A Novel Assay for Antibody-Dependent Cell ... - Journal of Virology

5 downloads 0 Views 2MB Size Report
Jun 27, 2012 - combinant gp120 and a canarypox vector acquired HIV-1 infec- tion at a reduced rate relative to the rate in recipients of placebo immunizations ...
A Novel Assay for Antibody-Dependent Cell-Mediated Cytotoxicity against HIV-1- or SIV-Infected Cells Reveals Incomplete Overlap with Antibodies Measured by Neutralization and Binding Assays Michael D. Alpert,a Lisa N. Heyer,a David E. J. Williams,a Jackson D. Harvey,a Thomas Greenough,b Maria Allhorn,c and David T. Evansa Department of Microbiology and Immunobiology, Harvard Medical School, New England Primate Research Center, Southborough, Massachusetts, USAa; Department of Medicine, University of Massachusetts Medical School, Worcester, Massachusetts, USAb; and Division of Infection Medicine, Department of Clinical Sciences, Lund University, Lund, Swedenc

The resistance of human immunodeficiency virus type 1 (HIV-1) to antibody-mediated immunity often prevents the detection of antibodies that neutralize primary isolates of HIV-1. However, conventional assays for antibody functions other than neutralization are suboptimal. Current methods for measuring the killing of virus-infected cells by antibody-dependent cell-mediated cytotoxicity (ADCC) are limited by the number of natural killer (NK) cells obtainable from individual donors, donor-to-donor variation, and the use of nonphysiological targets. We therefore developed an ADCC assay based on NK cell lines that express human or macaque CD16 and a CD4ⴙ T-cell line that expresses luciferase from a Tat-inducible promoter upon HIV-1 or simian immunodeficiency virus (SIV) infection. NK cells and virus-infected targets are mixed in the presence of serial plasma dilutions, and ADCC is measured as the dose-dependent loss of luciferase activity. Using this approach, ADCC titers were measured in plasma samples from HIV-infected human donors and SIV-infected macaques. For the same plasma samples paired with the same test viruses, this assay was approximately 2 orders of magnitude more sensitive than optimized assays for neutralizing antibodies—frequently allowing the measurement of ADCC in the absence of detectable neutralization. Although ADCC correlated with other measures of Env-specific antibodies, neutralizing and gp120 binding titers did not consistently predict ADCC activity. Hence, this assay affords a sensitive method for measuring antibodies capable of directing ADCC against HIV- or SIV-infected cells expressing native conformations of the viral envelope glycoprotein and reveals incomplete overlap of the antibodies that direct ADCC and those measured in neutralization and binding assays.

T

he inherent resistance of human immunodeficiency virus type 1 (HIV-1) to antibodies has confounded efforts to elicit neutralizing antibodies by vaccination and complicated the detection of antibodies that interfere with virus replication. The masking of antibody epitopes on the viral envelope glycoprotein (Env) enables persistent HIV-1 replication in the face of vigorous Envspecific antibody responses (32, 36, 65, 137, 138). Antibody epitopes in the native Env trimer are occluded by glycosylation (66, 69, 91, 102, 108, 133, 144), oligomerization of the gp120 and gp41 Env subunits (12, 47, 88, 89, 115, 136), the recessed nature of the CD4 binding site (17, 73), the spatial dispersion of the coreceptor binding site prior to CD4 engagement (16, 74, 128, 135), and the thermodynamics of conformational changes associated with receptor binding (72, 92). As a consequence of these features, no vaccine approach under consideration for clinical development has elicited detectable antibodies capable of neutralizing primary isolates of HIV-1 or simian immunodeficiency virus (SIV) that are representative of the circulating HIV-1 isolates confronting these vaccines (10, 15, 24, 25, 41, 68, 80, 86, 95, 103, 110, 114, 118, 127). Antibodies mediate antiviral immunity through numerous functions in addition to neutralization. The constant (Fc) region of IgG interacts with Fc receptors expressed on leukocytes and with complement. These interactions can contribute to antiviral immunity by inactivating and clearing virions (1, 121), orchestrating the homing of effector cells (37, 42, 56, 78, 90, 93, 94, 98, 99, 113, 131), inhibiting virus replication (23, 31, 33, 37, 45, 55, 70, 98, 128), and killing virus-infected cells by complement-dependent

November 2012 Volume 86 Number 22

cytotoxicity (CDC) (120) or by antibody-dependent cell-mediated cytotoxicity (ADCC) (71, 75, 112). These nonneutralizing effector functions may be key components of antiviral immunity (58). It is important to measure the antibodies that bind Env despite the presence of features that confer resistance to antiviral immunity. Enzyme-linked immunoadsorbent assays (ELISAs) are routinely used to sensitively measure antibodies that bind to gp120 monomers or gp140 trimers, but these recombinant forms of Env expose epitopes that are normally occluded in the native, membrane-bound Env trimer that exists on virions and virus-infected cells (12, 15, 26, 34, 47, 48, 54, 73, 88, 89, 100, 111, 115, 116, 136, 137, 141). When neutralization of primary viruses is undetectable, neutralization assays are often performed using T-cell lineadapted viruses, which have lost features that confer resistance to antibodies as an adaptation to chronic propagation on T-cell lines in vitro (11, 30, 83, 87, 105, 134). Therefore, ELISAs using recombinant forms of Env and neutralization assays using T-cell lineadapted viruses measure antibodies that may not belong to the

Received 27 June 2012 Accepted 17 August 2012 Published ahead of print 29 August 2012 Address correspondence to David T. Evans, [email protected]. Copyright © 2012, American Society for Microbiology. All Rights Reserved. doi:10.1128/JVI.01650-12

Journal of Virology

p. 12039 –12052

jvi.asm.org

12039

Alpert et al.

subset that is relevant for immunity against circulating HIV-1 isolates. These drawbacks also apply to current methods for measuring antibodies that direct ADCC. ADCC is typically measured using target cells coated with gp120, gp140, or peptides (10, 19–22, 38– 40, 46, 51–53, 59, 61, 64, 67, 101, 104, 124, 139) or chronically infected T-cell lines (38, 40, 51, 53, 104). ADCC assays based on target cells coated with recombinant forms of Env or chronically infected T-cell lines therefore measure antibodies that may not direct ADCC against cells infected with primary isolates. Practical considerations also place limitations on current methods for measuring antibodies that direct ADCC. These assays rely on natural killer (NK) cells expressing the low-affinity IgG receptor CD16 (Fc␥RIIIA), which are the predominant effectors of ADCC (125, 130). However, the number of NK cells that can be obtained from an individual donor restricts the number of samples that can be processed in parallel. Donor-to-donor variation in the frequency and cytolytic activity of CD16⫹ NK cells and the susceptibility of CD4⫹ T cells to infection can adversely affect the consistency of assays that depend on primary cells (104, 124). Furthermore, assays performed with primary cells can be cumbersome to perform, requiring the repeated isolation of effector and target cells. As a surrogate for measuring ADCC, inhibition of virus replication in the presence of effector cells and antibody can be evaluated by using assays for antibody-dependent cell-mediated viral inhibition (ADCVI) (40, 43–45, 59, 139). However, ADCVI assays have the disadvantages associated with dependence on primary cells and often do not clearly discriminate between neutralizing and nonneutralizing antibody activities. Thus, current assays for the cell-mediated effects of antibodies have a number of drawbacks that limit their utility for measuring ADCC against HIV- or SIV-infected cells. We therefore developed a novel ADCC assay based on immortalized cell lines—an NK cell line that expresses either human or rhesus macaque CD16 and a CD4⫹ target cell line that expresses luciferase from a Tat-inducible promoter upon HIV-1 or SIV infection. The dose-dependent loss of luciferase in the presence of NK cells and serial dilutions of plasma or serum indicates the killing of virus-infected cells by ADCC. Using this assay, we show that although ADCC is correlated with other measures of virusspecific antibodies, there is incomplete overlap with the antibodies measured by neutralization and binding assays. MATERIALS AND METHODS Human and nonhuman primate samples. Human plasma samples were obtained from volunteers who had given written consent and were collected according to protocols approved by the Institutional Review Boards of the United States Army, the Ministry of Public Health of Thailand, the Royal Thai Army Medical Department, and Mahidol University (109, 110). Additional plasma samples were obtained from volunteers who had provided written informed consent for studies of HIV-specific immune responses through protocols approved by the Institutional Review Board of the University of Massachusetts Medical School. All human plasma samples were deidentified. Samples from rhesus macaques (Macaca mulatta) were obtained from animals housed in a biocontainment facility at the New England Primate Research Center (NEPRC) and given care in accordance with standards of the Association for Assessment and Accreditation of Laboratory Animal Care and the Harvard Medical School Animal Care and Use Committee. The animal samples used here were collected under experimental protocols approved by the Harvard Medical Area Standing Committee on Animals for the purpose of other vaccine

12040

jvi.asm.org

studies and conducted in accordance with the Guide for the Care and Use of Laboratory Animals (8). All plasma samples were collected in sodium citrate anticoagulant and heat inactivated for 30 min at 56°C prior to use in assays. Cell lines. The effector cells were derived from the CD16-negative human NK cell line KHYG-1 (Japan Health Sciences Foundation) (140). KHYG-1 cells were transduced with a pQCXIP-derived retroviral vector that expresses human CD16 (V158 variant of FCGR3A) or with a pQCXIN-derived retroviral vector that expresses rhesus macaque CD16 (FCGR3A variant 7; PubMed accession number ABN69102) (84). Transduced KHYG-1 cells were selected by propagation in medium containing puromycin or G418, respectively, and then subjected to limiting-dilution cloning. Individual clones were selected for use in the ADCC assay based on the criteria of CD16 expression, low background cytotoxicity against infected and uninfected target cells in the absence of antibody, and high ADCC activity. The NK cell lines were maintained at a density of 1 ⫻ 105 to 4 ⫻ 105 cells per ml in R10 cell culture medium consisting of RPMI medium (Invitrogen) supplemented with 10% fetal bovine serum (FBS) (Invitrogen), 25 mM HEPES (Invitrogen), 2 mM L-glutamine (Invitrogen), 0.1 mg/ml Primocin (InvivoGen), 1 ␮g/ml cyclosporine (CsA) (Sigma), and interleukin-2 (IL-2) (AIDS Research and Reference Reagent Program). The human CD16⫹ KHYG-1 cells were cultured at 5 U/ml IL-2, whereas the macaque CD16⫹ KHYG-1 cells were cultured at 10 U/ml IL-2. CsA was prepared as a 10,000⫻ stock solution in dimethyl sulfoxide (DMSO) (Sigma) at a concentration of 10 mg/ml. The NK cell lines were cryopreserved in 0.25 to 0.5 ml of recovery freezing medium (Invitrogen) supplemented with 1 ␮g/ml CsA. After cryopreservation, thawed cells were washed twice in 37°C R10 medium supplemented with 1 ␮g/ml CsA. FK-506 (Sigma) at 1 ␮g/ml can be used as an alternative to CsA. Target cells were derived from CEM.NKR-CCR5 CD4⫹ T cells (60, 129) (AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH, contributed by Alexandra Trkola). The CEM.NKR-CCR5 cells were transduced with a pLNSX-derived retroviral vector, which contains the gene for firefly luciferase downstream from the SIV long terminal repeat (LTR) promoter (83). After propagation in the presence of G418, a limiting-dilution clone was selected based on the criteria of supporting rapid virus replication and high luciferase expression in SIVmac239-infected cells relative to the baseline level of luciferase expression in uninfected cells. Although luciferase expression from the SIV LTR is upregulated in cells infected by HIV-1, CCR5-tropic HIV-1 strains had suboptimal infectivity in this cell line. Infection by CCR5-tropic HIV-1 strains was greatly enhanced by transduction of these cells with a pQCXIP-derived vector that expresses human CCR5. A limiting-dilution clone was selected from CEM.NKR-CCR5 cells transduced with pQCXIPCCR5, which expressed significantly higher levels of CCR5 than the original CEM.NKR-CCR5 clone. Target cell clones were maintained at densities of 5 ⫻ 104 to 5 ⫻ 105 cells per ml in R10 culture medium. ADCC assay. Target cells were infected 4 days prior to each assay. Infections were conducted by spinoculation to overcome the relative resistance of CEM.NKR-CCR5 cells to infection (96). Spinoculation was performed in round-bottom 12 by 75 mm tubes by resuspending 5 ⫻ 105 target cells in medium containing the following amounts of infectious virus, as measured by SIV p27 or HIV-1 p24 antigen capture ELISA: 200 ng p27 of SIVmac239, simian-human immunodeficiency virus (SHIV) chimera SHIVSF162P3, or SHIVKB9, 200 ng p24 of HIV-1NL4-3, 300 ng p24 of HIV-192TH023, or 1,000 ng p24 of HIV-1YU2. Mixtures of target cells and viral inoculum were subjected to centrifugation for 2 h at 1,200 ⫻ g at 25°C. Afterward, the viral inoculum was removed, and the target cells were cultured in R10 medium as described above. A 5- to 20-fold induction of luciferase expression over the background level was generally observed for populations of target cells infected under these conditions. Immediately before the assembly of ADCC assays, infected target cells were washed three times in R10 medium. ADCC assays were performed in round-bottom, tissue culture-treated

