation of Incredibly Complex Evaluations (SPICE) software
package. In peripheral blood, the polyfunctional profile of those
responses significantly differed between WT and DEL bNAb-treated
monkeys (p = 0.04 for pie chart comparison) (Fig. [236]5B). DEL
bNAb-treated monkeys had significantly higher levels of CD8^+ CD69^+ T
cells producing IFNγ alone than WT bNAb-treated and untreated monkeys
(p = 0.02 for both WT vs. DEL bNAb treatment and untreated vs. DEL bNAb
treatment) (Fig. [237]5C top). Although DEL bNAb-treated monkeys also
had higher levels of IFNγ/MIP-1β double-producing CD8^+ CD69^+ T cells
(p = 0.001 for WT vs. DEL bNAb treatment and p = 0.01 for untreated vs.
DEL bNAb treatment) (Fig. [238]5C top), the levels were overall too low
and likely negligible. In contrast, during and shortly after the DEL
bNAb infusions, plasma levels of MIP-1β as measured by Luminex were
noteworthy and higher when compared to the other monkey groups
(p = 0.03 and p = 0.05 at day 17 post-challenge and p = 0.007 and
p = 0.04 at week 3 post-challenge for untreated vs. DEL bNAb treatment
and WT vs. DEL bNAb treatment, respectively) (Supplementary
Fig. [239]6D, E). The polyfunctional profile of the LN responses was
comparable among the monkey groups (Fig. [240]5B). While a very small
population, WT bNAb-treated monkeys had significantly higher levels of
MIP-1β/TNF/CD107a triple-expressing LN CD8^+ CD69^+ T cells when
compared to untreated monkeys (p = 0.03) (Fig. [241]5C bottom). Within
each monkey group, SIV Gag-specific CD8^+ T cell responses
significantly differed between peripheral blood and LNs, in that
degranulation alone accounted for the vast majority of the CD8^+ T cell
responses in LNs (p < 0.001, p = 0.002, and p < 0.001, for pie chart
comparisons of PBMCs vs. LN cells in untreated, WT bNAb-treated, and
DEL bNAb-treated groups, respectively) (Fig. [242]5B).
In conclusion, these data indicate that although early bNAb
administration delayed the development of virus-specific cellular
immunity, likely by delaying exposure to viral antigens, circulating
virus-specific CD8^+ T cell responses developing upon DEL bNAb
treatment had a distinct profile when compared to those developing upon
WT bNAb treatment or in the absence of treatment.
LN immune cells from DEL bNAb-treated monkeys had distinct transcriptomic
profiles at week 8 post-challenge
Given the presence of infused bNAbs in LNs at weeks 2 and 8
post-challenge (Fig. [243]4), we next investigated the immunological
effects of bNAbs binding to LN cells through their Fc domain. Rhesus
macaque monocytes and NK cells express FcγRs^[244]48, so we
hypothesized that the bNAbs used in this study could bind to these cell
types and alter their transcriptomic profiles. To test this hypothesis,
we incubated LN cells from naïve NHPs with each bNAb and assessed
binding to monocytes, NK cells, and T cells by flow cytometry. Like in
the confocal microscopy experiments (Fig. [245]4), we used an
anti-human immunoglobulin (Ig) kappa light chain antibody to recognize
VRC07-523-LS and VRC07-523-LS/DEL and an anti-human Ig lambda light
chain antibody to recognize PGT121 and PGT121/DEL. All tested bNAbs
bound to LN monocytes and NK cells, and only minimally to T cells
(Supplementary Fig. [246]7). In addition, more monocytes were bound to
the DEL bNAbs than to their WT counterparts (Supplementary
Fig. [247]7), consistent with the higher affinity of DEL bNAbs to
rhesus FcγRII and FcγRIII^[248]40.
We then FACS-sorted LN monocytes and NK cells (Supplementary
Fig. [249]8A) as well as different subsets of CD4^+ and CD8^+ T cells
(Supplementary Fig. [250]8B) from untreated and bNAb-treated monkeys
(both infected and uninfected) before challenge and 2 and 8 weeks after
challenge and conducted transcriptomic analysis using bulk RNA
sequencing. T cell sorts were performed on samples from the second
batch of monkeys (4 monkeys per group) due to scarcity of LN samples
from the first batch of monkeys (6 monkeys per group) (Fig. [251]1).
Gene set enrichment analysis (GSEA) revealed that LN monocytes and NK
cells from treated monkeys significantly upregulated antiviral IFN-α
signaling at week 8 post-challenge, but not earlier in infection, when
compared to the pre-challenge timepoint (Fig. [252]6, A and B). At week
2 post-challenge, monocytes from DEL bNAb-treated monkeys and NK cells
from both WT and DEL bNAb-treated monkeys significantly upregulated TNF
signaling via NF-κB (Fig. [253]6A, B and Supplementary Fig. [254]9A,
B). However, such upregulation only persisted at week 8 in DEL
bNAb-treated monkeys, whereas it was absent from WT bNAb-treated and
untreated monkeys at the same timepoint (Fig. [255]6A, B and
Supplementary Fig. [256]9A, B). Also at week 8 post-challenge,
IL-6/JAK/STAT3 signaling was upregulated exclusively in monocytes from
DEL bNAb-treated monkeys (Fig. [257]6A). Taken together, these data
indicate that at week 8 post-challenge, only DEL bNAb treatment induced
proinflammatory signatures in LN monocytes and NK cells.
Fig. 6. Transcriptomic programs in LN cells from SHIV[AD8-EO]-challenged
monkeys on or off early bNAb therapy.
