Previous Article | Next Article ![]()
Journal of Virology, September 2006, p. 9217-9225, Vol. 80, No. 18
0022-538X/06/$08.00+0 doi:10.1128/JVI.02746-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Division of Infectious Diseases, Department of Medicine, University of California, Irvine School of Medicine, Irvine, California 92697-4028,1 Department of Medicine, Infectious Diseases Division, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15261,2 California National Primate Research Center, University of California, Davis, Davis, California 956163
Received 30 December 2005/ Accepted 30 June 2006
|
|
|---|
|
|
|---|
We have found that nonneutralizing antibodies appear very early after human immunodeficiency virus type 1 (HIV-1) infection in humans and that the antiviral activity of the antibodies correlates with the fall in viremia during acute infection (9). These antibodies inhibit HIV-1 yield from infected CD4 lymphocytes but only in the presence of NK effector cells. Since virus inhibition was due to both cytolytic and noncytolytic mechanisms, we distinguished this antibody function from ADCC and termed it antibody-dependent cell-mediated virus inhibition (ADCVI). ADCVI is closely related to ADCC, since the components (i.e., antibody and effector cells acting on infected target cells) are the same; however, ADCVI is a measure of the combined ability of antibody and effector cells to limit virus infection rather than to cause target cell death.
In the current study, we turned our attention to measuring ADCVI activity in the rhesus macaque-SIV model. We determined if SIV-specific ADCVI antibody was present in plasma of infected animals or in serum that prevented infection of newborn macaques. In addition, we measured the ADCVI activity of three macaque MAbs. Finally, we determined if rhesus macaque peripheral blood mononuclear cells (rhPBMCs) were able to function as ADCVI effector cells.
|
|
|---|
Animals.
All rhesus macaques (Macaca mulatta) were housed at the California National Primate Research Center in accordance with American Association for Accreditation of Laboratory Animal Care standards and with adherence to the "Guide for Care and Use of Laboratory Animals" (17). Two groups of animals were used for collection of plasma or PBMCs. The first group consisted of SIV-negative healthy adult macaques that are part of the California National Primate Research Center blood donor pool. The second group consisted of animals that were infected with uncloned SIVmac251 (animals 29276, 33088, and 33091) or RT-SHIV (a chimeric virus containing the reverse transcriptase of HIV-1, animal 30577) (37). These animals were started on prolonged tenofovir treatment between 2 and 20 weeks after infection; although their virus developed reduced in vitro susceptibility to tenofovir (associated with development of the K65R mutation in the reverse transcriptase gene), viremia declined and remained undetectable (<125 RNA copies/ml plasma) for
3 years (19; unpublished data). At the time of sample collection for the present study, their maintenance tenofovir regimen was 0.8 to 10 mg/kg of body weight per day, given subcutaneously. For blood collections, SIV-infected animals were immobilized with 10 mg/kg ketamine-HCl (Parke-Davis, Morris Plains, NJ), administered intramuscularly.
Virus. In all assays, uncloned SIVmac251 was used to infect target cells. This is a further passage in rhPBMCs of the SIVmac251 stock used in an oral challenge of neonatal rhesus macaques following passive infusion of SIV hyperimmune serum (SIV-HIS) and in studies of the effect of tenofovir on viremia control (42, 43). This SIVmac251 is highly virulent in vivo, is difficult to control immunologically, and often leads to persistent viremia and a relatively rapid disease course (40-43).
Serum and plasma. As a source of antibody in ADCVI assays, heat-inactivated serum or plasma samples were used. Pooled serum (SIV-HIS) was obtained from two juvenile and four adult rhesus macaques who successfully controlled viremia after immunization with SIVmac1A11 and challenge with SIVmac251. SIV-HIS was shown to prevent infection in six of six neonatal rhesus macaques following oral challenge with uncloned SIVmac251 (42). Plasma was also obtained from adult rhesus macaques who had been treated with tenofovir as described above.
Anti-SIV MAbs. MAbs 310A, 311.E, and C26 were isolated from rhPBMCs as described previously (7, 30).