Journal of Virology

ADCC against HIV-1- and SIV-Infected Cells

polystyrene 96-well plates, with each well containing 104 target cells and 105 effector cells in a 200-␮l final volume. Assays were performed in R10 culture medium containing 10 U IL-2 per ml, with no CsA. Effector cells were combined with washed target cells immediately before addition to assay plates. The outermost wells of each plate were filled with phosphatebuffered saline (PBS) to minimize differences due to evaporation toward the edges of the plate. The second column of wells on each plate contained NK effector cells and uninfected target cells but no antibody and defined 0% relative light units (RLU). The third column contained NK cells and infected targets but no antibody and defined 100% RLU. Serial, 2-fold, triplicate dilutions of plasma or monoclonal antibody were added to columns 4 through 11 of each assay plate. Once targets, effectors, and serially diluted antibody were combined, assay plates were incubated for 8 h at 37°C and 5% CO2. After an 8-h incubation, a 150-␮l volume of cells was resuspended and mixed by pipette with 50 ␮l of the luciferase substrate reagent BriteLite Plus (Perkin Elmer) in black 96-well plates. Luciferase activity was read approximately 2 min later using a Wallac Victor3 plate reader (Perkin Elmer). Two methods are described for quantifying the data, 50% ADCC titers and area under the curve (AUC) values for ADCC. We estimated 50% ADCC titers as previously described for virus neutralization assays (4). The 50% intercept was calculated using the adjacent %RLU values above and below 50% RLU. AUC values for ADCC were calculated from log10transformed %RLU values by the trapezoidal method. Log10%RLU values were subtracted from log10100 at each antibody dilution tested. To obtain an area, the sum of these differences was multiplied by the dilution factor, log102. Any %RLU values ⬍1% were replaced with 1%, and any negative sums (i.e., due to %RLU values of ⬎100) were replaced with zeroes. Spearman correlation coefficients (RS) and tests for statistical significance were calculated using Prism version 4.1b (GraphPad Software). Depletion of macaque antibodies reactive with human cellular antigens. Some macaques produce antibodies that bind antigens expressed on human cells. To measure these antibodies, we performed ADCC assays by detecting the background level of luciferase expressed by uninfected target cells. Macaque antibodies reactive with human cells were depleted by adsorption onto uninfected CEM.NKR-CCR5 cells. A pellet of 107 CEM.NKR-CCR5 cells was resuspended in each plasma sample. After a 20-min incubation at room temperature with occasional gentle tapping, plasma was separated from cells by centrifugation, removed, and used to resuspend a fresh pellet of 107 CEM.NKR-CCR5 cells. Six iterations of this procedure were sufficient to deplete detectable antibodies from most samples, but 20 rounds of adsorption onto uninfected target cells were required for samples from macaques immunized with single-cycle SIV, which is produced in human embryonic kidney 293T cells. Assessment of ADCC by flow cytometry. Parallel assays were performed to compare flow cytometry and luciferase as measures of cytotoxicity. Target cells were infected 4 days earlier with vesicular stomatitis virus (VSV) G trans-complemented single-cycle SIV, which expresses enhanced green fluorescent protein (eGFP) from the nef locus. On the day of the assay, infected and uninfected target cells were both labeled with the membrane-labeling dye PKH26 according to the protocol provided by the manufacturer (Sigma). After an 8-h incubation in the presence of macaque CD16⫹ NK cells and serial dilutions of plasma from an animal infected with SIVmac239, the luciferase activity of one plate was measured as described above. The cells on a second plate were resuspended, washed twice with PBS, and incubated at room temperature in the dark for 30 min in the presence of an amine-reactive dye, Live/Dead fixable far red stain (Molecular Probes, Invitrogen). Cells were subsequently washed in 2% FBS in PBS and resuspended in 2% formaldehyde in PBS. Data were collected on a FACSCalibur flow cytometer (BD Biosciences). Neutralization assays. Neutralization assays were performed using triplicate serial dilutions of plasma. Neutralization of SIVmac239 and SIVmac251TCLA was measured using a C8166 reporter cell line, which expresses secreted alkaline phosphatase (SEAP) from an SIV LTR promoter (83). The sensitivity to detect neutralization of SIVmac239 was maximized

November 2012 Volume 86 Number 22

by minimizing the amount of virus input necessary to achieve consistent results, which was a concentration of infectious virus containing 0.5 ng SIV p27 per well in flat-bottom, tissue culture-treated polystyrene 96-well plates. Neutralization assays using SIVmac251TCLA were performed with 2 ng SIV p27 per well. Plasma and virus were incubated together for 1 h at 37°C before adding 15,000 C8166-SEAP cells per well. SEAP activity was measured after 3 days using a Phospha-Light SEAP detection kit (Applied Biosystems). Neutralization assays against HIV-1YU2 were performed in TZM-bl cells (132, 133), using an inoculum of infectious virus containing 2 ng HIV-1 p24 per well. TZM-bl cells were seeded to flat-bottom 96-well plates the day before each neutralization assay at a density of 5,000 cells per well. As for the SIV neutralization assays, antibody dilutions and viruses were incubated for 1 h at 37°C before being combined with reporter cells. Luciferase activity in TZM-bl cells was measured 3 days later, using BriteLite Plus luciferase substrate (Perkin Elmer). The antibody titers that neutralized 50% of the virus infection were calculated using the same formula used to calculate 50% ADCC titers. ELISAs. Maxisorb ELISA plates (Nunc) were coated with recombinant, 6-His-tagged SIVmac239 or HIV-1HxB2 gp120 produced in 239T cells (Immune Technology) at a concentration of 0.5 ␮g/ml in a 0.1 M sodium bicarbonate (pH 9.5) buffer. Plates were blocked overnight at 4°C in PBS containing 0.5% Tween 20 (Sigma) and 5% nonfat dry milk. Antibody samples were diluted in PBS containing 0.5% Tween 20 and 5% nonfat dry milk. Bound antibodies were measured using a horseradish peroxidase-conjugated goat anti-monkey/-human IgG antibody (Santa Cruz Biotechnology). A statistically defined endpoint titer was calculated from the mean plus three standard deviations for preimmune plasma samples from nine macaques at a 1:100 dilution, which was the highest plasma concentration tested (49). Viruses. VSV G trans-complemented single-cycle SIVmac239 expressing eGFP from the nef position (35), SIVmac239, HIV-1 NL4-3, and HIV1YU2 viruses were produced by transfection of 293T cells. SIVmac251TCLA was propagated in MT4 cells (83). SHIVSF162P3 (AIDS Research and Reference Reagent Program, NIAID, NIH, contributed by Janet Harouse, Cecilia Cheng-Mayer, Ranajit Pal, and the DAIDS, NIAID) was expanded in rhesus peripheral blood mononuclear cells (PBMC) that were depleted of CD8⫹ cells by using anti-CD8 magnetic beads (Invitrogen) and activated with phytohemagglutinin (PHA) (Sigma). HIV-192TH023 (AIDS Research and Reference Reagent Program, NIAID, NIH, contributed by the UNAIDS Network for HIV Isolation and Characterization) was initially grown in CD8-depleted, PHA-activated human PBMC and then expanded by short-term culture in CEM.NKR-CCR5 cells. The SHIVKB9 used here was passaged extensively in CEM.NKR-CCR5 cells to generate a strain with higher infectivity in this target cell line. Purified IgG. To purify IgG from plasma, samples were incubated on a rotator overnight at 4°C with protein A beads (GE Healthcare). These were washed first with PBS containing 0.5 M NaCl and second in PBS. IgG was eluted off the protein A beads with 3 M MgCl2 and subsequently dialyzed into PBS to remove the MgCl2. Dialyzed IgG samples were concentrated using Amicon centrifugal filtration units with a nominal molecular weight limit of 10 kDa (Millipore). Deglycosylation of IgG was achieved by treating 1 ml of plasma from an SIVmac239-infected macaque with 20 ␮g of endoglycosidase S (EndoS) for 2.5 h at 37°C (2, 3), and the IgG was purified away from the EndoS enzyme using protein A beads, as described above.

RESULTS

CD16 expression on NK cell lines. Limiting-dilution clones of KHYG-1 NK cells were selected that express comparable levels of human or rhesus macaque CD16 (Fig. 1A). The levels of CD16 expressed on these clones were compared to the levels expressed on primary NK cells from five human donors (Fig. 1B) and from five rhesus macaques (Fig. 1C). The NK cells included in this comparison are the CD56dim population in humans and the CD56⫺/dim population in macaques, which include the majority

jvi.asm.org 12041

Alpert et al.