[258]Fig. 6
[259]Open in a new tab
Heatmaps showing the modulation of pathways in sorted LN monocytes (A),
NK cells (B), and T cell subsets (C) from untreated and bNAb-treated
monkeys at week 2 or 8 post-SHIV[AD8-EO] challenge when compared to
pre-challenge or week 2 post-challenge, as determined by transcriptomic
analysis. Monkeys were either left untreated or were treated at days 3,
10, and 17 post-SHIV[AD8-EO] challenge with either VRC07-523-LS and
PGT121 or VRC07-523-LS/DEL and PGT121/DEL. Pathways are indicated in
rows (database: MSigDB Hallmark) and the comparisons of interest
(monkey group & timepoint 1 – monkey group & timepoint 2) in columns.
Red cells indicate significant enrichment of a pathway among genes
induced in monkey group & timepoint 1 (GSEA nominal p ≤ 0.05 and
NES > 0); blue cells indicate significant enrichment of a pathway among
genes downregulated in monkey group & timepoint 1 (GSEA nominal
p ≤ 0.05 and NES < 0); and gray cells correspond to non-significant
enrichment in monkey group & timepoint 1 (GSEA nominal p > 0.05), when
compared to monkey group & timepoint 2. A two-sided permutation test
was used to calculate the nominal p-value and the Benjamini-Hochberg
procedure was used to calculate the adjusted p-value (A–C). Source data
are provided in the Source Data file. CM central memory, EM effector
memory, fCD8^+ follicular CD8^+, GC germinal center, mem. memory, NES
normalized enrichment score, pre pre-challenge, Tfh T follicular
helper, UnTx untreated, wk week.
Although bNAbs only minimally bound to T cells (Supplementary
Fig. [260]7), we hypothesized that, at week 8 post-challenge, T cells
could be indirectly affected by DEL bNAbs downstream of monocyte
engagement. To assess the effect of DEL bNAbs on LN T cells at week 8
post-challenge without having viremia as a confounding factor, we
compared the transcriptomic programs of T cells from DEL bNAb-treated
monkeys at week 8 post-challenge to those from untreated monkeys at
week 2 post-challenge since their viral loads were comparable. Samples
from DEL bNAb-treated monkeys at week 2 post-challenge were used to
assess the effect of DEL bNAbs longitudinally; however, monkeys were
still aviremic at that timepoint. GSEA on the T cell transcriptomic
dataset of untreated and DEL bNAb-treated monkeys revealed that
antiviral IFN-α signatures were upregulated in all studied T cell types
only when the virus was already detectable in plasma and LNs (i.e., at
week 2 post-challenge in untreated monkeys and at week 8, but not week
2, post-challenge in DEL bNAb-treated monkeys) (Fig. [261]6C).
Interestingly, exclusively in DEL bNAb-treated monkeys at week 8
post-challenge, cell cycle-related genes were significantly upregulated
in most non-naïve CD4^+ and CD8^+ T cell subsets (Fig. [262]6C and
Supplementary Fig. [263]10A), and IL-2/STAT5 signaling and TNF
signaling via NF-κB were significantly downregulated in virtually all T
cell subsets (Fig. [264]6C and Supplementary Fig. [265]10B). These
expression patterns were not only observed when comparing DEL
bNAb-treated monkeys between the timepoint of detectable virus (week 8
post-challenge) and no detectable virus (week 2 post-challenge), but
also when comparing timepoints of detectable virus between DEL
bNAb-treated and untreated monkeys (weeks 8 and 2 post-challenge,
respectively) (Fig. [266]6C). This implicates DEL bNAb treatment, and
not viremia, as the main driver of these expression patterns. In
conclusion, LN immune cells of DEL bNAb-treated monkeys showed distinct
transcriptomic profiles at week 8 post-challenge.
Discussion
In recent years, potent anti-HIV-1 bNAbs have been discovered and
extensively characterized. Not only can bNAbs neutralize the virus
through their Fab domain, but they also have the potential to trigger
various host antiviral effector functions through their Fc domain,
which makes them promising candidates for HIV-1 prevention and
treatment strategies. Although anti-HIV-1 bNAbs with enhanced or
ablated Fc-mediated effector functions have been infused in NHPs during
chronic SHIV infection^[267]39–[268]41, administration of such bNAbs in
the acute phase of SHIV or HIV infection has not been explored. In this
study, we investigated the virological and immunological effects of
therapeutically administering to NHPs the anti-HIV-1 bNAbs VRC07-523-LS
and PGT121 either in their WT conformation or with the DEL mutation,
which increases binding affinity to rhesus FcγRIII and FcγRII and
Fc-mediated effector functions in vitro^[269]40,[270]42. bNAbs were
administered to NHPs very early after SHIV[AD8-EO] infection,
specifically at days 3, 10, and 17 post-challenge. Early triple
administration of WT or DEL bNAbs delayed the emergence of virus in
both plasma and LNs as opposed to untreated monkeys. This delay was
shorter when DEL bNAbs were administered when compared to WT bNAbs,
consistent with the shorter duration of DEL bNAbs in circulation, and
viremia was eventually apparent in all monkey groups. Circulating
virus-specific CD8^+ T cell responses developing thereafter had a
distinct polyfunctional profile in DEL bNAb-treated monkeys,
characterized by higher frequencies of CD8^+ CD69^+ T cells producing
IFNγ alone when compared to WT bNAb-treated and untreated monkeys. In
LNs, the presence of both WT and DEL bNAbs was demonstrable as early as
2 weeks after challenge and was sustained up to week 8, but while WT
bNAbs were similarly distributed between follicular and extrafollicular
areas, DEL bNAbs mostly bound to cells in the extrafollicular areas. In
DEL bNAb-treated monkeys at week 8 post-challenge, LN monocytes and NK
cells upregulated proinflammatory signaling pathways and LN T cells
downregulated NF-κB signaling in response to TNF. Overall, early
administration of bNAbs with increased affinity to FcγRs affected both
innate and adaptive cellular immunity. The use of such bNAbs should be
considered when designing future passive bNAb therapies aimed at
harnessing different components of the immune system and modulating
host immune responses.