IgG and F(ab')2. Immunoglobulin G (IgG) was separated from serum after binding to protein G columns (Amersham Pharmacia, Piscataway, NJ), elution from the columns, and pH neutralization according to the manufacturer's instructions. The purity of the IgG was assessed by polyacrylamide gel electrophoresis, and total protein content was determined by the Bradford method with a human IgG standard. For some experiments, F(ab')2 was made from SIV-HIS or control IgG samples by pepsin digestion, followed by purification and concentration with Amicon centrifugal concentrators (Millipore, Bedford, MA.). F(ab')2 purity was verified by polyacrylamide gel electrophoresis.
Cells. Target cells infected with SIVmac251 for ADCVI assays included CEMx174 cells and rhPBMCs. rhPBMCs were separated from whole blood by Ficoll-Hypaque centrifugation, depleted of CD8+ cells (see below), and stimulated with Staphylococcus endotoxin A (0.5 µg/ml) and recombinant human interleukin 2 (Proleukin; Chiron, Emeryville, CA) for 3 days in medium (RPMI 1640 supplemented with 10% fetal bovine serum, penicillin, streptomycin, and L-glutamine) prior to infection. Effector cells consisted of human PBMCs (huPBMCs) or rhPBMCs. huPBMCs were separated from whole blood by Ficoll-Hypaque centrifugation and used the same day. rhPBMCs were obtained from healthy animals and were used as effector cells within 48 h after separation without freezing.
CD8+ and CD14+ cell depletions. rhPBMCs used for target cells were depleted of CD8+ cells using Dynabeads CD8 (Dynal Biotech, Oslo, Norway) according to the manufacturer's instructions. In some experiments, rhPBMC effector cells were also depleted of CD8+ cells using the same method.
CD14+ cells were depleted from rhPBMC effector cells using CD14-conjugated magnetic beads (StemCell Technologies, Vancouver, British Columbia). In some experiments, the positively selected CD14+ cells were used as effector cells.
Cytometry. rhPBMCs or CD8+-cell-depleted rhPBMCs were analyzed using four-color flow cytometry. A first aliquot was stained with peridinin chlorophyll protein-conjugated anti-human CD8 (clone SK1; Becton Dickinson Immunocytometry, Inc., San Jose, Calif.), fluorescein isothiocyanate-conjugated anti-human CD3 (clone SP34; Becton Dickinson Pharmingen), phycoerythrin-conjugated anti-human CD4 (clone M-T477; Becton Dickinson Pharmingen), and allophycocyanin-conjugated anti-human CD20 (clone L27; Becton Dickinson). In some experiments, a second aliquot was stained with fluorescein isothiocyanate-conjugated anti-human CD16 MAb (clone 3G8; Becton Dickinson Pharmingen), allophycocyanin-conjugated anti-huCD8 MAb (clone SK.1; Becton Dickinson), peridinin chlorophyll protein-conjugated anti-CD3 (clone SP34), and phycoerythrin-conjugated anti-CD14 (clone M5E2; Becton Dickinson Pharmingen). Red blood cells were lysed, and the samples were fixed in paraformaldehyde by the Coulter Q-prep system (Coulter Corporation, Hialeah, Fla.). Flow cytometry was performed on a FACSCalibur flow cytometer (Becton Dickinson). Lymphocytes were gated by forward and side light scatter and were analyzed with Cellquest software (Becton Dickinson).