FIG 1 CD16 expression on cloned NK cell lines versus primary NK cells. (A) CD16 expression on the parental KHYG-1 NK cell line (solid gray) and the limiting-dilution clones of KHYG-1 cells transduced with retroviral vectors expressing human CD16 (red) or rhesus macaque CD16 (blue). (B) Expression levels on the human CD16⫹ KHYG-1 clone (solid red) versus levels on primary CD56dim NK cells from 5 people (colored lines). (C) Expression levels on the rhesus CD16⫹ KHYG-1 clone (solid blue) versus levels on primary CD56⫺/dim NK cells from 5 macaques (colored lines). The populations of primary NK cells shown are CD20⫺, CD3⫺, HLA-DR⫺, CD8⫹, and NKG2A⫹.

of CD16⫹ cytolytic NK cells in the peripheral blood of these species (28, 107). This comparison demonstrates that the levels of CD16 expressed on the cloned NK cell lines are similar to or slightly lower than the levels expressed on the CD16⫹ populations of primary human and macaque NK cells. Luciferase as an indicator of cytotoxicity. We derived a target cell line from CEM.NKR-CCR5 CD4⫹ T cells (60, 129) by stable transduction with a Tat-inducible luciferase reporter gene and evaluated the potential for the luciferase expressed by virus-infected target cells to serve as an indicator of cytotoxicity. A time course experiment was performed to determine the kinetics of the loss of luciferase activity from virus-infected target cells in the presence of NK cells and virus-specific antibody (Fig. 2A). Target cells infected with SIVmac239 were incubated alone or in the presence of the macaque CD16⫹ NK cell line and plasma from naïve or SIVmac239-infected animals. Luciferase activity was measured in cells and supernatants collected every hour for 12 h. Relative to the luciferase activity in SIVmac239-infected target cells that were incubated with NK cells, there were no differences in the luciferase activities measured in the absence of NK cells or in the presence of NK cells and plasma from a naïve animal (Fig. 2A). In contrast, there was a loss of luciferase activity over time from the SIVmac239-infected cells that were incubated with NK cells and plasma from an SIVmac239-infected animal, consistent with the elimination of virus-infected cells expressing luciferase. A simultaneous increase in the luciferase activity detectable in the supernatant was also measured during the first 4 h. This initial accumulation of luciferase in the cell culture supernatant is consistent with the lysis of the majority of virus-infected cells occurring during the first few hours of the assay. However, this luciferase activity declined to baseline levels after 8 h, suggesting that the enzyme is rapidly inactivated or degraded in the supernatant. Hence, this experiment associates the loss of luciferase activity from virusinfected cells with a loss of membrane integrity and suggests that an 8-h incubation maximizes the sensitivity of detection of a decrease in luciferase activity due to the lysis of virus-infected cells. As an alternative approach to confirm that the loss of luciferase from virus-infected cells reflects cytotoxicity, we performed parallel assays using both luciferase and flow cytometry to measure target cell killing. Target cells were infected with a virus derived

12042

jvi.asm.org

from SIVmac239 that is limited to a single cycle of infection and expresses enhanced green fluorescent protein (eGFP) from the nef locus (35). Prior to the addition of macaque CD16⫹ NK cells, the target cells were stained with a membrane labeling dye (PKH26) to facilitate discrimination between the effector and target cell populations by flow cytometry (51). Infected target cells were incubated in the presence of NK cells and serial dilutions of plasma from an animal infected with SIVmac239. After 8 h, the luciferase activity was measured for one set of samples, and the second set was stained with an amine-reactive dye that is excluded by the intact cellular membranes of live cells. In these parallel assays, the percentage of maximum luciferase activity corresponded to the percentage of maximum live eGFP-positive (eGFP⫹) cells over a range of plasma dilutions (Fig. 2B). The percentage of dead target cells stained by the amine-reactive dye was lower among uninfected target cells (1.8%) than among infected target cells that were also incubated in the absence of plasma (2.87%) (Fig. 2C and D), due in part to the cytopathic effects of SIVmac239 infection. In the presence of plasma diluted 102- to 103.5-fold, the loss of live eGFP⫹ target cells was associated with an increase in the number of dead target cells stained by the amine-reactive dye (Fig. 2E to H). Over the higher 104- to 105.5-fold dilutions of plasma, the live eGFP⫹ cell population increased in frequency, approaching the maximum percentage of live eGFP⫹ cells observed in the absence of plasma (Fig. 2I to L). Thus, live eGFP⫹ target cells were eliminated in the presence of NK cells and plasma in a dose-dependent manner that corresponded to the loss of luciferase activity, while dead target cells identified by the amine-reactive dye increased in number. The loss of luciferase from virus-infected cells therefore corresponds to the killing of infected target cells by ADCC. Specificity of ADCC activity. Experiments were performed to confirm that the ADCC activity observed in this assay was Env specific, IgG mediated, and Fc receptor dependent. A reciprocal Env mismatch experiment tested the ability of plasma samples from macaques infected with SIVmac239, or with the simian-human immunodeficiency virus chimera SHIVKB9, to direct the killing of cells infected with these viruses (Fig. 3A and B). Plasma from an animal infected with SIVmac239 directed the killing of cells infected with SIVmac239 but not SHIVKB9 (Fig. 3A). Likewise, plasma from an SHIVKB9-infected macaque directed the killing of

Journal of Virology

ADCC against HIV-1- and SIV-Infected Cells

FIG 2 Luciferase activity as an indicator of cytotoxicity and target cell membrane integrity. Relative light units (RLU) indicate luciferase activity. Wells containing infected target cells and NK cells with no plasma define 100% RLU, whereas wells containing uninfected target cells and NK cells define 0% RLU. (A) Luciferase activity was lost over a 12-h time course from target cells infected with SIVmac239 (filled symbols) in the presence of the NK cell line and plasma from SIVmac239-infected (red) but not from naïve (blue) animals, concomitant with an increase in the low level of luciferase activity detectable in supernatants (open symbols). (B) ADCC activity against target cells infected with single-cycle SIV-eGFP was quantified by measuring luciferase activity or the frequency of cells positive for eGFP but negative for an amine-reactive dye. The dashed line indicates 50% maximal RLU or eGFP⫹ cells. (C and D) Uninfected target cells incubated with NK cells and no plasma were eGFP negative (C), whereas infected target cells incubated under the same conditions were eGFP⫹(D). (E to L) The frequencies of dead and living eGFP⫹ target cells were determined after an 8-h incubation in the presence of NK cells and plasma diluted (fold) 102 (E), 102.5 (F), 103 (G), 103.5 (H), 104 (I), 104.5 (J), 105 (K), or 105.5 (L).

cells infected with SHIVKB9 but not SIVmac239 (Fig. 3B). Based on these observations and the fact that Env is the only viral protein expressed on the cell surface, the cytotoxicity measured using this assay is Env specific. The cytotoxicity of the cloned NK cell lines expressing macaque or human CD16 was compared with that of the parental KHYG-1 cell line. Macaque CD16⫹ NK cells but not parental KHYG-1 cells, killed SIVmac239-infected target cells in the presence of plasma from a macaque infected with SIVmac239 (Fig. 3C). Likewise, the human CD16⫹ NK cell line was cytotoxic against HIV-1NL4-3-infected target cells in the presence of HIV-1⫹ patient plasma, whereas the parental KHYG-1 cell line was not (Fig. 3D). These results demonstrate that CD16 expression by the NK cell line is required for the killing of virus-infected cells in the presence of plasma. To verify that the IgG fraction of plasma is responsible for the cytotoxicity measured in this assay, IgG was purified from the plasma of an SIVmac239-infected macaque using protein A beads. The IgG and plasma were compared for the ability to direct the killing of SIVmac239-infected target cells by the macaque CD16⫹ NK cell line (Fig. 3E), using input amounts of IgG and plasma that were normalized to have equivalent titers by gp120 ELISA. An identically formatted assay was performed using IgG purified

November 2012 Volume 86 Number 22

from the pooled plasma of HIV-1⫹ patients, the human CD16⫹ NK cell line, and target cells infected with HIV-1NL4-3 (Fig. 3F). In both experiments, luciferase activity was lost to a similar extent at nearly all dilutions of plasma and IgG tested (Fig. 3E and F). These comparisons demonstrate that the cytotoxicity measured in this assay is mediated by the IgG fraction of plasma. The interaction between IgG and Fc receptors, including CD16, is dependent upon the presence of a single N-linked glycan attached to the Fc region of the antibody (94). Streptococcus pyogenes encodes an enzyme, endoglycosidase S (EndoS), which specifically hydrolyzes this N-linked glycan (3, 27). Treatment of IgG with EndoS inhibits ADCC and complement-mediated lysis (2). To confirm that the cytotoxicity measured in our assay is dependent upon an Fc receptor-mediated interaction, we compared EndoS-treated IgG with untreated IgG (Fig. 3G). Indeed, EndoStreated IgG from an SIVmac239-infected macaque had no detectable ability to mediate cytotoxicity, whereas the untreated IgG directed the killing of SIVmac239-infected cells. Removal of the N-linked glycan from IgG heavy chains was verified by SDSPAGE, followed by staining total protein with Coomassie blue or blotting with a lectin specific for ␣-mannose residues, Lens culinaris agglutinin (LCA) (Fig. 3H). The LCA blot showed that treatment with EndoS removed nearly all of the N-linked glycan from

jvi.asm.org 12043

Alpert et al.

FIG 3 Cytotoxicity is Env-specific, CD16-dependent, and mediated by IgG. (A and B) The luciferase activity in cells infected with SIVmac239 or SHIVKB9 was measured after an 8-h incubation in the presence of the macaque CD16⫹ NK cell line and serial dilutions of plasma from macaques infected with SIVmac239 (A) or SHIVKB9 (B). The dashed line indicates 50% ADCC activity. (C and D) The parental KHYG-1 cell line was compared to the NK cell clones expressing macaque CD16 (C) or human CD16 (D) for cytotoxicity directed by SIVmac239-infected macaque plasma against SIVmac239-infected cells or by HIV-1⫹ patient plasma against HIV-1NL4-3-infected cells, respectively. (E) Plasma and IgG purified from the same sample using protein A beads were compared for their ability to direct the killing of SIVmac239-infected cells. (F) Cytotoxicity against HIV-1NL4-3 was compared for HIV-1⫹ patient plasma versus IgG purified from the same plasma pool on protein A beads. These plasma and IgG samples were normalized for binding to SIVmac239 or HIV-1HxB2 gp120 in ELISAs. (G) EndoS-treated IgG from an SIVmac239-infected macaque was compared with the untreated IgG samples for killing of target cells infected with SIVmac239. (H) A Coomassie stain (Coo.) of total protein (left 2 lanes) and a Western blot using Lens culinaris agglutinin (LCA) to detect the Fc region N-linked glycan (right 2 lanes) were performed to confirm its removal by EndoS. Inf., infected.

the IgG sample. These observations indicate that the cytotoxicity measured by this assay requires the IgG heavy chain to be glycosylated. Therefore, the cytotoxicity measured in this assay is Env specific, dependent upon the Fc receptor CD16, and mediated by IgG. Antibodies reactive with human cellular antigens. Some rhesus macaques have antibodies that bind antigens present on human cells, which can interfere with the detection of virus-specific ADCC. The ability of plasma samples to direct ADCC against uninfected target cells was assessed by measuring changes from the baseline level of luciferase activity. At a 1:32 dilution of plasma, 43% of macaques (45 of 105) had antibodies with at least 20% ADCC activity against uninfected target cells (Fig. 4A). Notably, 12% of macaques exceeded 80% ADCC activity against uninfected target cells. In contrast, only 5.7% of HIV-1-negative human volunteers (22 of 386) exceeded 20% ADCC activity against cells infected with HIV-1 at a 1:32 dilution of plasma (Fig. 4B). Therefore, antibodies reactive with human cellular antigens are a significant technical concern for measuring virus-specific ADCC activity in nonhuman primates but are less of a concern for human subjects. Macaque antibodies that react with human cellular antigens can be depleted by adsorption onto uninfected target cells. Six rounds of depletion were sufficient to eliminate detectable ADCC activity against uninfected cells by plasma from 17 of 18 animals (Fig. 4C). However, immunization of macaques with material produced in human cells can elicit antibody responses against human cellular antigens (29, 50, 106, 123), and these antibodies require more iterations of the depletion procedure to eliminate. Pooled plasma from animals immunized with single-cycle SIV produced by transfection of human embryonic kidney 293T cells (63) exhibited similarly potent ADCC activity against target cells