To our knowledge, this is the first study where anti-HIV-1 bNAbs with
increased affinity to FcγRs were administered in the acute phase of
SHIV infection and where the presence, localization, and cellular
effects of infused bNAbs were investigated in detail in the LNs of
treated monkeys. The early bNAb therapy strategy followed in this study
consisted of triple administration of a combination of two anti-HIV-1
bNAbs starting at day 3 post-intrarectal SHIV[AD8-EO] challenge and
resulted in delayed emergence of measurable virus in both plasma and
LNs. According to our previous study where we identified the
Fiebig-equivalent stages of SHIV[AD8-EO] infection^[271]49, these
findings mean that plasma and LN viremia can be successfully delayed if
administration of bNAbs starts in the eclipse phase of infection. Our
early bNAb strategy differs from that used by Bolton et al. ^[272]28,
where monkeys received either WT VRC07-523 and PGT121 or WT VRC01 on
day 10 after SHIV[SF162P3] challenge, followed by daily ART initiated
11 days after bNAb infusion. In that study, bNAbs were administered
when viremia was already detectable in plasma, and although they
lowered peak viremia, they did so to a similar extent than daily ART
initiated on day 10 post-SHIV[SF162P3] challenge in another group of
monkeys^[273]28. Moreover, bNAb administration did not prevent viral
rebound post-ART interruption^[274]28. In our study, plasma virus
eventually emerged in treated monkeys as strong ADA responses against
the infused bNAbs developed and plasma bNAb levels declined. Of note,
ADA responses against the DEL bNAbs were stronger than those against
the WT bNAbs. ADA responses occurred in NHPs since bNAbs from a
different species (human) were used and were also reported following
infusion of PGT121 and N6-LS in NHPs^[275]50; however, they have seldom
been detected in human clinical trials that used human
bNAbs^[276]23,[277]51–[278]53. At the time of peak viremia, we found no
evidence of viral escape mutations to the infused bNAbs. This could be
due to the use of two bNAbs targeting distinct sites of HIV-1 Env,
which decreased selective pressure on the viral population, and/or the
limited time that bNAbs were maintained in circulation (5–12 weeks
post-challenge for WT bNAbs and 2–9 weeks post-challenge for DEL
bNAbs).
Detailed investigation of virus-specific CD8^+ T cell responses at week
20 post-challenge, when plasma virus was detectable in all animal
groups and responses were similar in magnitude across groups, revealed
a distinct response profile in peripheral blood of DEL bNAb-treated
monkeys when compared to WT bNAb-treated and untreated monkeys. This
profile was set apart for its significantly higher levels of SIV
Gag-specific CD8^+ CD69^+ T cells producing IFNγ alone. Similar viral
burdens among monkey groups rules out an effect of viremia alone in the
different response patterns detected, and one can hypothesize that DEL
bNAbs had a direct effect on FcγR-expressing antigen-presenting cells
(APCs), which ultimately affected CD8^+ T cell responses. In fact, RNA
sequencing analyses revealed that in LN monocytes, only DEL bNAb
treatment induced proinflammatory signatures characterized by
upregulation of TNF signaling via NF-kB at weeks 2 and 8 post-challenge
and IL-6/JAK/STAT3 signaling at week 8 post-challenge. In the LNs, it
is possible that such proinflammatory signatures detected in monocytes
as well as NK cells boosted ADCP and ADCC activities in vivo, similar
to what has been demonstrated in vitro for antibodies carrying the DEL
mutation^[279]40,[280]42. However, those effector functions, even if
present, were not sufficient to ultimately induce long-term viral
control in DEL bNAb-treated monkeys, as discussed below.
The early bNAb therapy regimen employed in our study was based on an
earlier study published by Nishimura et al. ^[281]29 where
administration of WT 3BNC117 and 10–1074 at days 3, 10, and 17
post-intrarectal or intravenous SHIV[AD8-EO] challenge led to long-term
control of infection in approximately half of the monkeys. Despite many
similarities between that study and ours in the challenge virus,
challenge route, and dose, timing, and route of bNAb infusions, monkeys
that received WT bNAbs in our study did not control virus long-term. It
is possible that the different stock of SHIV[AD8-EO] used for challenge
and the different bNAbs infused (even though they targeted similar
viral epitopes as the ones used by Nishimura et al. ^[282]29) played a
role in the different outcome of these studies.
The long-term viral control achieved by Nishimura et al.^[283]29 was
interpreted as being mediated by CD8^+ T cells since depletion of this
cell type in controller monkeys led to rapid induction of plasma
viremia. Thus, this study supported the hypothesis that bNAbs given
early in infection before natural antibodies arise may form immune
complexes, which, once cross-presented to CD8^+ T cells, may boost
antiviral immune responses^[284]30 and ultimately lead to viral
control. With the inclusion in our study of a group of monkeys infused
with bNAbs carrying the DEL mutation, we aimed to test if increasing
the affinity of bNAbs to FcγRs would potentiate the uptake of immune
complexes into APCs and induction of downstream antiviral effects.