ADCVI assay. The ADCVI assay was based on methods described previously using human cells and antibody (9, 10). CEMx174 target cells were infected with uncloned SIVmac251 at a multiplicity of infection of 0.01. After adsorption for 1 h, cells were washed and incubated in 5% CO2 at 37°C for 48 h in medium. CD8+-cell-depleted rhPBMC target cells were first stimulated with Staphylococcus endotoxin A and interleukin 2 for 72 h, washed, infected with uncloned SIVmac251 (multiplicity of infection of 0.01), washed after 1 h, and incubated in 5% CO2 at 37°C for 48 h in medium. Just prior to use in the ADCVI assay, target cells were washed, and 5 x 104 were added to 96-well round-bottom microtiter plates. Various dilutions of serum, plasma, IgG, or F(ab')2 were added to target cells along with effector cells at various effector:target (E:T) ratios. Effector cells were either huPBMCs, rhPBMCs, CD8+-cell-depleted rhPBMCs, CD14+-cell-depleted rhPBMCs, or positively selected rhCD14+ cells. After 5 or 7 days of incubation at 37°C in 5% CO2, supernatant fluid was collected and assayed for p27 by enzyme-linked immunosorbent assay (ELISA) (Zeptometrix, Buffalo, NY). Virus inhibition due to ADCVI was calculated as follows: % inhibition = 100[1 ([p27p]/[p27n])], where [p27p] and [p27n] are the concentrations of p27 in supernatant fluid from wells containing a source of SIV-positive or -negative antibody, respectively.
|
|
|---|
R)-bearing NK cells, in addition to cytotoxic T lymphocytes (CTLs) (19). The animals treated with tenofovir developed binding antibody responses to SIV as well as low neutralizing antibody titers using CEM-CCR5 cells and rhPBMC-grown SIVmac251 (40, 43; unpublished data). Therefore, we sought to determine whether an interaction between antibody and Fc
R-bearing cells could underlie viremia control. Plasma samples from two animals were tested for ADCVI activity using CEMx174 cells infected for 48 h with SIVmac251 as target cells and fresh huPBMCs as effector cells; plasma pooled from three uninfected animals was used as a source of SIV-negative antibody. Potent ADCVI activity was demonstrated in the plasma of both SIV-infected animals (Fig. 1). Note that plasma was left on the infected target cells throughout the assay period, which would allow antibody to neutralize cell-free virus emerging from the infected cells. Consistent with that possibility, plasma did inhibit virus in the absence of effector cells. However, virus inhibition in the presence of effector cells occurred at dilutions at least 10-fold greater than without effector cells (Fig. 1). Plasma from two other animals showed a similar increase in antiviral activity in the presence of huPBMCs effector cells; in these cases, antibody inhibition was measured relative to a medium control rather than to SIV-negative plasma (data not shown).
![]() View larger version (6K): [in a new window] |
FIG. 1. Inhibition of viral yield from SIVmac251-infected cells by plasma from tenofovir-treated, SIV-infected rhesus macaques is enhanced in the presence of huPBMC effector cells. CEMx174 cells were infected with uncloned SIVmac251 for 48 h, washed, and added to huPBMC effector cells (effector:target = 10:1) at indicated dilutions of plasma from SIV-infected macaque 29276 (A) or 30577 (B) or an uninfected animal. Seven days later, p27 was measured in supernatant fluid and percent inhibition was calculated as described in Methods. Each result represents the mean from two separate experiments, each run in duplicate, and error bars represent standard errors.
|
rhPBMCs serve as ADCVI effector cells. If ADCVI antibody is involved in control of viremia, functioning ADCVI effector cells must be present in vivo. We next tested the ability of rhPBMCs to serve as effector cells in the ADCVI assay. In the first set of experiments, fresh rhPBMCs from three SIV-negative animals were used with SIVmac251-infected CEMx174 target cells at E:T ratios of 10:1, 1:1, and 1:10. Plasma (1:200) from one of the animals controlling viremia with tenofovir (29276) was used as the antibody source. The rhPBMCs from all three animals mediated ADCVI activity, although the effector-cell function varied between the animals (Fig. 2). To our knowledge this is the first demonstration that rhPBMCs can act as effector cells in the presence of antibody to reduce the yield of SIV from infected cells.