12044

jvi.asm.org

infected with SIVmac239 or SHIVSF162P3 (Fig. 4D). However, only the Env-specific ADCC activity against SIVmac239-infected target cells was detectable after adsorbing antibodies to uninfected target cells 20 times (Fig. 4D). To determine whether the depletion procedure reduces virus-specific ADCC activity, we measured ADCC against SIVmac239-infected cells before and after 20 rounds of depletion using plasma from two SIVmac239-infected animals that lacked detectable ADCC activity against uninfected cells (Fig. 4E and F). Both animals lacked discernible ADCC activity against SHIVSF162P3, confirming the Env specificity of the signal measured. However, multiple depletions had little or no effect on the ADCC activity measured against SIVmac239-infected target cells. Therefore, rhesus macaques often have antibodies that recognize human cellular antigens, but these antibodies can be depleted without an appreciable loss of virus-specific ADCC activity. Preventing the loss of ADCC activity associated with cryopreservation. Cryopreservation of the NK cell line under standard conditions leads to a loss of cytotoxicity (Fig. 5A). We hypothesized that this loss of cytotoxicity may relate to the ability of calcium flux to induce an unresponsive or anergic state, which can be prevented in T cells by the calcineurin inhibitor cyclosporine (CsA) (62, 79). Indeed, the loss of cytolytic activity after cryopreservation of the NK cell line could be prevented by CsA (Fig. 5A). To further our understanding of the causes of unresponsiveness following cryopreservation, we performed additional experiments using macaque CD16⫹ KHYG-1 cells that were taken off CsA 3 weeks after being thawed. These cells were cultured for an additional 3 weeks in the absence of CsA to minimize the influence of any residual CsA on subsequent experiments. The ADCC activities of these cells were compared after they were frozen and thawed, treated with freezing medium but not frozen, or treated with ionomycin in the presence or absence of the calcineurin in-

Journal of Virology

ADCC against HIV-1- and SIV-Infected Cells

FIG 4 Antibody reactivity with target cells. (A) ADCC activity against uninfected target cells was measured at a 1:32 dilution of plasma from 105 rhesus macaques. (B) The baseline level of ADCC activity at a 1:32 dilution of plasma from 386 naïve human volunteers was tested against target cells infected with HIV-192TH023. (C) ADCC activity titers of plasma samples against uninfected target cells before (gray) and after (black) 6 rounds of depletion were determined. The dashed line indicates 50% ADCC activity. (D) Pooled plasma from macaques immunized with single-cycle SIV produced in human cells had ADCC activity against cells infected with SIVmac239 (black filled symbols) with potency similar to that of SHIVSF162P3 (gray filled symbols). After 20 rounds of depletion, ADCC activity was detectable against cells infected with SIVmac239 (black open symbols) but not SHIVSF162P3 (gray open symbols). (E and F) Performing 20 rounds of depletion on plasma from 2 animals that lacked detectable ADCC activity against uninfected target cells did not appreciably reduce the ADCC activity against SIVmac239-infected target cells. Inf., infected; Undep., undepleted; Dep., depleted.

hibitors CsA and FK-506 (Fig. 5B to D). Both CsA and FK-506 were able to prevent the induction of unresponsiveness by cryopreservation (Fig. 5B). However, treatment of these cells with freezing medium alone, without actually freezing the cells, resulted in a loss of ADCC activity that could be prevented with CsA and FK-506 (Fig. 5C). Likewise, CsA and FK-506 also prevented the loss of cytolytic activity observed after stimulating the macaque CD16⫹ NK cell line with ionomycin (Fig. 5D). These observations suggest that treatment with freezing medium or ionomycin induces the negative regulation of cytolytic activity through

a pathway that can be inhibited by CsA and FK-506. Thus, the induction of unresponsiveness by cryopreservation can be avoided by freezing and culturing the NK cells in the presence of calcineurin inhibitor CsA or FK-506. Reproducibility. Reproducibility was assessed by performing identical ADCC assays on 10 different days over a time period of 2 months. In each assay, the ADCC titer of purified HIV-1⫹ patient IgG (HIVIG; obtained from the AIDS Research Reference Reagent Program) against target cells infected 4 days earlier with HIV1NL4-3 was determined (Fig. 6A). The similarity of the HIVIG titra-

FIG 5 Freezing medium causes a loss of cytotoxicity that can be prevented by calcineurin inhibitors. (A) ADCC before and after cryopreservation of the macaque CD16⫹ NK cell line under standard conditions and after cryopreservation in the presence of cyclosporine (CsA). (B to D) ADCC by macaque CD16⫹ NK cells taken through a freeze-thaw cycle (B), treated with freezing medium but not frozen (C), or treated with ionomycin (D) in the presence or absence of calcineurin inhibitor CsA or FK-506. The dashed line indicates 50% ADCC activity for a plasma sample from an SIVmac239-infected macaque against SIVmac239-infected target cells.

November 2012 Volume 86 Number 22

jvi.asm.org 12045

Alpert et al.

FIG 7 Neutralization and ADCC of a primary HIV-1 isolate. (A and B) Plasma samples from one healthy volunteer (gray) and 10 HIV-1⫹ patients not taking antiretroviral drugs were tested for neutralization of HIV-1YU2 in the TZM-bl assay (118, 132, 133) (A) and for ADCC against target cells infected with HIV-1YU2 (B). Dashed lines indicate 50% neutralization (A) or ADCC activity (B).

FIG 6 Reproducibility of the assay. (A) ADCC activity titers of HIVIG reagent against cells infected with HIV-1NL4-3 were determined on 10 different days. The dashed line indicates 50% ADCC activity. (B) The CV was determined at each concentration of HIVIG for the data obtained on different days and among the triplicate samples tested on the same day. (C) The 50% ADCC titers ranged from 1.4 to 7.3 ␮g/ml, with a standard deviation of 2.5 ␮g/ml (indicated by error bar) and a CV of 65%.

tion curves indicates that the NK cell line maintained a relatively stable level of cytolytic activity over the course of 2 months. The variability among tests performed on the same day was quantified as described elsewhere (104), by determining the coefficient of variation (CV) at each dilution tested (Fig. 6B). The CVs for the three wells that comprise each triplicate reading on the same day were 3.7 to 13.8%, averaging 10% at the higher concentrations of HIVIG, where ADCC activity was greater than 50%, but declining to 5% at lower concentrations of HIVIG. We also calculated the CV for assays performed on different days (Fig. 6B). The CV values for assays performed on the same day were lower than for those performed on different days, which ranged from 30% at the highest HIVIG concentration tested to approximately 5% at lower concentrations of HIVIG. The HIVIG concentrations estimated to have 50% ADCC activity on 10 different days ranged from 1.4 to 7.3 ␮g/ml (standard deviation, 2.5 ␮g/ml) and had a CV of 65% (Fig. 6C). These results suggest that although similar data were obtained on different days, direct comparisons of ADCC activity among different samples should be made using assays performed on the same day whenever possible. Neutralization versus ADCC against a primary HIV-1 isolate. ADCC and neutralization of a primary HIV-1 isolate were compared using plasma samples from HIV-1⫹ patients. The virus used for these comparisons, HIV-1YU2, is a neutralization-resistant, CCR5-tropic primary isolate that was cloned directly from patient tissue without in vitro expansion (18, 77, 114). Plasma samples from 10 HIV-1⫹ patients in the United States, who were not taking antiretroviral therapy, and from one uninfected volun-

12046

jvi.asm.org

teer were tested for neutralization of HIV-1YU2 (Fig. 7A). All of the HIV-1⫹-patient plasma samples had detectable antibodies that neutralized HIV-1YU2, which inhibited virus infection by 50% at 13- to 472-fold dilutions of plasma, with a mean 50% neutralizing antibody titer of 82. The same plasma samples were tested for their ability to direct ADCC against HIV-1YU2-infected cells (Fig. 7B). On average, the 50% ADCC titers were 115-fold higher than the 50% neutralization titers for the samples where both ADCC and neutralization were measurable (P ⫽ 0.0156, 2-tailed Wilcoxon matched-pairs test). Two of the samples lacked detectable ADCC activity, whereas the others had 50% ADCC titers ranging from 46 to 12,344, with a mean of 3,992. One of the two samples that lacked detectable ADCC activity had a 50% neutralizing antibody titer of 64. Another plasma sample had a low level of ADCC activity that did not cross the 50% threshold. Thus, ADCC titers were generally higher than neutralizing antibody titers, but there was not a direct correspondence between the antibody titers measured by these assays. ADCC in comparison to neutralization and gp120-binding assays. The ability of plasma to direct ADCC against cells infected with a primary SIV isolate, SIVmac239, was compared with neutralization of the same virus, neutralization of T-cell line-adapted SIVmac251TCLA, and binding to recombinant SIVmac239 gp120 protein by ELISA. The plasma samples used for these comparisons were collected from 37 macaques in the chronic phase of SIVmac239 infection, an average of 35 weeks postchallenge (range, 16 to 52 weeks). Since all of the animals were infected with SIVmac239, differences in ADCC are not due to extensive viral diversity, which would complicate similar comparisons using samples from HIV-1⫹ patients. All of the macaque plasma samples were depleted of antibodies reactive with human cellular antigens. Most of these plasma samples directed ADCC against cells infected with SIVmac239 (Fig. 8A). Excluding 11 animals that lacked measurable 50% ADCC titers, the average 50% ADCC titer was 16,162. In assays performed in parallel, the plasma samples had low to undetectable ADCC activity against cells infected with SHIVKB9, indicating that the ADCC activity measured against SIVmac239-infected cells was Env specific. Thus, most, but not all, of the animals chronically infected with SIVmac239 made Envspecific antibodies capable of directing ADCC against cells infected with this virus. Although calculating 50% or endpoint dilution titers produces numerical values with a wide dynamic range, plasma is undiluted in vivo. Therefore, a measure that reflects the extent of target cell

Journal of Virology

ADCC against HIV-1- and SIV-Infected Cells

FIG 8 ADCC versus neutralization of SIVmac239, neutralization of SIVmac251TCLA, and binding to gp120. (A) Plasma samples from 37 unvaccinated rhesus macaques chronically infected with SIVmac239 were tested for ADCC against cells infected with SIVmac239 and against cells infected with SHIVKB9 (gray). The dashed line indicates 50% ADCC activity. (B) ADCC activity against SIVmac239-infected cells was quantified by calculating 50% ADCC titers and area under the curve (AUC) values for ADCC, and these measures were compared. (C and D) The same plasma samples were also tested for neutralization of SIVmac239 (C), and the neutralization and ADCC titers were compared (D). The dashed line indicates 50% neutralization. (E and F) The titers for neutralization of SIVmac251TCLA in plasma samples were determined (E), and the neutralization titers against SIVmac251TCLA were compared with ADCC titers against SIVmac239-infected cells (F). (G and H) Binding to SIVmac239 gp120 was measured by ELISA (G), and the endpoint ELISA titers (H), where each titration curve crosses the dashed line, were compared with ADCC titers against SIVmac239-infected cells. Neut., neutralization.