While we demonstrated an effect on APCs and CD8^+ T cells, these
monkeys did not control virus long-term. Virus emerged earlier in DEL
bNAb-treated monkeys than in WT bNAb-treated monkeys, consistent with
the poorer PK profile of DEL bNAbs. This could be due to either the
intrinsic nature of DEL bNAbs or their higher affinity to FcγRs, which
precluded them from being detected in the plasma. Peak plasma levels of
PGT121/DEL were lower (around 2.5-fold) than those of PGT121, and
higher affinity of PGT121/DEL to FcγRs with significantly higher levels
of LN cells bound to PGT121/DEL than to PGT121 may have contributed to
the relatively lower plasma levels of this DEL bNAb. However,
VRC07-523-LS/DEL peak plasma levels were substantially lower (around
7.5-fold) than those of its WT counterpart and unlikely to be only
explained by the higher affinity of VRC07-523-LS/DEL to FcγRs. Despite
the substantially lower plasma levels of VRC07-523-LS/DEL when compared
to its WT counterpart, it is noteworthy that VRC07-523-LS/DEL was
present between weeks 1 and 3 post-challenge at plasma levels that
were, on average, 6- to 13- fold above the IC[80] and IC[50] titers for
in vitro neutralization of SHIV[AD8-EO], respectively. In addition,
when VRC07-523-LS/DEL was administered to SHIV[SF162P3]-chronically
infected monkeys, it led to a decrease in plasma viral loads, despite
also being detected in that study at lower levels than its WT
counterpart^[285]40. In our study, DEL bNAbs did not reduce overall
viral burden, but the response profile of SIV Gag-specific CD8^+ T
cells and the proinflammatory signatures of LN monocytes and NK cells
induced in DEL bNAb-treated monkeys but not in untreated monkeys should
motivate further research on the cellular effects and mechanisms of
action of Fc-modified bNAbs.
Overall, long-term viral control was not achieved in either WT or DEL
bNAb-treated monkeys. Importantly, our data suggest that immune
complexes may not have formed to the extent required to achieve
protection due to a temporal and spatial dissociation between virus and
bNAbs in the LNs. In fact, in both treated groups, bNAbs reached the
LNs before the virus. Both WT and DEL bNAbs were present in the LNs at
weeks 2 and 8 post-challenge, but virus RNA^+ cells and virions were
only detected at week 8 post-challenge and only in the DEL bNAb-treated
group, reflecting the pattern of virus emergence in the plasma. In
addition, when both bNAbs and virus were present in the LNs of DEL
bNAb-treated monkeys, their location differed, in that the virus seemed
to predominate in the follicles while DEL bNAbs were mostly detected in
the extrafollicular areas. Based on these findings, future studies are
required to better understand the traffic and dynamics of infused bNAbs
in LNs. Matching the emergence and localization of virus and bNAbs in
LNs will be crucial to better assess the role of immune complexes in
early bNAb therapy strategies.
It is noteworthy that the DEL mutation ablates binding to the
complement proteins C1q and C3c^[286]40. The fact that no difference in
long-term viral control was observed in monkeys that received WT or DEL
bNAbs after SHIV[AD8-EO] infection may suggest that complement does not
play an important role in the control of infection. Likewise,
complement was found to be redundant in antibody-mediated protection of
rhesus macaques against SHIV[SF162P3]^[287]36. Nonetheless, the role of
complement in viral control upon early bNAb therapy warrants further
focused investigation through the use of bNAb variants that enhance or
abolish binding to complement proteins without affecting binding to
FcγRs.
Therapeutic strategies initiated in the acute phase of SHIV or HIV
infection offer the opportunity to tackle early viral events and
potentially change the course of the disease. In this study, we
initiated bNAb interventions in monkeys during the eclipse phase of
SHIV[AD8-EO] infection and used two anti-HIV-1 bNAbs, VRC07-523-LS and
PGT121, which have already shown to be safe and well tolerated in
clinical trials when administered in their WT
conformation^[288]51,[289]54. Monkeys treated early after infection
with WT or DEL bNAbs experienced a delay in viremia that was
proportional to the duration of the infused bNAbs in plasma. Treatment
with DEL bNAbs exerted distinct effects in innate and adaptive immune
cells when compared to WT bNAb-treated or untreated monkeys. Overall,
our study encourages the design of bNAb strategies that further explore
the effects and mechanisms of action of bNAbs with increased affinity
to FcγRs.
Methods
Study design
All animal procedures and experiments were conducted according to NIH
regulations and standards on the humane care and use of laboratory
animals as well as the Animal Care and Use Committees of the NIH
Vaccine Research Center (VRC) and Bioqual, Inc. (Rockville, MD). Thirty
male and female rhesus macaques (Macaca mulatta) of Indian origin and
aged 2 to 4 years at the time of challenge were housed at Bioqual, Inc.
(Rockville, MD). The animals were inoculated intrarectally with 1000
TCID[50] of SHIV[AD8-EO]^[290]55, as previously described^[291]56. At
days 3, 10, and 17 post-challenge, 20 monkeys were administered
intravenously with a combination of either VRC07-523-LS and PGT121
(n = 10) or VRC07-523-LS/DEL and PGT121/DEL (n = 10), with each bNAb at
10 mg/kg. Ten other monkeys were left untreated. The monkeys did not
express the MHC class I alleles Mamu-A*01, Mamu-B*08, and Mamu-B*17,
which are associated with spontaneous viral control, and were assigned
to each group based on the genotype of their FcγRII and FcγRIII genes
(FCGR2 and FCGR3, respectively) for a balanced FCGR genotype
representation across groups (Supplementary Table [292]8). For the
CD8^+ T cell depletion experiments, some monkeys were intravenously
infused once with 50 mg/kg of the anti-CD8β mAb CD8b255R1 (NIH Nonhuman
Primate Reagent Resource). Peripheral blood samples and LN biopsies
were collected at multiple timepoints after challenge, and plasma,
PBMCs, and LN cells were isolated and stored as previously described in
ref. ^[293]49. Specifically, the following LNs were sampled at each
timepoint: left inguinal LNs before challenge; right inguinal LNs at
week 2 post-challenge; left axillary LNs from WT and DEL bNAb-treated
monkeys and left or right axillary LNs from untreated monkeys at week 8
post-challenge, and right axillary LNs at week 20 post-challenge.
Antibodies for infusions
VRC07-523-LS and PGT121 were generated as previously
described^[294]43,[295]51. The LS^[296]45 and DEL^[297]42 mutations
were introduced in the heavy chain plasmids by site-directed
mutagenesis (GenScript). Antibodies were purified by protein
A-Sepharose affinity chromatography (Cytiva). The anti-CD8β mAb
CD8b255R1 is a rhesus recombinant IgG1 mAb that was engineered and
produced by the Nonhuman Primate Reagent Resource (NIH Nonhuman Primate
Reagent Resource Cat# PR-2557, RRID:AB_2716321). All antibody
preparations had low endotoxin levels (< 0.1 endotoxin unit/mg).
