9G4 autoreactivity is increased in HIV-infected patients and correlates with HIV broadly neutralizing serum activity.

Abstract

The induction of a broadly neutralizing antibody (BNAb) response against HIV-1 would be a desirable feature of a protective vaccine. Vaccine strategies thus far have failed to elicit broadly neutralizing antibody responses; however a minority of HIV-infected patients do develop circulating BNAbs, from which several potent broadly neutralizing monoclonal antibodies (mAbs) have been isolated. The findings that several BNmAbs exhibit autoreactivity and that autoreactive serum antibodies are observed in some HIV patients have advanced the possibility that enforcement of self-tolerance may contribute to the rarity of BNAbs. To examine the possible breakdown of tolerance in HIV patients, we utilized the 9G4 anti-idiotype antibody system, enabling resolution of both autoreactive VH4-34 gene-expressing B cells and serum antibodies. Compared with healthy controls, HIV patients had significantly elevated 9G4+ serum IgG antibody concentrations and frequencies of 9G4+ B cells, a finding characteristic of systemic lupus erythematosus (SLE) patients, both of which positively correlated with HIV viral load. Compared to the global 9G4-IgD--memory B cell population, the 9G4+IgD--memory fraction in HIV patients was dominated by isotype switched IgG+ B cells, but had a more prominent bias toward "IgM only" memory. HIV envelope reactivity was observed both in the 9G4+ serum antibody and 9G4+ B cell population. 9G4+ IgG serum antibody levels positively correlated (r = 0.403, p = 0.0019) with the serum HIV BNAbs. Interestingly, other serum autoantibodies commonly found in SLE (anti-dsDNA, ANA, anti-CL) did not correlate with serum HIV BNAbs. 9G4-associated autoreactivity is preferentially expanded in chronic HIV infection as compared to other SLE autoreactivities. Therefore, the 9G4 system provides an effective tool to examine autoreactivity in HIV patients. Our results suggest that the development of HIV BNAbs is not merely a consequence of a general breakdown in tolerance, but rather a more intricate expansion of selective autoreactive B cells and antibodies.