elimination at higher concentrations of plasma may represent physiological conditions more accurately than a titer (119). Therefore, as an alternative approach for quantifying the data, we calculated area under the curve (AUC) values for ADCC from log10-transformed %RLU values. This approach gives greater relative weight to differences in the extent of target cell elimination at higher antibody concentrations and yields values that are proportional to log10-transformed 50% ADCC titers (Fig. 8B). These two methods for quantifying ADCC activity produced well-correlated data sets, with a Spearman correlation coefficient (RS) of 0.8824 (P ⬍ 0.0001). Two of the 11 plasma samples that lacked detectable 50% ADCC titers had low but measurable AUC values for ADCC, suggesting that the latter quantification method may also facilitate

November 2012 Volume 86 Number 22

comparisons that include samples with low ADCC activity. Therefore, AUC values for ADCC may have advantages over 50% titers for certain comparisons, although both methods are well correlated. The plasma samples from 37 macaques chronically infected with SIVmac239 were tested for their ability to neutralize SIVmac239. Sixteen of the 37 animals had detectable 50% neutralizing antibody titers against SIVmac239, with a mean titer of 66 (Fig. 8C). All 16 animals that neutralized SIVmac239 also directed ADCC against SIVmac239-infected cells. The 50% ADCC titers in these animals were an average of 418-fold higher than the 50% neutralizing antibody titers (P ⫽ 0.0002, 2-tailed Wilcoxon matched-pairs test). Thus, for the same plasma sample paired with the same test virus, the ADCC assay was a more sensitive measure of Env-specific antibodies than an optimized neutralization assay by at least 2 orders of magnitude. Perhaps due to this sensitivity, 10 animals that lacked detectable 50% neutralizing antibody titers had measurable 50% ADCC titers. Thus, the absence of detectable neutralization does not imply the absence of ADCC. However, neutralization and ADCC titers were well correlated (P ⬍ 0.0001, RS ⫽ 0.7296) (Fig. 8D). The comparison between ADCC and neutralization assays using SIVmac239 suggests that these measures are related but that the ADCC assay is considerably more sensitive than the neutralization assay in most cases. These plasma samples were also tested for neutralization of SIVmac251TCLA, a strain that has lost its resistance to neutralizing antibodies as a result of adaptation to an immortalized T-cell line (83). Only 6 of the 37 macaques lacked measurable 50% neutralization titers against SIVmac251TCLA (Fig. 8E). ADCC against SIVmac239-infected cells correlated with neutralization of SIVmac251TCLA (P ⫽ 0.004, RS ⫽ 0.5527) (Fig. 8F), albeit not as well as neutralization of SIVmac239. Five animals with detectable SIVmac251TCLA neutralization titers lacked measurable ADCC titers against SIVmac239-infected cells. Three of these five animals had SIVmac251TCLA neutralization titers that exceeded 20,000, suggesting that neutralization of SIVmac251TCLA is not predictive of ADCC. Thus, although ADCC against SIVmac239-infected cells and neutralization of SIVmac251TCLA correlate in most cases, titers of neutralizing antibody to SIVmac251TCLA are not a surrogate for ADCC activity. These plasma samples were also tested for antibodies that bind SIVmac239 gp120 by ELISA (Fig. 8G). The ELISA titers against gp120 were correlated with the ADCC titers against SIVmac239infected cells (P ⫽ 0.0206, RS ⫽ 0.3794) (Fig. 8H), although the relationship was weaker than for neutralization of SIVmac239 or SIVmac251TCLA. As noted for neutralization of SIVmac251TCLA, five samples that lacked measurable ADCC titers had gp120 ELISA titers. Four of these five samples bound gp120 at or above a titer of 20,000, suggesting that these animals made abundant antibodies to gp120 but failed to direct ADCC against SIVmac239-infected cells. Overall, ADCC against SIVmac239-infected cells correlated best with neutralization of SIVmac239, and least with binding to monomeric gp120. The relative strength of the relationship between ADCC against SIVmac239-infected cells and neutralization of SIVmac239 is probably due to the measurement of antibodies that recognize membrane-bound Env trimers of the same neutralization-resistant primary SIV isolate in both assays. Moreover, these comparisons suggest that neutralization of a T-cell lineadapted virus and binding to recombinant gp120 are not synony-

jvi.asm.org 12047

Alpert et al.

mous with ADCC against cells infected with a primary virus isolate. DISCUSSION

We have developed a practical method for measuring the capacity of antibodies to direct ADCC against cells infected with HIV-1 or SIV. The use of immortalized NK cell lines in combination with luciferase as an indicator of cytotoxicity enables the titration of many samples in parallel on a routine basis. The dose-dependent elimination of infected target cells by ADCC over serial antibody dilutions generates a curve, which facilitates quantitative comparisons over a large dynamic range. Thus, the potential for antibodies to direct ADCC against physiologically relevant targets can be quantified in a format analogous to state-of-the-art methods for measuring neutralizing antibodies (118, 132, 133). The use of this approach indicates that there is incomplete overlap between the subsets of antibodies that direct ADCC against virus-infected cells and the antibodies measured in neutralization and gp120-binding assays. HIV-1 vaccine approaches under consideration for clinical trials have not yet succeeded in eliciting antibodies capable of neutralizing primary isolates of HIV-1 or SIV that are representative of naturally transmitted HIV-1 isolates (10, 15, 24, 25, 41, 68, 80, 86, 95, 103, 110, 114, 118, 127). Whether antibodies capable of neutralizing primary isolates of HIV-1 can be elicited by vaccination remains unclear (15, 86, 122). Therefore, assays capable of measuring antiviral activity in the absence of detectable neutralization may be particularly valuable for the evaluation of HIV-1 vaccine candidates. Since the ADCC assay reported here is approximately 2 orders of magnitude more sensitive than optimized assays for neutralization of the same test viruses by the same plasma samples, it is often able to measure ADCC when neutralization is undetectable. The relative sensitivity of the ADCC assay may reflect differences in the proportions of Env trimers that must be bound by antibodies to detect neutralization versus ADCC and in the epitopes bound by these antibodies. Every functional Env trimer on a virion may need to be consistently occupied by antibody to detect neutralization (142, 143, 145), whereas a lower proportion of Env trimers bound by antibody may be sufficient to engage CD16. In addition, antibodies may be able to direct ADCC even if their association with Env does not interfere with entry or if they have a comparatively low affinity for functional Env trimers. Whereas neutralization assays measure only functional Env trimers, antibodies that bind nonfunctional conformations of Env may contribute to ADCC. Also, interactions between CD16 molecules and the Fc regions of antibodies may increase the avidity of interactions between those antibodies and Env. However, the sensitivity of the ADCC assay described here cannot be attributed to CD16 overexpression, since the levels of CD16 expressed by the NK cell lines were similar to or lower than the level of expression on primary NK cells. Therefore, the sensitivity of this assay probably reflects differences in the interactions between antibodies and Env that are necessary to detect neutralization versus ADCC. Recent observations from an HIV-1 vaccine clinical trial (RV144) are consistent with a protective role other than virus neutralization for antibodies. Study subjects immunized with recombinant gp120 and a canarypox vector acquired HIV-1 infection at a reduced rate relative to the rate in recipients of placebo immunizations (110). Although antibodies capable of neutraliz-

12048

jvi.asm.org

ing HIV-1 primary isolates were not detected, 98.6% of vaccine recipients had Env-specific antibodies that were detectable by gp120 ELISA. These antibodies have been hypothesized to account for vaccine protection (5), since vaccine recipients lacked significant CD8⫹ T-cell responses (110). In a follow-up study designed to identify immune correlates of vaccine protection in the RV144 trial, IgG antibodies capable of binding to the V1V2 region of gp120 correlated with protection and Env-specific IgA antibodies correlated with increased risk of infection (57). There was a nonsignificant trend toward a lower risk of HIV-1 infection among the vaccinated study subjects with higher levels of ADCC activity measured using the assay described here. Moreover, among vaccinated study subjects with low or undetectable Envspecific IgA responses, there was a borderline significant relationship between ADCC activity and lower risk of infection. Thus, although ADCC was not identified as a correlate of protection, these results point to a role for Env-specific antibody responses in the modest protection observed in the RV144 trial. The antibodies that bind Env despite the presence of features that have evolved to interfere with humoral immunity are important to measure, since these are the antibodies that are likely to possess antiviral activity in vivo. In contrast, many of the antibodies that bind recombinant forms of Env or neutralize T-cell lineadapted viruses recognize epitopes that may not be relevant to vaccine protection. Several previously published methods for measuring ADCC, or other Fc receptor-mediated antibody functions, are based on targets coated with peptides or recombinant forms of Env (1, 10, 19–22, 38–40, 46, 51–53, 59, 61, 64, 67, 101, 104, 124, 139). Other methods have relied on viruses that are not representative of neutralization-resistant primary isolates due to chronic propagation in T-cell lines (38, 40, 51, 53, 83, 104, 134) or the use of CXCR4 as a coreceptor (14, 39, 43, 139). Although these assays are influenced by differences in the Fc regions of antibodies, they have many of the same disadvantages as ELISAs for Envspecific antibodies or neutralization assays with T-cell lineadapted viruses. Indeed, a secondary analysis of ADCC activity in the RV144 trial revealed a nonsignificant trend toward higher risk of HIV-1 infection among the volunteers with higher ADCC activity against gp120-coated target cells—precisely the opposite of the trend observed with the assay described here (57). The opposite relationships between HIV-1 infection and ADCC activity against HIV-1-infected versus gp120-coated target cells suggests that these approaches measure different subsets of antibodies. Likewise, RV144 participants with higher neutralizing antibody titers against neutralization-sensitive HIV-1 isolates, including Tcell line-adapted strains, also may have a higher risk of HIV-1 infection (57). Therefore, assays that do not consider features of the HIV-1 Env protein that have evolved to resist antibody responses may not be appropriate for assessing correlates of vaccine protection or for optimizing vaccine antigens. Differences among the antibodies measured using neutralization-resistant primary isolates versus T-cell line-adapted viruses or recombinant gp120 protein may account for differences in the rank order of titers measured using the same plasma samples. ADCC against target cells infected with SIVmac239 correlated best with neutralization of SIVmac239, less well with neutralization of SIVmac251TCLA, and least with binding to recombinant SIVmac239 gp120 by ELISA. Thus, the assay with the strongest relationship to ADCC against SIVmac239-infected cells was neutralization of SIVmac239, which measures antibody binding to the native Env