FCGR genotyping
Total RNA was isolated from fresh, whole NHP blood using the QIAamp RNA
Blood Mini Kit (Qiagen) as per the manufacturer’s instructions and then
reverse transcribed using oligo(dT)[20] primers (Invitrogen) and
SuperScript™ III Reverse Transcriptase (Invitrogen) under the following
conditions: 25 °C for 5 min, 50 °C for 1 h, and 70 °C for 15 min. FCGR
complementary DNA (cDNA) was amplified by PCR using the Platinum™ Taq
DNA Polymerase High Fidelity (Invitrogen) and previously described
primer pairs that allow amplification of rhesus macaque FCGR2 cDNA
(CD32.1 (fw)/CD32.2 (rv) as well as CD32.3 (fw)/CD32.8 (rv)) and FCGR3
cDNA (FCG3aF (fw)/FCG3aR (rv)) (Sigma)^[298]57. PCR conditions were as
follows: initial denaturation at 94 °C for 2 min followed by 40 cycles
of denaturation at 94 °C for 15 s, annealing at 56 °C for 30 s, and
extension at 68 °C for 1 min. Final extension occurred at 72 °C for
10 min. The size of amplified FCGR fragments was confirmed by gel
electrophoresis using 2% agarose gels (Embi Tec). PCR products were
sequenced (ACGT, Inc.) and sequencing analysis was performed using
Geneious Prime software v.2020.0.5 (Biomatters Ltd.) based on
previously reported macaque FCGR sequences^[299]58.
Viral load measurements
Plasma viremia was measured as previously described^[300]59, using
well-established qRT-PCR assays that quantify SIV Gag RNA with a
detection limit of 15 (standard assay) or 1 (ultrasensitive assay)
copy/mL. Cell-associated SIV Gag RNA and DNA were quantified as
previously described^[301]60, using snap-frozen dry PBMC pellets
obtained right after isolation of PBMCs from peripheral blood. Data
from this assay were normalized to 10^6 cell equivalents based on
concurrent quantification of the CCR5 genomic sequence.
bNAb PK and ADA analyses
Plasma levels of infused bNAbs were measured by ELISA, as previously
described^[302]40. High-binding, 96-well half-area microplates
(Corning) were coated for 1 h at 37 °C with 2 μg/mL in
phosphate-buffered saline (PBS) of either resurfaced stabilized core 3
(RSC3) or ST09 (manufactured at the VRC, NIAID, NIH), proteins
displaying the CD4-binding site and V3-glycan site^[303]9,[304]61 for
capture of VRC07-523-LS (or its DEL mutant) and PGT121 (or its DEL
mutant), respectively. Plates were then blocked for 1 h at room
temperature (RT) with blocking buffer (Tris buffered saline with 2%
bovine serum albumin (BSA), 5% skim milk, and 0.1% Tween 20) and then
washed in PBS with 0.05% Tween 20. Serial dilutions of heat-inactivated
plasma in blocking buffer were added to the plates and incubated for
1 h at RT. Plates were subsequently washed and incubated for 30 min at
RT with a horseradish peroxidase (HRP)-conjugated anti-human IgG
secondary antibody cross-adsorbed against rhesus IgG (Southern Biotech)
to detect bound bNAbs. After washing, binding was visualized with
SureBlue 3,3′,5,5′-tetramethylbenzidine microwell substrate (SeraCare).
Color reactions were stopped with 1 N sulfuric acid (Fisher Chemical)
and absorbances were measured at 450 nm in a SpectraMax Plus 384
microplate reader equipped with the SoftMax Pro software v.7.1
(Molecular Devices, LLC). Every plate included control plasma samples
from naïve monkeys spiked with known concentrations of purified bNAb
and 4-parameter logistic standard curves of purified bNAb from which
plasma bNAb concentrations were calculated.
ADA responses were measured in a similar ELISA assay, with the
exception that plates were initially coated with 2 μg/mL in PBS of
purified bNAb and the secondary antibody was an HRP-conjugated
anti-monkey IgG with minimal reactivity to human antibodies (Southern
Biotech). Data were reported as endpoint titers interpolated from
sigmoidal, 4-parameter logistic sample dilution curves at a
predetermined cut point of five times the average absorbance of the
blank wells. Every plate included as positive controls standard curves
of 5C9 or 9E9 (manufactured at the VRC, NIAID, NIH), anti-idiotype mAbs
that recognize CD4-binding site-specific and V3-glycan-specific bNAbs,
respectively.
Full-length SHIV[AD8-EO] Env deep sequencing
Sequencing
RNA was automatically extracted from virions in plasma using RNAdvance
Viral Reagent Kit (Beckman Coulter) and epMotion® 5073t (Eppendorf),
and immediately reverse transcribed as previously described^[305]46.