![]() View larger version (17K): [in a new window] |
FIG. 2. rhPBMCs inhibit viral yield from SIVmac251-infected CEMx174 target cells in the presence of plasma from SIV-infected rhesus macaques. SIV-infected CEMx174 target cells (see Methods) were incubated with plasma (1:200) from an SIV-infected or uninfected animal and fresh rhPBMCs from healthy animals at the indicated E:T ratio. Seven days later, p27 was measured in the supernatant fluid and percent inhibition was calculated as described. Data represent means for duplicate samples.
|
![]() View larger version (9K): [in a new window] |
FIG. 3. rhPBMC effector cells inhibit viral yield from SIVmac251-infected, autologous CD8+-cell-depleted target cells. rhPBMCs were obtained from the five healthy animals indicated and either depleted of CD8+ cells for use as target cells or left unfractionated for use as effector cells. Plasma from a single SIV-infected animal (or an uninfected animal) was used as a source of antibody at a dilution of 1:800. Target cells were infected as described in Methods, and the autologous effector cells were added at the indicated E:T ratio. p27 was measured by ELISA 5 days after the addition of plasma and effector cells.
|
chain have established a role for CD8+ cells in the control of lentivirus infection (19, 21, 33, 41). However, both rhesus NK cells and CTLs express CD8, and rhesus NK cells express Fc
Rs (2, 44). Thus, CD8+-cell depletion leads to a reduction in NK cell numbers and could reduce ADCVI effector activity (41). We directly tested the latter possibility by measuring ADCVI effector cell function in rhPBMCs from five animals before and after in vitro CD8+-cell depletion, which removed >97% of CD8+ cells (Table 1). Using autologous combinations of rhesus effector and target cells, CD8+-cell depletion resulted in a reduction of PBMC effector cell function from two of the five animals (26687 and 26268); in two other animals (27962 and 29311), there was a smaller decrease in effector cell function (Fig. 4). |
View this table: [in a new window] |
TABLE 1. Flow cytometry analysis of rhPBMC effector cells before and after CD8+-cell depletion
|
![]() View larger version (11K): [in a new window] |
FIG. 4. CD8+ cells from some animals have ADCVI effector cell function. CD8+-cell-depleted rhPBMC target cells infected with SIVmac251 for 48 h were mixed with effector cells consisting of either fresh, unfractionated rhPBMCs or CD8+-cell-depleted rhPBMCs at the E:T ratios indicated. Plasma from SIV-infected or uninfected animals was used as a source of antibody. Five days after the addition of effector cells and plasma, p27 was measured in the supernatant fluid by ELISA.
|
Rs (6), and human monocytes have been shown to mediate ADCC (14, 18). We therefore determined the ability of rhPBMCs depleted of cells bearing the monocyte marker CD14 to inhibit virus yield from autologous CD8+-cell-depleted rhPBMC target cells. CD14+-cell depletion resulted in less ADCVI activity than did either the unfractionated rhPBMCs or the CD8+-cell-depleted effector cells (Fig. 5A, compare with animals 26687 and 29311 in Fig. 4). Furthermore, CD14+ cells positively selected from rhPBMCs resulted in 89 to 97% inhibition at an E:T ratio of 1:1 with three of three animals studied (Fig. 5B). Thus, both monocytes and to a lesser extent CD8+ cells contribute to the ADCVI effector cell function of rhPBMCs. In some assays, rhPBMCs contained a substantial number of neutrophils (data not shown), and it is possible that those or other cells also contribute to ADCVI activity.
![]() View larger version (13K): [in a new window] |
FIG. 5. CD14+ are ADCVI effector cells in rhPBMCs. rhPBMCs from uninfected animals, depleted of CD8+ cells in vitro and infected with SIVmac251 for 48 h, were used as target cells. Target cells were mixed with effector cells consisting of either fresh, unfractionated rhPBMCs or CD14+ cell-depleted rhPBMCs at the E:T ratios indicated (A) or with positively selected CD14+ cells at an E:T ratio of 1:1 (B). Plasma from SIV-infected or uninfected animals was used as a source of antibody; in addition, SIV-IgG was used with CD14+ effector cells. Five days after the addition of effector cells and plasma, p27 was measured in the supernatant fluid by ELISA.