Full text

Page 1

9G4 Autoreactivity Is Increased in HIV-Infected Patients
and Correlates with HIV Broadly Neutralizing Serum
Activity
James J. Kobie1*, Danielle C. Alcena2, Bo Zheng1, Peter Bryk1, Jonelle L. Mattiacio2, Matthew Brewer2,
Celia LaBranche3, Faith M. Young4, Stephen Dewhurst2, David C. Montefiori3, Alexander F. Rosenberg1,
Changyong Feng5, Xia Jin6, Michael C. Keefer6, Ignacio Sanz1
1Division of Allergy, Immunology and Rheumatology, University of Rochester Medical Center, Rochester, New York, United States of America, 2Department of
Microbiology and Immunology, University of Rochester Medical Center, Rochester, New York, United States of America, 3Department of Surgery, Laboratory for AIDS
Vaccine Research and Development, Duke University Medical Center, Durham, North Carolina, United States of America, 4J.P. Wilmot Cancer Center, University of
Rochester Medical Center, Rochester, New York, United States of America, 5Department of Biostatistics and Computational Biology, University of Rochester Medical
Center, Rochester, New York, United States of America, 6Division of Infectious Disease, University of Rochester Medical Center, Rochester, New York, United States of
America
Abstract
The induction of a broadly neutralizing antibody (BNAb) response against HIV-1 would be a desirable feature of a protective
vaccine. Vaccine strategies thus far have failed to elicit broadly neutralizing antibody responses; however a minority of HIV-
infected patients do develop circulating BNAbs, from which several potent broadly neutralizing monoclonal antibodies
(mAbs) have been isolated. The findings that several BNmAbs exhibit autoreactivity and that autoreactive serum antibodies
are observed in some HIV patients have advanced the possibility that enforcement of self-tolerance may contribute to the
rarity of BNAbs. To examine the possible breakdown of tolerance in HIV patients, we utilized the 9G4 anti-idiotype antibody
system, enabling resolution of both autoreactive VH4-34 gene-expressing B cells and serum antibodies. Compared with
healthy controls, HIV patients had significantly elevated 9G4+ serum IgG antibody concentrations and frequencies of 9G4+ B
cells, a finding characteristic of systemic lupus erythematosus (SLE) patients, both of which positively correlated with HIV
viral load. Compared to the global 9G42IgD2 memory B cell population, the 9G4+IgD2 memory fraction in HIV patients
was dominated by isotype switched IgG+ B cells, but had a more prominent bias toward ‘‘IgM only’’ memory. HIV envelope
reactivity was observed both in the 9G4+ serum antibody and 9G4+ B cell population. 9G4+ IgG serum antibody levels
positively correlated (r=0.403, p=0.0019) with the serum HIV BNAbs. Interestingly, other serum autoantibodies commonly
found in SLE (anti-dsDNA, ANA, anti-CL) did not correlate with serum HIV BNAbs. 9G4-associated autoreactivity is
preferentially expanded in chronic HIV infection as compared to other SLE autoreactivities. Therefore, the 9G4 system
provides an effective tool to examine autoreactivity in HIV patients. Our results suggest that the development of HIV BNAbs
is not merely a consequence of a general breakdown in tolerance, but rather a more intricate expansion of selective
autoreactive B cells and antibodies.
Citation: Kobie JJ, Alcena DC, Zheng B, Bryk P, Mattiacio JL, et al. (2012) 9G4 Autoreactivity Is Increased in HIV-Infected Patients and Correlates with HIV Broadly
Neutralizing Serum Activity. PLoS ONE 7(4): e35356. doi:10.1371/journal.pone.0035356
Editor: Aftab A. Ansari, Emory University School of Medicine, United States of America
Received November 2, 2011; Accepted March 14, 2012; Published April 18, 2012
Copyright: ? 2012 Kobie et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work supported by National Institutes of Health (NIH) grants R01AI084808, R21AI078459, and R37AI049660 to IS, the University of Rochester
Developmental Center for AIDS Research grant P30 AI078498 (NIH/NIAID), the University of Washington Center for AIDS Research grant 5P30AI027757 (NIH/
NIAID), and the Collaboration for AIDS Vaccine Discovery grant 38619 from the Bill and Melinda Gates Foundation to DM. The funders had no role in study design,
data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have read the journal’s policy and have the following conflicts: co-author Xia Jin is an Academic Editor of PLoS ONE. This
does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials.
* E-mail: james_kobie@urmc.rochester.edu
Introduction
HIV infection is a major global health issue, and there is
a critical need for a protective vaccine. The primary focus for
humoral-mediated protection is the induction of neutralizing
antibodies that recognize the HIV Envelope glycoprotein (Env).
Although antibodies that recognize Env readily develop in HIV-1 -
infected patients and can be induced by vaccination, these
antibodies primarily recognize immunodominant, highly variable
domains [1], consequently conferring little to no protection from
the rapidly evolving virus. A minority of HIV patients develop
serum antibodies that can neutralize a broad range of HIV isolates
[2,3,4]. These broadly neutralizing antibodies (BNAbs) typically
do not arise before three years post-infection [5,6], and their
occurrence correlates with viral load (VL) [2,5,7], suggesting that
long-term antigen-driven evolution of the humoral response may
be required for their development.
The limited incidence of persons producing HIV-reactive
BNAbs in response to infection may in part result from proper
enforcement of immunological tolerance for cross-reactive self-
antigens. A relationship between autoreactive antibody and HIV
BNAb development has been highlighted by several observations.
PLoS ONE | www.plosone.org1April 2012 | Volume 7 | Issue 4 | e35356