Journal of Virology

ADCC against HIV-1- and SIV-Infected Cells

trimer on the surface of virions. Samples with different levels of ADCC activity had similar SIVmac251TCLA neutralization and gp120 ELISA titers, suggesting that antibodies capable of binding nonnative forms of Env may be elicited more consistently than antibodies that recognize the native trimer. Several animals neutralized SIVmac251TCLA or bound gp120 by ELISA at titers of 20,000 or above but lacked detectable ADCC activity against SIVmac239-infected cells. The antibodies produced by these animals probably recognized surfaces of Env that are available on SIVmac251TCLA virions or monomeric gp120 but not on SIVmac239 virions or SIVmac239-infected cells. Differences in IgG isotype or glycosylation are also potentially responsible for differences in the ADCC activities of samples with similar titers in other assays (13, 94). Therefore, the differences we observed using different types of assays suggest that the antibodies measured using T-cell lineadapted viruses or recombinant gp120 can differ from those that direct ADCC against cells infected with neutralization-resistant primary isolates. The main advantages of the approach described here stem from its use of an NK cell line. In contrast to ADCC assays performed using NK cells isolated from different donors (104), ADCC assays performed on 10 different days using immortalized NK cells had consistent levels of ADCC activity. The CV values for ADCC activity measured on different days were approximately 30% among tests performed at higher antibody concentrations and 65% for 50% ADCC titers. At higher antibody concentrations, the coefficients of variation among individual wells tested on the same day were 3.7 to 13.8%, which is similar to reported CV values of 7.6 to 22.9% or 6.7 to 11.4% for assays performed on the same day with primary NK cells as effectors and gp120-coated or chronically infected T-cell lines as target cells, respectively (104). Hence, the use of immortalized effector and target cell lines improves the reproducibility of ADCC assays. Cryopreservation of the NK cell lines resulted in a loss of cytolytic activity. However, we were able to circumvent this problem by freezing the NK cell lines in the presence of calcineurin inhibitors. The logic that led to our use of calcineurin inhibitors stemmed from a model in which cryopreservation stimulates the NK cell line to negatively regulate effector function. Stimulation can induce an unresponsive or anergic state in T and B cells (117), and chronic stimulation may contribute to NK cell dysfunction in the context of HIV-1 infection (6, 7, 37, 70, 81, 82, 85, 126). T-cell, B-cell, and Fc receptor cross-linking induces an influx of calcium into the cytoplasm (97). In the presence of cytoplasmic calcium, calmodulin activates the protein phosphatase calcineurin, which dephosphorylates nuclear factor of activated T cells (NFAT), exposing nuclear localization signals that promote the redistribution of NFAT to the nucleus, where it regulates gene expression (9). This pathway is thought to be important for anergy induction, since calcium flux does not render T cells unresponsive in NFAT1 knockout mice or in the presence of the calcineurin inhibitor CsA (62, 79). We speculated that cryopreservation may induce this negative regulatory pathway, since treatment of cells with 8% DMSO can induce calcium flux (76) and the freezing medium used here contains 10% DMSO as a cryoprotectant. Indeed, the addition of freezing medium was sufficient to induce a loss of ADCC activity that could be prevented with CsA or FK-506. Consistent with our model, CsA and FK-506 also prevented the induction of unresponsiveness after treating the NK cell line with ionomycin to mobilize an influx of calcium into the cytoplasm. Since

November 2012 Volume 86 Number 22

these calcineurin inhibitors are used as immunosuppressant drugs in clinical settings (9), it is perhaps counterintuitive that their use would promote cytotoxic effector function. However, these observations are consistent with a model in which treatment with freezing medium or ionomycin stimulates the NK cells to induce a calcineurin-dependent negative regulatory pathway, which can be blocked by CsA or FK-506. This solution to maintaining the cytolytic activity of the NK cell line through cryopreservation enabled the routine and reproducible measurement of ADCC activity. We have shown that the antibodies that direct ADCC are correlated with the antibodies measured by other types of assays but that these relationships are not absolute. Therefore, the best practice going forward may be to measure both virus-specific neutralization and ADCC titers. Since neutralization assays using primary HIV-1 isolates do not currently detect vaccine-elicited antibody responses, the assay described here may be particularly useful for evaluating the ability of antibody responses to recognize physiologically relevant forms of Env. ACKNOWLEDGMENTS We thank Ronald C. Desrosiers for providing plasma samples from SIVinfected macaques. We also thank Michelle Connole and Jacqueline Gillis, in the Division of Immunology, NEPRC, for flow cytometry services. We gratefully acknowledge the generosity of the Ministry of Public Health, Thailand, and the Thai AIDS Vaccine Evaluation Group in providing the RV144 clinical trial materials through the Henry M. Jackson Foundation for the Advancement of Military Medicine. Materials from the RV144 clinical trial are under the joint trusteeship of the Ministry of Public Health, Thailand and the US Army, on behalf and with the consent of the MOPH-TAVEG collaboration, and provided to collaborators through the Henry M. Jackson Foundation for the Advancement of Military Medicine. This work was supported by grants AI071306, AI063993, and RR000168/OD01103 from the National Institutes of Health, by supplement 3P01AI071306-04S1 from the American Recovery and Reinvestment Act (ARRA) of 2009, by contract 689633 from the Henry M. Jackson Foundation for the Advancement of Military Medicine and the U.S. Military HIV Research Program (MHRP), and by the University of Massachusetts Center for AIDS Research, grant P30AI04284. DTE is an Elizabeth Glaser Scientist of the Elisabeth Glaser Pediatric AIDS Foundation. A patent application for the in vitro and in vivo use of EndoS has been submitted by Genovis AB and Hansa Medical AB (US 2010/0135981). M. Allhorn is one of the listed inventors on these applications and has a royalty agreement with Genovis AB for products based on the in vitro use of EndoS.

REFERENCES 1. Ackerman ME, et al. 2011. A robust, high-throughput assay to determine the phagocytic activity of clinical antibody samples. J. Immunol. Methods 366:8 –19. 2. Allhorn M, et al. 2010. The IgG-specific endoglycosidase EndoS inhibits both cellular and complement-mediated autoimmune hemolysis. Blood 115:5080 –5088. 3. Allhorn M, Olin AI, Nimmerjahn F, Collin M. 2008. Human IgG/Fc gamma R interactions are modulated by streptococcal IgG glycan hydrolysis. PLoS One 3:e1413. doi:10.1371/journal.pone.0001413. 4. Alpert MD, et al. 2010. Envelope-modified single-cycle simian immunodeficiency virus selectively enhances antibody responses and partially protects against repeated, low-dose vaginal challenge. J. Virol. 84:10748 – 10764. 5. Alter G, Moody MA. 2010. The humoral response to HIV-1: new insights, renewed focus. J. Infect. Dis. 202(Suppl 2):S315–S322. 6. Alter G, et al. 2006. Low perforin and elevated SHIP-1 expression is associated with functional anergy of natural killer cells in chronic HIV-1 infection. AIDS 20:1549 –1551. 7. Alter G, et al. 2005. Sequential deregulation of NK cell subset distribu-

jvi.asm.org 12049

Alpert et al.

8. 9. 10. 11. 12.

13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

27. 28. 29. 30.

31. 32. 33. 34.

tion and function starting in acute HIV-1 infection. Blood 106:3366 – 3369. Anonymous. 1996. Guide for the care and use of laboratory animals, p 86 –123. National Academy Press, Washington, DC. Baine I, Abe BT, Macian F. 2009. Regulation of T-cell tolerance by calcium/NFAT signaling. Immunol. Rev. 231:225–240. Barouch DH, et al. 2012. Vaccine protection against acquisition of neutralization-resistant SIV challenges in rhesus monkeys. Nature 482: 89 –93. Bou-Habib DC, et al. 1994. Cryptic nature of envelope V3 region epitopes protects primary monocytotropic human immunodeficiency virus type 1 from antibody neutralization. J. Virol. 68:6006 – 6013. Broder CC, et al. 1994. Antigenic implications of human immunodeficiency virus type 1 envelope quaternary structure: oligomer-specific and -sensitive monoclonal antibodies. Proc. Natl. Acad. Sci. U. S. A. 91: 11699 –11703. Bruhns P, et al. 2009. Specificity and affinity of human Fcgamma receptors and their polymorphic variants for human IgG subclasses. Blood 113:3716 –3725. Bunnik EM, Quakkelaar ED, van Nuenen AC, Boeser-Nunnink B, Schuitemaker H. 2007. Increased neutralization sensitivity of recently emerged CXCR4-using human immunodeficiency virus type 1 strains compared to coexisting CCR5-using variants from the same patient. J. Virol. 81:525–531. Burton DR, et al. 2004. HIV vaccine design and the neutralizing antibody problem. Nat. Immunol. 5:233–236. Chen B, et al. 2005. Structure of an unliganded simian immunodeficiency virus gp120 core. Nature 433:834 – 841. Chen L, et al. 2009. Structural basis of immune evasion at the site of CD4 attachment on HIV-1 gp120. Science 326:1123–1127. Choe H, et al. 1996. The beta-chemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates. Cell 85:1135–1148. Chung AW, et al. 2011. Immune escape from HIV-specific antibodydependent cellular cytotoxicity (ADCC) pressure. Proc. Natl. Acad. Sci. U. S. A. 108:7505–7510. Chung AW, et al. 2011. Activation of NK cells by ADCC responses during early HIV infection. Viral Immunol. 24:171–175. Chung AW, et al. 2011. Activation of NK cells by ADCC antibodies and HIV disease progression. J. Acquir. Immune Defic. Syndr. 58:127–131. Chung AW, Rollman E, Center RJ, Kent SJ, Stratov I. 2009. Rapid degranulation of NK cells following activation by HIV-specific antibodies. J. Immunol. 182:1202–1210. Cocchi F, et al. 1995. Identification of RANTES, MIP-1 alpha, and MIP-1 beta as the major HIV-suppressive factors produced by CD8⫹ T cells. Science 270:1811–1815. Cohen J. 1994. The HIV vaccine paradox. Science 264:1072–1074. Cohen J. 1993. Jitters jeopardize AIDS vaccine trials. Science 262:980 – 981. Cole KS, et al. 2001. Characterization of neutralization epitopes of simian immunodeficiency virus (SIV) recognized by rhesus monoclonal antibodies derived from monkeys infected with an attenuated SIV strain. Virology 290:59 –73. Collin M, Olsen A. 2001. EndoS, a novel secreted protein from Streptococcus pyogenes with endoglycosidase activity on human IgG. EMBO J. 20:3046 –3055. Cooper MA, Fehniger TA, Caligiuri MA. 2001. The biology of human natural killer-cell subsets. Trends Immunol. 22:633– 640. Cranage MP, et al. 1993. Studies on the specificity of the vaccine effect elicited by inactivated simian immunodeficiency virus. AIDS Res. Hum. Retroviruses 9:13–22. Daar ES, Li XL, Moudgil T, Ho DD. 1990. High concentrations of recombinant soluble CD4 are required to neutralize primary human immunodeficiency virus type 1 isolates. Proc. Natl. Acad. Sci. U. S. A. 87: 6574 – 6578. Deng H, et al. 1996. Identification of a major co-receptor for primary isolates of HIV-1. Nature 381:661– 666. Desrosiers RC. 1999. Strategies used by human immunodeficiency virus that allow persistent viral replication. Nat. Med. 5:723–725. Dragic T, et al. 1996. HIV-1 entry into CD4⫹ cells is mediated by the chemokine receptor CC-CKR-5. Nature 381:667– 673. Earl PL, et al. 1994. Native oligomeric human immunodeficiency virus type 1 envelope glycoprotein elicits diverse monoclonal antibody reactivities. J. Virol. 68:3015–3026.