|
![]() View larger version (10K): [in a new window] |
FIG. 6. Pooled serum from SIVmac1A11-immunized adult rhesus macaques (SIV-HIS) that prevents SIVmac251 infection of newborn macaques inhibits SIVmac251 in the presence of huPBMC effector cells. CEMx174 cells were infected with SIVmac251 for 48 h and added to huPBMCs (effector:target ratio = 10:1) at the indicated dilutions of SIV-HIS (or SIV-negative serum). p27 was measured in supernatant fluid 5 days later. Data represent the means from two or three separate experiments, each done in duplicate.
|
R interaction was required for inhibition, F(ab')2 was made from the IgG. There was little virus inhibition by F(ab')2 either in the presence or absence of effector cells (Fig. 7). These results indicate that antibodies with little inhibitory effect on cell-free virus may, in the presence of effector cells, be very potent inhibitors of viral yield from infected cells. Furthermore, inhibition is largely dependent on intact IgG.
![]() View larger version (16K): [in a new window] |
FIG. 7. The IgG fraction of SIV-HIS (SIV-IgG) inhibits SIVmac251 yield from infected CEMx174 cells in the presence of huPBMC effector cells, whereas F(ab')2 has little antiviral activity. Virus inhibition was measured as described in Materials and Methods.
|
![]() View larger version (10K): [in a new window] |
FIG. 8. Nonneutralizing rhesus MAbs (20 µg/ml) inhibit SIVmac251 in the presence of huPBMC effector cells. Virus inhibition was measured as described in Methods using SIVmac251-infected CEMx174 cells as target cells (effector:target ratio = 10:1).
|
|
|
|---|
To our knowledge, this study is the first demonstration of ADCVI in nonhuman primates. ADCVI has been described previously for measles virus and for HIV-1 using human antibody and human target and effector cells (9, 10). As has been observed for human systems with HIV-1, ADCVI in the SIV/macaque model requires intact antibody, rather than F(ab')2 and effector cells bearing Fc receptors; thus, there is a reliance on Fc-FcR interactions. ADCVI against HIV-1 results from a combination of target cell death and the production of ß-chemokines, both due to antibody cross-linking of Fc receptors on NK cells (or monocyte/macrophages) (9). Although we have not directly determined the mechanisms of virus inhibition in the SIV/macaque model, by analogy with HIV-1, it is likely that both cytolytic and noncytolytic antiviral factors also play a role in ADCVI against SIV. With regard to cytolytic mechanisms, it has recently been demonstrated that vaccine-induced rhesus macaque antibody in the presence of human effector cells is capable of mediating the death of target cells (measured by 51chromium release) adsorbed with SIV (13).
We have shown that plasma from SIVmac251-infected macaques who control viremia has potent ADCVI activity (Fig. 1). These animals have been treated long term with tenofovir. However, their virus isolates are resistant to tenofovir in vitro, and in vivo depletion of CD8+ cells results in marked increases in viremia levels, indicating that cellular immunity plays a beneficial role (41). Since ADCVI requires effector cells, some of which may be CD8+, it is possible that ADCVI is a component of viremia control in these animals.
Our data indicate that CD14+ cells are an important ADCVI effector cell in rhPBMCs. As in humans, CD14 is a marker of monocytes and macrophages, and in both humans and rhesus macaques, such cells bear Fc receptors (6). In humans, monocytes or macrophages mediate ADCC, and Fc receptor cross-linking can lead to the production of ß-chemokines (8, 14, 18, 20, 26). In addition, CD14+ cells are phagocytic, and it is possible that some of the ADCVI activity might be due to Fc receptor-mediated phagocytosis of immune complexes formed from virus released from the infected cells (16).