Reverse transcription was conducted using the SuperScript IV Reverse
Transcriptase (ThermoFisher Scientific) and a reverse transcription
primer (CCCGCGTGGCCTCCTGAATTATNNNNNNNNTRTAATAAATCCCTTCCAGTCCCCCC) under
the following conditions: 50 °C for 10 min, 85 °C for 10 min, and 4 °C
hold. The cDNA was treated with proteinase K (Sigma-Aldrich) at 55 °C
for 25 min with shaking at 1000 rpm for residual protein removal. The
cDNA was then purified with a 2.2:1 volumetric ratio of RNAClean XP
solid phase reverse immobilization beads (Beckman Coulter). Copy
numbers of resulting cDNAs were determined by limiting dilution PCR
using fluorescence-assisted clonal amplification^[306]62. Purified cDNA
molecules were amplified by PCR using the Advantage 2 PCR kit (Takara
Bio), forward primer (GAGCAGAAGACAGTGGCAATGA), and reverse primer
(CCCGCGTGGCCTCCTGAATTAT), with each primer at a final concentration of
400 nM. PCR conditions were as follows: initial denaturation at 95 °C
for 1 min, 31 cycles of denaturation at 95 °C for 10 s, annealing at
64 °C for 30 s, and extension at 68 °C for 7 min, and final extension
at 68 °C for 10 min. Amplified DNA products of 3 kb (2.5 kb Env gene)
were incorporated into sequencing libraries using the SMRTbell Express
Template Prep Kit 2.0 (Pacific Biosciences) and Barcoded Overhang
Adapter kit 8A and 8B (Pacific Biosciences) to sequence multiple
samples. The libraries were treated by primer annealing and polymerase
binding using the Sequel II Binding Kit 2.0 and Internal Control 1.0
(Pacific Biosciences), and then sequenced by a Sequel II system
(Pacific Biosciences) with a 20 h movie time under circular consensus
sequencing (CCS) mode.
Data analysis
For initial data processing, CCS were generated from SMRT sequencing
data with minimum predicted accuracy of 0.99 and minimum number of
passes of 3 in Pacific Biosciences SMRT Link (v11.0.0.146107)^[307]63.
CCS reads were then demultiplexed using Pacific Biosciences barcode
demultiplexer (lima) to identify barcode sequences. The resulting FASTA
files were reoriented into 5’-3’ direction using the vsearch -orient
command in vsearch (v2.21.1). Cutadapt (v4.1) was used to trim forward
and reverse primers. Length filtering was performed to remove reads
shorter than 2800 nt or longer than 4000 nt. Appropriately sized reads
were then binned by their 8-base UMI sequences. UMI bins with 10 or
more CCS reads were kept for preliminary analysis. For each bin, reads
were clustered with vsearch -cluster_fast based on 99% sequence
identity. Only bins that yielded a single, predominant cluster (i.e.,
where the largest cluster was 1) comprised by at least half of the
bin’s reads and 2) at least twice as large as the second largest
cluster) with at least 10 CCS reads were kept. The cluster consensus
sequence generated by the vsearch -cluster_fast was then used as a
reference to map the cluster’s reads with minimap2 (v2.24). The
commands bcftools mpileup -X pacbio-ccs and bcftools consensus were
used to determine the final consensus sequences for each bin. Consensus
sequences were filtered by searching the BLAST nt database, and
non-SHIV sequences thus identified were discarded.
For final single-genome sequence (SGS) determination, putative fake UMI
bins (bins that arise due to PCR and/or sequencing errors) were
identified and removed with a network approach as previously described
in ref. ^[308]46. Given two distinct bins a and b with read counts n[a]
and n[b], respectively (assume n[a] ≥ n[b]), a and b are connected by
an edge if they have edit distance 1 and satisfy the following count
criterion: n[a] ≥ 2n[b]–1. To resolve the network formed above, we
applied the adjacency method^[309]64, which iteratively consolidates
smaller bins into larger bins that meet the above criteria. Despite
high CCS read accuracy, some errors may persist, particularly those
that may arise during conversion of RNA to cDNA (e.g., reverse
transcription errors). As such, variant calling was performed using a
model describing technical error rates. Given a reverse transcription
error rate R = 1 × 10^-4, a target insert length L, and a number of
recovered SGS sequences N, the probability of observing a technical
variant with at least c occurrences in the sample can be written as
[MATH: PC≥c=1−BinomCDF(
c∣N,R)
L :MATH]
. To determine a cutoff for variant calling, the smallest value c[v]
for which
[MATH: PC≥c<0.01 :MATH]
was determined, and the minimum number of occurrences to call a variant
was set as c[v] + 1. Variant calling with this minimum count threshold
was performed with a custom Python script included alongside the HT-SGS
pipeline. Indels were handled separately; at least three occurrences of
the exact same indel were criteria for inclusion as a real variant.
Variants not meeting these criteria were reverted to the consensus
base. To identify virus haplotypes defined by the data, we took the
consensus sequences of all UMI bins and collapsed non-unique sequences.
We considered the unique sequences bearing different combination of
mutations as individual haplotypes.
Neutralization assays
The experimental neutralizing activity of plasma samples was measured
using an established virus neutralization assay that utilizes TZM-bl
cells and a luciferase reporter gene, as previously described in
ref.^[310]55. Neutralization curves were fitted to a 5-parameter
nonlinear regression analysis, and the neutralizing activity was
reported as the reciprocal plasma dilution required to inhibit
infection by 50% (50% inhibitory dilution, ID[50]) or 80% (80%
inhibitory dilution, ID[80]). The neutralization potency of bNAbs was
measured by the same assay and reported as the bNAb concentration
required to inhibit infection by 50% (50% inhibitory concentration,
IC[50]) or 80% (80% inhibitory concentration, IC[80]). The predicted
ID[50] or ID[80] values of plasma samples were a sum of the predicted
ID[50] or ID[80] values for each of the two bNAbs that the monkeys
received. These values were obtained by dividing the plasma bNAb
concentration at each timepoint by the IC[50] or IC[80] value of the
bNAb against each virus.
To summarize the experimental neutralizing activity of plasma samples
at weeks 64 and 116 post-challenge against all 14 tested viruses, a
neutralization score was calculated for each monkey at each timepoint,
as previously described in ref.^[311]65. One point was assigned for
experimental ID[50] or ID[80] values between 40 and 99, two points for
values between 100 and 999, and three points for values equal or higher
than 1000 for each virus in the multiclade panel. These points were
then added together for an overall neutralization score.