12050

jvi.asm.org

35. Evans DT, Bricker JE, Desrosiers RC. 2004. A novel approach for producing lentiviruses that are limited to a single cycle of infection. J. Virol. 78:11715–11725. 36. Evans DT, Desrosiers RC. 2001. Immune evasion strategies of the primate lentiviruses. Immunol. Rev. 183:141–158. 37. Fauci AS, Mavilio D, Kottilil S. 2005. NK cells in HIV infection: paradigm for protection or targets for ambush. Nat. Rev. Immunol. 5:835– 843. 38. Ferrari G, et al. 2011. An HIV-1 gp120 envelope human monoclonal antibody that recognizes a C1 conformational epitope mediates potent antibody-dependent cellular cytotoxicity (ADCC) activity and defines a common ADCC epitope in human HIV-1 serum. J. Virol. 85:7029 –7036. 39. Florese RH, et al. 2009. Contribution of nonneutralizing vaccineelicited antibody activities to improved protective efficacy in rhesus macaques immunized with Tat/Env compared with multigenic vaccines. J. Immunol. 182:3718 –3727. 40. Florese RH, et al. 2006. Evaluation of passively transferred, nonneutralizing antibody-dependent cellular cytotoxicity-mediating IgG in protection of neonatal rhesus macaques against oral SIVmac251 challenge. J. Immunol. 177:4028 – 4036. 41. Flynn NM, et al. 2005. Placebo-controlled phase 3 trial of a recombinant glycoprotein 120 vaccine to prevent HIV-1 infection. J. Infect. Dis. 191: 654 – 665. 42. Foreman KE, et al. 1994. C5a-induced expression of P-selectin in endothelial cells. J. Clin. Invest. 94:1147–1155. 43. Forthal DN, Gilbert PB, Landucci G, Phan T. 2007. Recombinant gp120 vaccine-induced antibodies inhibit clinical strains of HIV-1 in the presence of Fc receptor-bearing effector cells and correlate inversely with HIV infection rate. J. Immunol. 178:6596 – 6603. 44. Forthal DN, et al. 2006. Rhesus macaque polyclonal and monoclonal antibodies inhibit simian immunodeficiency virus in the presence of human or autologous rhesus effector cells. J. Virol. 80:9217–9225. 45. Forthal DN, Landucci G, Daar ES. 2001. Antibody from patients with acute human immunodeficiency virus (HIV) infection inhibits primary strains of HIV type 1 in the presence of natural-killer effector cells. J. Virol. 75:6953– 6961. 46. Fouda GG, et al. 2011. HIV-specific functional antibody responses in breast milk mirror those in plasma and are primarily mediated by IgG antibodies. J. Virol. 85:9555–9567. 47. Fouts TR, Binley JM, Trkola A, Robinson JE, Moore JP. 1997. Neutralization of the human immunodeficiency virus type 1 primary isolate JR-FL by human monoclonal antibodies correlates with antibody binding to the oligomeric form of the envelope glycoprotein complex. J. Virol. 71:2779 –2785. 48. Fouts TR, Trkola A, Fung MS, Moore JP. 1998. Interactions of polyclonal and monoclonal anti-glycoprotein 120 antibodies with oligomeric glycoprotein 120-glycoprotein 41 complexes of a primary HIV type 1 isolate: relationship to neutralization. AIDS Res. Hum. Retroviruses 14: 591–597. 49. Frey A, Di Canzio J, Zurakowski D. 1998. A statistically defined endpoint titer determination method for immunoassays. J. Immunol. Methods 221:35– 41. 50. Goldstein S, et al. 1994. Immunization with whole inactivated vaccine protects from infection by SIV grown in human but not macaque cells. J. Med. Primatol. 23:75– 82. 51. Gomez-Roman VR, et al. 2006. A simplified method for the rapid fluorometric assessment of antibody-dependent cell-mediated cytotoxicity. J. Immunol. Methods 308:53– 67. 52. Gomez-Roman VR, et al. 2006. An adenovirus-based HIV subtype B prime/boost vaccine regimen elicits antibodies mediating broad antibody-dependent cellular cytotoxicity against non-subtype B HIV strains. J. Acquir. Immune Defic. Syndr. 43:270 –277. 53. Gomez-Roman VR, et al. 2005. Vaccine-elicited antibodies mediate antibody-dependent cellular cytotoxicity correlated with significantly reduced acute viremia in rhesus macaques challenged with SIVmac251. J. Immunol. 174:2185–2189. 54. Gorny MK, Zolla-Pazner S. 2003. Human monoclonal antibodies that neutralize HIV-1, p 37–51. LA-UR 04 – 8162. In Korber BTM, Brander C, Haynes BF, Koup R, Moore JP, Walker BD, Watkins DI (ed), HIV immunology and HIV/SIV vaccine databases 2003. Los Alamos National Laboratory, Theoretical Biology and Biophysics, Los Alamos, NM. 55. Guidotti LG, Chisari FV. 2001. Noncytolytic control of viral infections

Journal of Virology

ADCC against HIV-1- and SIV-Infected Cells

56. 57. 58. 59.

60.

61. 62.

63. 64. 65. 66.

67.

68. 69. 70. 71. 72. 73. 74.

75.

76. 77.

78.

by the innate and adaptive immune response. Annu. Rev. Immunol. 19:65–91. Guo RF, Ward PA. 2005. Role of C5a in inflammatory responses. Annu. Rev. Immunol. 23:821– 852. Haynes BF, et al. 2012. Immune-correlates analysis of an HIV-1 vaccine efficacy trial. N. Engl. J. Med. 366:1275–1286. Hessell AJ, et al. 2007. Fc receptor but not complement binding is important in antibody protection against HIV. Nature 449:101–104. Hidajat R, et al. 2009. Correlation of vaccine-elicited systemic and mucosal nonneutralizing antibody activities with reduced acute viremia following intrarectal simian immunodeficiency virus SIVmac251 challenge of rhesus macaques. J. Virol. 83:791– 801. Howell DN, Andreotti PE, Dawson JR, Cresswell P. 1985. Natural killing target antigens as inducers of interferon: studies with an immunoselected, natural killing-resistant human T lymphoblastoid cell line. J. Immunol. 134:971–976. Isitman G, Chung AW, Navis M, Kent SJ, Stratov I. 2011. Pol as a target for antibody dependent cellular cytotoxicity responses in HIV-1 infection. Virology 412:110 –116. Jenkins MK, Chen CA, Jung G, Mueller DL, Schwartz RH. 1990. Inhibition of antigen-specific proliferation of type 1 murine T cell clones after stimulation with immobilized anti-CD3 monoclonal antibody. J. Immunol. 144:16 –22. Jia B, et al. 2009. Immunization with single-cycle SIV significantly reduces viral loads after an intravenous challenge with SIVmac239. PLoS Pathog. 5:e1000272. doi:10.1371/journal.ppat.1000272. Johansson SE, et al. 2011. NK cell function and antibodies mediating ADCC in HIV-1-infected viremic and controller patients. Viral Immunol. 24:359 –368. Johnson WE, Desrosiers RC. 2002. Viral persistance: HIV’s strategies of immune system evasion. Annu. Rev. Med. 53:499 –518. Johnson WE, et al. 2003. Assorted mutations in the envelope gene of simian immunodeficiency virus lead to loss of neutralization resistance against antibodies representing a broad spectrum of specificities. J. Virol. 77:9993–10003. Kantakamalakul W, et al. 2006. A novel EGFP-CEM-NKr flow cytometric method for measuring antibody dependent cell mediatedcytotoxicity (ADCC) activity in HIV-1 infected individuals. J. Immunol. Methods 315:1–10. Keele BF, et al. 2008. Identification and characterization of transmitted and early founder virus envelopes in primary HIV-1 infection. Proc. Natl. Acad. Sci. U. S. A. 105:7552–7557. Koch M, et al. 2003. Structure-based, targeted deglycosylation of HIV-1 gp120 and effects on neutralization sensitivity and antibody recognition. Virology 313:387– 400. Kottilil S, et al. 2003. Innate immunity in human immunodeficiency virus infection: effect of viremia on natural killer cell function. J. Infect. Dis. 187:1038 –1045. Koup RA, et al. 1989. Antigenic specificity of antibody-dependent cellmediated cytotoxicity directed against human immunodeficiency virus in antibody-positive sera. J. Virol. 63:584 –590. Kwong PD, et al. 2002. HIV-1 evades antibody-mediated neutralization through conformational masking of receptor-binding sites. Nature 420: 678 – 682. Kwong PD, et al. 1998. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393:648 – 659. Labrijn AF, et al. 2003. Access of antibody molecules to the conserved coreceptor binding site on glycoprotein gp120 is sterically restricted on primary human immunodeficiency virus type 1. J. Virol. 77:10557– 10565. Lanier LL, Le AM, Civin CI, Loken MR, Phillips JH. 1986. The relationship of CD16 (Leu-11) and Leu-19 (NKH-1) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes. J. Immunol. 136:4480 – 4486. Larman MG, Sheehan CB, Gardner DK. 2006. Calcium-free vitrification reduces cryoprotectant-induced zona pellucida hardening and increases fertilization rates in mouse oocytes. Reproduction 131:53– 61. Li Y, et al. 1991. Molecular characterization of human immunodeficiency virus type 1 cloned directly from uncultured human brain tissue: identification of replication-competent and -defective viral genomes. J. Virol. 65:3973–3985. Lozada C, et al. 1995. Identification of C1q as the heat-labile serum

November 2012 Volume 86 Number 22

79. 80.

81. 82. 83. 84. 85. 86. 87.

88.

89. 90. 91. 92. 93. 94. 95. 96. 97. 98.

99. 100. 101.

102.