Page 2

In HIV patients, anti-CL serum antibodies correlate with
increased HIV neutralization breadth [8], and several HIV
broadly neutralizing monoclonal antibodies, including 2F5,
4E10, and 12A21 have been reported to have reactivity to self-
antigens including dsDNA, insulin, Ro, histones, centromere B,
and CL [9,10,11], although this still remains contentious [12,13].
Additionally, many patients with connective tissue autoimmune
disorders, including SLE and anti-phospholipid syndrome (APS),
exhibit limited HIV neutralizing activity [14,15]. Thus, during
normal B cell development, a proportion of B cells with the
potential to give rise to HIV BNAbs may be deleted or rendered
anergic by engagement of corresponding self-antigen, and thus
their development into mature B cells and antibody-secreting cells
may require self-tolerance to be subverted. However, during HIV
infection, substantial B cell hyperactivation manifested by poly-
clonal B cell activation and hypergammaglobulinemia [16], may
contribute to disruption of tolerance, leading to the development
of autoreactive antibodies in HIV patients, including those HIV
BNAbs with autoreactivity. In addition to increases in autoanti-
bodies including anti-CL, anti-dsDNA, anti-nuclear antibodies
(ANA) and others in HIV patients, dramatic alterations in B cell
homeostasis are reflected by the expansion of immature/
transitional B cells, exhausted tissue like-memory B cells [17]
and plasmablasts [18], and decreased resting memory and IgM
memory [16,19]. Many of these serological and cellular alterations
are reversed with anti-retroviral therapy [16], suggesting they
result from ongoing HIV viral replication.
Our group and others previously described an approach to
monitor the development of autoreactive B cells and antibodies in
SLE using the 9G4 anti-idiotype antibody [20,21,22,23]. The rat
anti-human monoclonal antibody 9G4 recognizes VH4-34 (pre-
viously designated VH4-21) -encoded antibodies and the B cells
expressing these antibodies as surface receptor (heretofore referred
to as 9G4+ antibodies and 9G4+ B cells respectively) [24]. Our
laboratory has extensively studied the 9G4 system to understand
human B cell tolerance and its breakdown in SLE [20,25]. The
advantage of the 9G4 system is predicated on a number of unique
and helpful features: 1) 9G4+ antibodies are intrinsically
autoreactive against N-acetylactosamine moietes expressed by
blood group antigens as well as by multiple tissue antigens and
glycoproteins [26,27,28,29], including B220/CD45R which is
expressed by many naı ¨ve B cells [26]. This reactivity is dependent
on the heavy chain framework 1-encoded idiotope recognized by
the 9G4 monoclonal antibody [24,30]. Many 9G4+ antibodies
also cross-react with glycolipids including LPS and gangliosides
and in association with favorable HCDR3s (heavy chain
complementary determining region 3), they may also recognize
DNA [24,31,32]; 2) 9G4+ B cells, despite their autoreactivity,
represent a sizable fraction (5–10%) of the normal repertoire in all
healthy subjects; yet, they are strictly censored from the germinal
centers (GC) and from expanding into the long-lived IgG memory
and plasma cell compartments [33]; 3) such censoring accounts for
the almost complete absence of serum 9G4+ antibodies in healthy
subjects [20,34]; 4) by contrast, GC censoring of 9G4+ B cells is
faulty in SLE and results in large expansions of 9G4+ IgG memory
and plasma cells in these patients [20,34].
In this study, we utilized the 9G4 system to examine both
autoreactive 9G4+ serum antibodies and B cells in HIV patients,