Imaging
For imaging studies, LN biopsies had residual fat removed and were
fixed overnight at RT in PBS with 4% paraformaldehyde (PFA). LNs were
then embedded in paraffin blocks and 5 μm-thick tissue sections were
mounted on Superfrost slides (ThermoFisher Scientific).
To detect SHIV[AD8-EO] RNA in LN sections, RNAscope™ in situ
hybridization (Advanced Cell Diagnostics) was performed as previously
described^[312]49. Following imaging, each SHIV[AD8-EO] RNA^+ cell and
virion were identified and marked in the LN sections using the Imaris’
built-in function Spots. This function first identifies fluorescent
signals based on a user-specified diameter and then marks them with a
large or small green sphere to facilitate visualization of SHIV[AD8-EO]
RNA^+ cells and virions, respectively. SHIV[AD8-EO] RNA^+ cells were
identified based on RNAscope signals measuring 7 μm in diameter. This
diameter was based on previous literature^[313]66,[314]67 and the
average diameter of SHIV[AD8-EO] RNA^+ cells randomly selected from our
previous studies. Each LN section was imaged through 5 scanning steps
(5 Z-stacks) that allowed us to scan through the tissue section
thickness at 5 different focal planes. To accept a 7 μm RNAscope signal
as a SHIV[AD8-EO] RNA^+ cell, it had to be present on at least 3 focal
planes and overlay with nuclear staining. These criteria were manually
checked for each individual 7 μm RNAscope signal. SHIV[AD8-EO] RNA^+
virions were identified based on RNAscope signals measuring ≤ 2 μm in
diameter. Given the diameter of HIV viral particles is in the 120 nm
range^[315]68, 2 μm was the lowest threshold in the Imaris’ Spots
function that allowed us to accurately capture all RNAscope signals
measuring up to 2 μm. To accept each of these RNAscope signals as a
SHIV[AD8-EO] RNA^+ virion, it had to be present on at least 3 focal
planes. For both large and small RNAscope signals described above,
additional background artifacts were manually identified and excluded
when the RNAscope signal overlapped with highly saturated nuclear
staining fluorescence.
To detect infused bNAbs in LN sections, multiplexed confocal imaging
was performed using antibodies that recognize either the kappa or
lambda light chain of human antibodies, given that VRC07-523-LS and
VRC07-523-LS/DEL express kappa light chains^[316]9,[317]43, and PGT121
and PGT121/DEL express lambda light chains^[318]47. Slides were baked
for 1 h at 60 °C, deparaffinized twice in xylene (2 min each), and then
rehydrated in decreasing concentrations of ethanol (100%, 95%, 80%, and
0%; 2 min each). Antigen retrieval was performed in Borg Decloaker, RTU
(Biocare Medical) for 15 min at 110 °C using a pressure cooker. Slides
were then permeabilized and blocked for 1 h with blocking solution (PBS
with 0.3% Triton X-100 and 1% BSA). Primary unconjugated antibody
against CD20 (clone L26, eBiosciences) was incubated overnight at 4 °C.
After washing steps (PBS 1X three times, 15 min each), slides were
incubated with the secondary antibody goat anti-mouse IgG2a (Brilliant
Violet 421-conjugated, Biolegend) for 2 h at RT. Following washing
steps, slides were blocked with 10% normal mouse/goat serum for 1 h at
RT. Directly conjugated antibodies against human Ig lambda light chain
(Alexa Fluor 488-conjugated, clone RM127, Novus Biologicals) and human
Ig kappa light chain (Alexa Fluor 647-conjugated, clone RM126, Novus
Biologicals) were incubated for 2 h at RT followed by additional wash
steps. Slides were then counterstained for nuclei with JOPRO
(Invitrogen) for 20 min at RT, then mounted with Fluoromount-G
(Southern Biotech) and allowed to cure for at least 30 min at 33 °C.
Images were acquired using a Nikon C2 confocal microscope (40X
objective, 1.40 NA) and processed with Imaris version 9.9.0 (Bitplane).
bNAb^+ cells were quantified using the Surface Wizard tool within
Imaris. Like in the RNAscope analysis described above, 7 μm was
assigned as the cell diameter. For each bNAb, the Surface generation
process required the following steps: 1) the Surface threshold was
adjusted so that while background signal was minimized, the Surface
expanded to cover the fluorescent signal; 2) the Surface filter was set
so that a number of voxels Img=1 filtered out background Surfaces with
a small number of voxels while retaining real Surfaces with a big
number of voxels; 3) manual clean-up of artifacts due to (a) tissue
autofluorescence; (b) oversaturated signal under multiple optical
configurations; (c) signal not associated with nuclear staining; and/or
(d) signal not detected in at least 3 confocal optical slices was
performed. bNAb^+ cells were quantified in whole LN sections as well as
in follicular and extrafollicular areas. To quantify events in the
follicular areas, defined based on CD20 expression, the Shortest
Distance to Surface statistics within the Object-Object tool of Imaris
was used. The number of events in the extrafollicular areas was
obtained by subtracting the number of follicular events from the number
of events in the whole LN section. The number of events was normalized
to the LN areas (total, follicular, or extrafollicular) to account for
differences in tissue size.
In vitro cell assays
For functional assays aimed at measuring SIV Gag-specific CD8^+ T cell
responses, PBMCs and LN cells were thawed and rested for 1 h at 37 °C /
5% CO[2] in complete medium consisting of RPMI-1640 medium supplemented
with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 μg/mL
streptomycin, and 100 U/mL penicillin (all from Gibco). Following
incubation, cells were handled as previously described^[319]49.
Briefly, cells were incubated for 6 h with either 4 μg/mL SIV Gag
peptide pool (NIH AIDS Reagent Program) or DMSO (unstimulated negative
control), along with brefeldin A (Sigma), monensin (GolgiStop, BD
Biosciences), and anti-CD107a antibody (clone H4A3, Biolegend) to
assess degranulation. Cells were then stained with flow cytometry
antibodies to measure their responses. Response levels of SIV
Gag-stimulated cells were background subtracted using the residual
response levels of unstimulated cells.