The role of rhesus CD8+ cells in mediating ADCVI is less clear. Since many rhesus NK cells express both CD8 and Fc
Rs (CD16) (2, 44), we fully expected that CD8+-cell depletion would remove a substantial portion of the ADCVI activity in rhPBMCs. However, only some animals achieved a reduction in ADCVI activity, despite a depletion technique that was at least 97% effective in eliminating CD8+ cells (Table 1). It is noteworthy that CD8+-cell depletion resulted in the relative enrichment of CD14+ cells, which we have demonstrated are ADCVI effector cells, and of CD8 CD16+ cells, which are likely to be effector cells as well (Table 1). Thus, the actual increase in other effector cell populations may have offset reductions in ADCVI due to CD8+-cell depletion. It remains unclear why CD8+-cell depletion had a more substantial effect on animal 26268 than on other animals. Although the percentage of CD3 CD8+ cells prior to depletion was highest for animal 26268 (Table 1), there was no overall correlation between predepletion CD3 CD8+-cell percentages and reductions in ADCVI effector cell activity. In any case, these results suggest that in vivo CD8+-cell depletion may have some impact on in vivo ADCVI activity. Since in vivo CD8+-cell depletion has demonstrated a role for CD8+ cells in controlling SIV in infected monkeys (19, 21, 33, 41) and since our data show that infected animals make ADCVI antibody, it is possible that viremia control is dependent, at least in part, on ADCVI activity.
We have also shown that rhesus MAbs directed against distinct epitopes on SIV gp120 can mediate ADCVI. Interestingly, one of the MAbs (3.10A) with ADCVI activity against SIVmac251 had no neutralizing activity either against the strain used to elicit the MAb (SIVmac17E) or against SIVmac251 (7; unpublished data). The other MAb with anti-SIVmac251 ADCVI activity, 3.11E, was able to neutralize SIVmac17E but not SIVmac251 (7; unpublished data). Finally, the third MAb tested, C26, neutralized SIVmac17E, did not neutralize SIVmac251 and did not mediate ADCVI against SIVmac251 (22; unpublished data). All of the MAbs bind to SIVmac251 gp120 (unpublished data). Thus, rhesus anti-SIV envelope MAbs may be discordant with respect to their biological functions, a finding similar to that noted for human anti-HIV-1 MAbs (1, 11).
The results with rhesus anti-SIV MAbs allow us to define, for the first time, ADCVI epitopes (7). In the case of MAb 3.10A, binding has been mapped to a linear epitope in the SIV V1 loop (amino acids 141 to 160) (7). MAb 3.11E binds to a conformational epitope, as yet not clearly defined (22). The non-ADCVI MAb, C26, also binds to a conformational epitope, possibly involving the SIV V4 loop (7). We do not know why C26, which binds to SIVmac251 and neutralizes SIVmac17E, does not mediate ADCVI against SIVmac251. Based on preliminary data, all three MAbs appear to be IgG1 (data not shown). However, it is possible that there is poor avidity between the Fc segment of MAb C26 and Fc receptors, related to alterations in Fc that result from antigen binding or to the presence of high fucose content (25).
Finally, our data indicate that SIV-HIS has potent anti-SIV activity, as long as effector cells and intact antibody, rather than F(ab')2, are present. Given the ability of SIV-HIS to protect neonatal macaques from oral challenge with SIVmac251 and the lack of ability of SIV-HIS to neutralize cell-free SIVmac251 (42), it is quite plausible that the ADCVI activity is responsible for the protective effect of SIV-HIS. This possibility is strengthened by the finding that SIV-HIS mediates ADCVI with rhesus PBMC effector and target cells, which reasonably simulates in vivo conditions. For logistical reasons, we were unable to determine the ADCVI activity of SIV-HIS using PBMCs from neonatal macaques, and it remains a possibility that such cells behave differently than the PBMCs from adult macaques used in our studies. Nonetheless, our finding of an association between ADCVI activity and protection against SIV infection, together with previous studies indicating both an in vitro and in vivo impact of ADCVI, ADCC, or macrophage phagocytosis on lentivirus infections, underscores the potentially critical influence of Fc-Fc receptor interactions on the antiviral activity of antibodies (3, 4, 9, 13).
In summary, we have shown, for the first time, that polyclonal and monoclonal antibody and effector cells from rhesus macaques are capable of mediating ADCVI against SIV. The finding of potent antiviral activity in poorly neutralizing serum that prevents SIV infection is consistent with a role for ADCVI in vivo and suggests that ADCVI might be an immunological correlate of protection in the rhesus macaque model.
|
|
|---|
3-vaccinated rhesus macaques. J. Virol. 79:8131-8141.This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»