and their relationship with clinical characteristics, HIV reactivity,
and the occurrence of HIV BNAbs.
Materials and Methods
Clinical Samples
Peripheral blood samples were obtained from HIV-1 -infected
patients at the University of Rochester Medical Center and the
University of Washington HIV clinics between 2004 and 2010.
Samples from healthy control (HC) subjects were obtained at the
University of Rochester. All subjects provided signed written
informed consent. Isolated peripheral blood mononuclear cells
and serum were cryopreserved before subsequent analysis. All
procedures and methods were approved by the University of
Rochester Research Subjects Review Board and the University of
Washington Institutional Review Board.
Flow Cytometry
For global B cell phenotypic analysis peripheral blood mono-
nuclear cells (PBMCs) were stained with anti-CD19-APC-Cy7
(SJ25C1, BD), anti-CD20-AlexaFluor 700 (2H7, Biolegend, San
Diego, CA), anti-CD3-PerCP-Cy5.5 (SP34-2, BD), anti-IgD-FITC
(IA6-2, BD), anti-IgG-PE (G18-145, BD), anti-IgM-PE-Cy5 (G20-
127, BD), anti-CD27-Qdot655 (CLB-27/1, Invitrogen), 9G4-
Pacific Blue, and Live/Dead fixable aqua dead cell stain
(Invitrogen). The 9G4 mAb was kindly provided by F.K.
Stevenson (University of Southhampton, Southhampton, United
Kingdom), and recognizes a framework 1 region–encoded idiotype
that is expressed by all unmutated, and close to 90% of mutated
VH4-34 B cells present in the normal repertoire [35–36]. 9G4
painting of B cells was assessed using methods similar to those
previously described [21,26]. Briefly, cells were incubated at either
37uC or 4uC for 30 minutes in complete media (RPMI 1640+10%
fetal bovine serum), and washed once with PBS, prior to staining
with fluorochrome-conjugated antibodies at 4uC for 60 minutes.
One-to-two million total events per sample were collected on an
LSRII instrument (BD Biosciences) and analysis performed in
a blinded manner using FlowJo software (Treestar, Inc, Ashland,
OR). Total PBMC were gated on lymphocytes using FSC and
SSC. Live/Dead stain and anti-CD3 were used to exclude dead
cells and T cells, respectively.
To measure gp140-specific B cells, PBMC were first incubated
for 30 minutes at 37uC to reduce 9G4 painting, then stained with
purified oligomeric HIV-1 SF162 (clade B) and KNH1144 (clade
A) gp140 directly conjugated to AlexaFluor660 and AlexaFluor
647, respectively and HIV-1 p24 (NIH AIDS Research and
Reference Reagent Program) directly conjugated to Alexa-
Fluor488, in addition to anti-CD19-PE-Cy7, anti-CD20-APC-
Cy7, anti-IgD-PE, anti-IgM-PerCP-Cy5.5, anti-CD3-PE-Cy5,
anti-CD14-PE-Cy5, 7AAD for dead cell exclusion, and biotiny-
lated 9G4 mAb/streptavidin Qdot800 (Invitrogen) at 4uC for
60 minutes. At least one million total CD19+ B cells (<10 million
total PBMC) were analyzed per sample. p24+ gp140+ double-
positive cells, a very rare population, were excluded from the
analysis because they represent B cells that non-specifically react
with gp140.
ELISA
The detection of IgG 9G4+ Abs was performed as previously
described [26]. Briefly, ELISA plates were coated with anti-human
IgG (Jackson ImmunoResearch, West Grove, PA), blocked with
2% nonfat dry milk/2% BSA for 1 h at 37uC, then washed with
0.1% Tween 20 in PBS. Samples were serially diluted in triplicate
in PBS containing 0.01% Tween 20 and 0.5% BSA, and
incubated for 90 minutes at 37uC. Plates were washed, and
binding was detected using biotinylated 9G4 mAb and streptavidin
alkaline phosphatase. A human VH4-34 9G4-reactive mAb,
9G4 Antibodies Correlate with HIV Neutralization
PLoS ONE | www.plosone.org2 April 2012 | Volume 7 | Issue 4 | e35356