To assess binding of bNAbs to LN cells in vitro, cells were thawed in
complete medium and incubated for 15 min at RT with 50 μg/mL
VRC07-523-LS, VRC07-523-LS/DEL, PGT121, or PGT121/DEL (in vitro bNAb
coating experiment). Cells incubated with PBS were used as negative
control. Cells were then washed in FACS buffer (PBS with 2% FBS) and
surface stained with flow cytometry antibodies.
Flow cytometry
For immunophenotypic analyses and following the in vitro bNAb coating
experiments, cells were stained as previously described in ref.^[320]49
with Zombie UV amine-reactive viability dye (Biolegend), followed by
primary conjugated antibodies to surface markers for 15 min at RT.
Cells were then washed in FACS buffer and fixed in 1% PFA prior to data
acquisition. For functional analyses, cells were stained with Aqua
amine-reactive viability dye (Invitrogen) for 10 min at RT, washed in
FACS buffer, surface stained with primary conjugated and biotinylated
antibodies for 15 min at RT, washed, incubated with streptavidin
(Brilliant Violet 421-conjugated, Biolegend) for 15 min at RT, and
washed again. Cells were then fixed and permeabilized in
Cytofix/Cytoperm (BD Biosciences) for 30 min at 4 °C, washed in
Perm/Wash buffer (BD Biosciences), and intracellularly stained with
primary conjugated antibodies for 30 min at 4 °C. After washing in
Perm/Wash buffer, cells were resuspended in FACS buffer for data
acquisition. Absolute cell counts using fresh, whole NHP blood were
determined using Trucount tubes (BD Biosciences) according to the
manufacturer’s instructions.
All antibodies are listed in Supplementary Table [321]9 and were used
at predetermined optimal titers. Data were acquired on a modified BD
FACSymphony equipped with 355-, 405-, 488-, 532-, and 628-nm lasers (BD
Biosciences) and analyzed using FlowJo software v.9.9.6 and v.10.8.1
(BD Biosciences). The SPICE software v.6.0 (Mario Roederer, VRC, NIAID,
NIH) was used for polyfunctionality analyses.
Transcriptomics of LN cells
Sorting & sequencing
LN cells were thawed in complete medium and stained for cell sorting.
Briefly, cells were stained with Aqua amine-reactive viability dye for
10 min at RT, washed in FACS buffer, and surface stained with primary
conjugated antibodies (Supplementary Table [322]9) for 15 min at RT.
After washing, cells were resuspended in complete medium and different
cell subsets were sorted into ice-cold FBS using either a modified
FACSAria III or a modified FACSymphony S6 cell sorter equipped with
355-, 405- (FACSAriaIII) or 407- (FACSymphony S6), 488-, 532-, and
628-nm lasers (BD Biosciences). Samples were kept on ice before and
right after the sortings. Cells were then pelleted by centrifugation,
lysed in RNAzol RT (Molecular Research Center, Inc.), and snap-frozen
in dry ice. Cell lysates were stored at −80 °C until further use.
Total RNA was isolated from cell lysates using the RNAzol RT protocol
as per the manufacturer’s instructions. RNA Illumina-ready libraries
were generated using NEBNext Ultra II RNA Prep reagents (New England
BioLabs). Using oligo dT beads (Invitrogen), poly-adenylated RNA was
purified and subsequently fragmented, followed by double-stranded cDNA
synthesis, end repair, and adapter ligation. The ligated DNA was then
barcoded and amplified by a limited cycle PCR and the libraries were
sequenced using a paired-end 150-base protocol on either a HiSeq 4000
or a NovaSeq 6000 (Illumina).
Data analysis
Illumina reads were aligned to the Macaca mulatta genome (Mmul_10 build
102) using the STAR aligner (version 2.7.10a) and counted using HTSeq
(2.0.2). Raw gene counts were then TMM normalized, and edgeR (version
3.38.4) framework was used for differential expression analysis (a
generalized linear model was fit to each gene, with bNAb therapy as the
independent variable and gene expression as the dependent variable).
Genes were ranked based on their signed (1: for genes upregulated and
-1: for genes downregulated) -log[10] p-value of the LR test
implemented in edgeR and submitted to pathway enrichment analysis using
GSEA (version 4.1.0, pre-ranked mode) and MSigDB (version 7.4).
ComplexHeatmap (version 2.12.1) was used for visualization.
Plasma MIP-1β quantification
Plasma levels of MIP-1β were measured in a bead-based multiplex assay
using the Luminex technology (Millipore), according to the
manufacturer’s instructions.
Statistical analyses
Statistical analyses of all but two datasets (the polyfunctionality and
transcriptomic datasets) were conducted using Prism software v.9
(GraphPad Software, LLC.). In Prism, the Mann–Whitney test was used to
compare two groups with unpaired samples; the Kruskal-Wallis test was
used to compare more than two groups with unpaired samples and was
followed by the Dunn’s multiple comparison test when the Kruskal-Wallis
p-value was significant; and correlations were evaluated using the
Spearman’s rank correlation test. To compare polyfunctionality data
depicted in pie charts, the permutation test was used in SPICE.
Two-sided p-values lower than 0.05 were considered significant. For
gene set enrichment analyses performed on the transcriptomic data, a
two-sided permutation test was used followed by a Benjamini-Hochberg
correction to adjust for multiple testing and an adjusted p-value equal
or lower than 0.05 was considered significant.
Reporting summary
Further information on research design is available in the [323]Nature
Portfolio Reporting Summary linked to this article.
Supplementary information
[324]Supplementary Information^ (4.7MB, pdf)
[325]Peer Review File^ (2.7MB, pdf)
[326]Reporting Summary^ (3.8MB, pdf)
Source data
[327]Source data^ (802.9KB, xlsx)
Acknowledgements