cofactor required for immune complexes to stimulate endothelial expression of the adhesion molecules E-selectin and intercellular and vascular cell adhesion molecules 1. Proc. Natl. Acad. Sci. U. S. A. 92:8378 – 8382. Macian F, et al. 2002. Transcriptional mechanisms underlying lymphocyte tolerance. Cell 109:719 –731. Mascola JR, et al. 1996. Immunization with envelope subunit vaccine products elicits neutralizing antibodies against laboratory-adapted but not primary isolates of human immunodeficiency virus type 1. The National Institute of Allergy and Infectious Diseases AIDS Vaccine Evaluation Group. J. Infect. Dis. 173:340 –348. Mavilio D, et al. 2003. Natural killer cells in HIV-1 infection: dichotomous effects of viremia on inhibitory and activating receptors and their functional correlates. Proc. Natl. Acad. Sci. U. S. A. 100:15011–15016. Mavilio D, et al. 2005. Characterization of CD56-/CD16⫹ natural killer (NK) cells: a highly dysfunctional NK subset expanded in HIV-infected viremic individuals. Proc. Natl. Acad. Sci. U. S. A. 102:2886 –2891. Means RE, Greenough T, Desrosiers RC. 1997. Neutralization sensitivity of cell culture-passaged simian immunodeficiency virus. J. Virol. 71:7895–7902. Miller CJ, et al. 2007. Antiviral antibodies are necessary for control of simian immunodeficiency virus replication. J. Virol. 81:5024 –5035. Mitchell WM, Forti RL, Vogler LB, Lawton AR, Gregg CR. 1983. Spontaneous and interferon resistant natural killer cell anergy in AIDS. AIDS Res. 1:221–229. Montefiori D, et al. 2007. Antibody-based HIV-1 vaccines: recent developments and future directions. PLoS Med. 4:e348. doi:10.1371/ journal.pmed.0040348. Moore JP, et al. 1993. Adaptation of two primary human immunodeficiency virus type 1 isolates to growth in transformed T cell lines correlates with alterations in the responses of their envelope glycoproteins to soluble CD4. AIDS Res. Hum. Retroviruses 9:529 –539. Moore JP, et al. 1995. Primary isolates of human immunodeficiency virus type 1 are relatively resistant to neutralization by monoclonal antibodies to gp120, and their neutralization is not predicted by studies with monomeric gp120. J. Virol. 69:101–109. Moore JP, Sodroski J. 1996. Antibody cross-competition analysis of the human immunodeficiency virus type 1 gp120 exterior envelope glycoprotein. J. Virol. 70:1863–1872. Morris MA, Ley K. 2004. Trafficking of natural killer cells. Curr. Mol. Med. 4:431– 438. Myers G, Lenroot R. 1992. HIV glycosylation: what does it portend? AIDS Res. Hum. Retroviruses 8:1459 –1460. Myszka DG, et al. 2000. Energetics of the HIV gp120-CD4 binding reaction. Proc. Natl. Acad. Sci. U. S. A. 97:9026 –9031. Nataf S, Davoust N, Ames RS, Barnum SR. 1999. Human T cells express the C5a receptor and are chemoattracted to C5a. J. Immunol. 162:4018 – 4023. Nimmerjahn F, Ravetch JV. 2008. Fcgamma receptors as regulators of immune responses. Nat. Rev. Immunol. 8:34 – 47. Nitayaphan S, et al. 2004. Safety and immunogenicity of an HIV subtype B and E prime-boost vaccine combination in HIV-negative Thai adults. J. Infect. Dis. 190:702–706. O’Doherty U, Swiggard WJ, Malim MH. 2000. Human immunodeficiency virus type 1 spinoculation enhances infection through virus binding. J. Virol. 74:10074 –10080. Oh-hora, M, Rao A. 2008. Calcium signaling in lymphocytes. Curr. Opin. Immunol. 20:250 –258. Oliva A, et al. 1998. Natural killer cells from human immunodeficiency virus (HIV)-infected individuals are an important source of CCchemokines and suppress HIV-1 entry and replication in vitro. J. Clin. Invest. 102:223–231. Ottonello L, et al. 1999. rC5a directs the in vitro migration of human memory and naive tonsillar B lymphocytes: implications for B cell trafficking in secondary lymphoid tissues. J. Immunol. 162:6510 – 6517. Parren PW, Burton DR, Sattentau QJ. 1997. HIV-1 antibody— debris or virion? Nat. Med. 3:366 –367. Patterson LJ, et al. 2008. Replicating adenovirus HIV/SIV recombinant priming alone or in combination with a gp140 protein boost results in significant control of viremia following a SHIV89.6P challenge in Mamu-A*01 negative rhesus macaques. Virology 374:322–337. Pikora C, Wittish C, Desrosiers RC. 2005. Identification of two Nlinked glycosylation sites within the core of the simian immunodefi-

jvi.asm.org 12051

Alpert et al.

103.

104. 105. 106. 107. 108. 109.

110. 111.

112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124.

ciency virus glycoprotein whose removal enhances sensitivity to soluble CD4. J. Virol. 79:12575–12583. Pitisuttithum P, et al. 2006. Randomized, double-blind, placebocontrolled efficacy trial of a bivalent recombinant glycoprotein 120 HIV-1 vaccine among injection drug users in Bangkok, Thailand. J. Infect. Dis. 194:1661–1671. Pollara J, et al. 2011. High-throughput quantitative analysis of HIV-1 and SIV-specific ADCC-mediating antibody responses. Cytometry A 79: 603– 612. Pugach P, et al. 2004. The prolonged culture of human immunodeficiency virus type 1 in primary lymphocytes increases its sensitivity to neutralization by soluble CD4. Virology 321:8 –22. Putkonen P, et al. 1993. Whole inactivated SIV vaccine grown on human cells fails to protect against homologous SIV grown on simian cells. J. Med. Primatol. 22:100 –103. Reeves RK, et al. 2010. CD16- natural killer cells: enrichment in mucosal and secondary lymphoid tissues and altered function during chronic SIV infection. Blood 115:4439 – 4446. Reitter JN, Means RE, Desrosiers RC. 1998. A role for carbohydrates in immune evasion in AIDS. Nat. Med. 4:679 – 684. Rerks-Ngarm S, Brown AE, Khamboonruang C, Thongcharoen P, Kunasol P. 2006. HIV/AIDS preventive vaccine ‘prime-boost’ phase III trial: foundations and initial lessons learned from Thailand. AIDS 20: 1471–1479. Rerks-Ngarm S, et al. 2009. Vaccination with ALVAC and AIDSVAX to prevent HIV-1 infection in Thailand. N. Engl. J. Med. 361:2209 –2220. Roben P, et al. 1994. Recognition properties of a panel of human recombinant Fab fragments to the CD4 binding site of gp120 that show differing abilities to neutralize human immunodeficiency virus type 1. J. Virol. 68:4821– 4828. Rook AH, et al. 1987. Sera from HTLV-III/LAV antibody-positive individuals mediate antibody-dependent cellular cytotoxicity against HTLV-III/LAV-infected T cells. J. Immunol. 138:1064 –1067. Rot A, von Andrian UH. 2004. Chemokines in innate and adaptive host defense: basic chemokinese grammar for immune cells. Annu. Rev. Immunol. 22:891–928. Salazar-Gonzalez JF, et al. 2009. Genetic identity, biological phenotype, and evolutionary pathways of transmitted/founder viruses in acute and early HIV-1 infection. J. Exp. Med. 206:1273–1289. Sattentau QJ, Moore JP. 1995. Human immunodeficiency virus type 1 neutralization is determined by epitope exposure on the gp120 oligomer. J. Exp. Med. 182:185–196. Scheid JF, et al. 2009. Broad diversity of neutralizing antibodies isolated from memory B cells in HIV-infected individuals. Nature 458:636 – 640. Schwartz RH. 2003. T cell anergy. Annu. Rev. Immunol. 21:305–334. Seaman MS, et al. 2010. Tiered categorization of a diverse panel of HIV-1 Env pseudoviruses for assessment of neutralizing antibodies. J. Virol. 84:1439 –1452. Shen L, et al. 2008. Dose-response curve slope sets class-specific limits on inhibitory potential of anti-HIV drugs. Nat. Med. 14:762–766. Spear GT, Landay AL, Sullivan BL, Dittel B, Lint TF. 1990. Activation of complement on the surface of cells infected by human immunodeficiency virus. J. Immunol. 144:1490 –1496. Spear GT, Sullivan BL, Landay AL, Lint TF. 1990. Neutralization of human immunodeficiency virus type 1 by complement occurs by viral lysis. J. Virol. 64:5869 –5873. Stamatatos L, Morris L, Burton DR, Mascola JR. 2009. Neutralizing antibodies generated during natural HIV-1 infection: good news for an HIV-1 vaccine? Nat. Med. 15:866 – 870. Stott EJ. 1991. Anti-cell antibody in macaques. Nature 353:393. Sun Y, et al. 2011. Antibody-dependent cell-mediated cytotoxicity in simian immunodeficiency virus-infected rhesus monkeys. J. Virol. 85: 6906 – 6912.

12052

jvi.asm.org

125. Tanneau F, et al. 1990. Primary cytotoxicity against the envelope glycoprotein of human immunodeficiency virus-1: evidence for antibodydependent cellular cytotoxicity in vivo. J. Infect. Dis. 162:837– 843. 126. Tarazona R, et al. 2002. Selective depletion of CD56(dim) NK cell subsets and maintenance of CD56(bright) NK cells in treatment-naive HIV-1-seropositive individuals. J. Clin. Immunol. 22:176 –183. 127. Thongcharoen P, et al. 2007. A phase 1/2 comparative vaccine trial of the safety and immunogenicity of a CRF01_AE (subtype E) candidate vaccine: ALVAC-HIV (vCP1521) prime with oligomeric gp160 (92TH023/ LAI-DID) or bivalent gp120 (CM235/SF2) boost. J. Acquir. Immune Defic. Syndr. 46:48 –55. 128. Trkola A, et al. 1996. CD4-dependent, antibody-sensitive interactions between HIV-1 and its co-receptor CCR-5. Nature 384:184 –187. 129. Trkola A, Matthews J, Gordon C, Ketas T, Moore JP. 1999. A cell line-based neutralization assay for primary human immunodeficiency virus type 1 isolates that use either the CCR5 or the CXCR4 coreceptor. J. Virol. 73:8966 – 8974. 130. Tyler DS, et al. 1989. GP120 specific cellular cytotoxicity in HIV-1 seropositive individuals. Evidence for circulating CD16⫹ effector cells armed in vivo with cytophilic antibody. J. Immunol. 142:1177–1182. 131. von Andrian UH, Mackay CR. 2000. T-cell function and migration. Two sides of the same coin. N. Engl. J. Med. 343:1020 –1034. 132. Wei X, et al. 2002. Emergence of resistant human immunodeficiency virus type 1 in patients receiving fusion inhibitor (T-20) monotherapy. Antimicrob. Agents Chemother. 46:1896 –1905. 133. Wei X, et al. 2003. Antibody neutralization and escape by HIV-1. Nature 422:307–312. 134. Wrin T, Loh TP, Vennari JC, Schuitemaker H, Nunberg JH. 1995. Adaptation to persistent growth in the H9 cell line renders a primary isolate of human immunodeficiency virus type 1 sensitive to neutralization by vaccine sera. J. Virol. 69:39 – 48. 135. Wu L, et al. 1996. CD4-induced interaction of primary HIV-1 gp120 glycoproteins with the chemokine receptor CCR-5. Nature 384:179 –183. 136. Wyatt R, et al. 1997. Analysis of the interaction of the human immunodeficiency virus type 1 gp120 envelope glycoprotein with the gp41 transmembrane glycoprotein. J. Virol. 71:9722–9731. 137. Wyatt R, et al. 1998. The antigenic structure of the HIV gp120 envelope glycoprotein. Nature 393:705–711. 138. Wyatt R, Sodroski J. 1998. The HIV-1 envelope glycoproteins: fusogens, antigens, and immunogens. Science 280:1884 –1888. 139. Xiao P, et al. 2010. Multiple vaccine-elicited nonneutralizing antienvelope antibody activities contribute to protective efficacy by reducing both acute and chronic viremia following simian/human immunodeficiency virus SHIV89.6P challenge in rhesus macaques. J. Virol. 84:7161– 7173. 140. Yagita M, et al. 2000. A novel natural killer cell line (KHYG-1) from a patient with aggressive natural killer cell leukemia carrying a p53 point mutation. Leukemia 14:922–930. 141. Yang X, et al. 2000. Modifications that stabilize human immunodeficiency virus envelope glycoprotein trimers in solution. J. Virol. 74:4746 – 4754. 142. Yang X, Kurteva S, Lee S, Sodroski J. 2005. Stoichiometry of antibody neutralization of human immunodeficiency virus type 1. J. Virol. 79: 3500 –3508. 143. Yang X, Kurteva S, Ren X, Lee S, Sodroski J. 2005. Stoichiometry of envelope glycoprotein trimers in the entry of human immunodeficiency virus type 1. J. Virol. 79:12132–12147. 144. Yuste E, et al. 2008. Glycosylation of gp41 of simian immunodeficiency virus shields epitopes that can be targets for neutralizing antibodies. J. Virol. 82:12472–12486. 145. Zwick MB, Burton DR. 2007. HIV-1 neutralization: mechanisms and relevance to vaccine design. Curr. HIV Res. 5:608 – 624.

Journal of Virology