Page 3

75D9, was used as a standard. For detection of gp140-reactive IgG
and 9G4+ Abs, ELISA plates were coated with purified oligomeric
HIV-1 YU2 gp140 [37] at 2 mg/ml and detected with either
horseradish peroxidase conjugated anti-human IgG (Jackson
ImmunoResearch) or biotinylated 9G4 Ab and streptavidin-
HRP. To detect anti-CL IgG Abs, plates were coated with
50 mg/ml bovine cardiolipin (Sigma) in hexanes, and detected
with anti-IgG HRP. For quantitative purposes the polyclonal
human anti-human CL Ab (Lifespan Biosciences, Seattle, WA)
was used at 1:100 dilution, and assigned a relative units value of
1000, for which all serum samples were normalized to; positive/
negative cutoff was determined as the upper 95% CI for serum
from healthy control subjects. Anti-dsDNA IgG ELISA and Anti-
Nuclear Antigen (ANA) IgG ELISA were performed, testing
samples in duplicate using kits according to manufacturer’s (Inova
Diagnostics, Sand Diego, CA) recommendations.
HIV-1 Neutralizing Activity
The neutralizing activity of sera was determined as previously
described [38]. Briefly, a panel of clade B Tier 1 (SF162) and Tier
2 (6353.3, QH0692.42, SC422661.8, PVO.4, and AC10.0.29)
pseudotyped viruses were incubated with heat-inactivated sera and
then added to TZM-bl indicator cells. SVA-MLV (murine
retrovirus Env-pseudovirus) was included as a negative control to
confirm the absence of anti-retroviral therapy (ART). Neutralizing
antibody titers were determined as the reciprocal serum dilution at
which luciferase expression was reduced by 50% (ID50). To
enable correlative analysis a single, composite BNAb score was
determined using the geometric mean ID50 of the serum Tier 2
neutralizing activity, similar to that previously described [2], and if
the ID50 was ,20 for an individual virus, the value for that virus
was set to 1 to enable calculation.
Statistical Analysis
Two-tailed t test or Mann-Whitney test were used as
appropriate to compare HC and HIV groups. Spearman two-
tailed correlation co-efficient was used to measure the correlation
of two variables. Statistical analyses were performed using Prism
5.0 software (GraphPad Software, La Jolla, CA) and significance
was taken as p,0.05.
Results
HIV-infected patient population
Peripheral blood lymphocytes and serum samples were obtained
from a diverse group of 90 HIV-1 seropositive patients from the
University of Rochester and the University of Washington Center
for AIDS Research (CFAR) cohorts. Most HIV patients (84%) had
no history of anti-retroviral treatment (ART naı ¨ve) and the
remainder were not receiving ART at time of sampling as per
standard of care (Table 1). Samples from HIV-1 -negative healthy
control (HC) subjects were obtained at the University of
Rochester.
HIV patients have elevated 9G4+ serum antibodies and
9G4+ B cells
Increased autoreactive antibodies are frequently observed in
HIV patients, and therefore we asked if this breakdown in
tolerance could be monitored using the 9G4 system as we and
others have previously used for SLE [21,26,33,39]. Indeed, HIV
patients exhibited significant (p,0.01) and approximately 5-fold
greater 9G4+ IgG serum antibody compared with HC subjects
(Fig. 1A), with approximately 40% of HIV patients having 9G4+
IgG serum antibody levels exceeding the 95% CI of HC (not
shown). HIV patients also exhibited a significant (p,0.05) 4-fold
increase in the frequency of 9G4+ total peripheral blood B cells
(Fig. 1B), with approximately 65% of HIV patients exceeding the
95% CI of HC (not shown). These results indicate that increased
9G4+ serum IgG and 9G4+ B cells are commonly found in many,
but not all HIV patients, and are consistent with elevated
autoreactive antibodies previously observed in HIV patients [8].
9G4+ antibody titers and 9G4+ B cell frequencies
correlate with CD4 and viral load in HIV patients
On average, the HIV patients in our study had elevated 9G4+
IgG serum and 9G4+ B cells; however, a substantial continuum
was observed (Fig. 1). To determine what factors might contribute
to the development of 9G4+ IgG serum antibody and 9G4+ B cells
in these patients, regression analysis was performed. Neither age
nor time since HIV diagnosis correlated with 9G4+ serum IgG
antibody or 9G4+ B cell frequency (data not shown). CD4+ T cell
count negatively correlated with 9G4+ IgG serum (r=20.202,
p=0.125) and 9G4+ B cells (r=20.473, p,0.0001) (Fig. 1C+ +D),
although only reached significance for 9G4+ B cells. HIV VL was
significantly positively correlated with both 9G4+ IgG serum
antibody (r=0.265, p=0.042) and 9G4+ B cells (r=0.374,
p=0.004) (Fig. 1E+ +F). Interestingly, HIV VL did not correlate
with the abundance of other autoreactive serum antibodies
including anti-dsDNA, ANA, or anti-CL (Fig S1). These results
suggest that HIV viral replication may selectively promote the
development of autoreactive 9G4+ B cells and 9G4+ serum
antibody.
Phenotype of 9G4+ B cells in HIV patients
It is striking that for many HIV patients, greater than 50% of
their total B cell compartment is 9G4+. We previously reported
that in SLE, 9G4+ serum antibodies can bind to B220, which is
expressed on the surface of naı ¨ve B cells and some memory B cells
[26]. This high frequency of 9G4+ B cells in some of the HIV
patients may result from measuring both the B cells actually
expressing VH4-34-encoded BCR, and also the indirect binding of
the 9G4 reagent to serum VH4-34-encoded autoreactive antibody
bound to B220 on the B cell, as previously described in SLE
patients and not HC subjects [21,26] and heretofore referred to as
painting. To determine if 9G4 painting of B cells may be occurring
in HIV patients, samples were incubated at 37uC to dissociate
bound 9G4+ serum antibody prior to 9G4 reagent staining at 4uC.
This approach revealed that similar to SLE [21,26], 9G4 painting
does occur in some HIV patients (Fig. 2A+ +B), and that soluble
Table 1. HIV-1 patient demographics.
HIV patients (n)90range
Age (YR)41.9+/29.922–70
Female (%)38.6
Time since HIV Dx (YR) 6.3+/25.60.1–23.3
ART naı ¨ve (%)84.1
Non-Caucasian (%)51.2
Black (%)48.2
Viral load (copies/ml) 35,500+/2111,629
549+/2237
mean +/2 SD indicated
45–942,000
CD4 (cells/ml)7–1,080
doi:10.1371/journal.pone.0035356.t001
9G4 Antibodies Correlate with HIV Neutralization
PLoS ONE | www.plosone.org3April 2012 | Volume 7 | Issue 4 | e35356

Page 4

autoreactive 9G4+ serum antibody is bound to a substantial
proportion of the total B cells. Interestingly, limited longitudinal
sampling (HIV055 and HIV056) indicated that 9G4 painting may
substantially fluctuate over time thereby suggesting episodic
stimulation of 9G4 antibodies rather than steady production
through long-lived plasma cells or continuous B cell stimulation
(Fig. 2B).
By eliminating 9G4 painting, the composition of the actual
9G4+ B cell population in HIV patients was determined (Fig. 2C).
The 9G4+ B cell population in HIV patients was comprised of
Figure 1. 9G4+ + serum antibodies and 9G4+ + B cells are increased in HIV patients, correlating with CD4 and viral load. Peripheral blood
was collected from ART-negative HIV patients and healthy control (HC) subjects, and serum and PBMC were isolated. A. 9G4+ serum IgG was
determined by ELISA. B. The frequency of 9G4+ B cells was determined by flow cytometry. C–F. Spearmann correlation of 9G4+ serum IgG and 9G4+
B cells with CD4 and VL was determined for HIV patients. Each symbol represents a unique patient. * p,0.05 (Mann Whitney test).
doi:10.1371/journal.pone.0035356.g001
9G4 Antibodies Correlate with HIV Neutralization
PLoS ONE | www.plosone.org4April 2012 | Volume 7 | Issue 4 | e35356

Page 5

9G4 Antibodies Correlate with HIV Neutralization
PLoS ONE | www.plosone.org5April 2012 | Volume 7 | Issue 4 | e35356