Transient CD4+ T Cell Depletion Results in Delayed Development of Functional Vaccine-Elicited Antibody Responses

ABSTRACT We have recently demonstrated that CD4+ T cell help is required at the time of adenovirus (Ad) vector immunization for the development of functional CD8+ T cell responses, but the temporal requirement for CD4+ T cell help for the induction of antibody responses remains unclear. Here we demonstrate that induction of antibody responses in C57BL/6 mice can occur at a time displaced from the time of Ad vector immunization by depletion of CD4+ T cells. Transient depletion of CD4+ T cells at the time of immunization delays the development of antigen-specific antibody responses but does not permanently impair their development or induce tolerance against the transgene. Upon CD4+ T cell recovery, transgene-specific serum IgG antibody titers develop and reach a concentration equivalent to that in undepleted control animals. These delayed antibody responses exhibit no functional defects with regard to isotype, functional avidity, expansion after boosting immunization, or the capacity to neutralize a simian immunodeficiency virus (SIV) Env-expressing pseudovirus. The development of this delayed transgene-specific antibody response is temporally linked to the expansion of de novo antigen-specific CD4+ T cell responses, which develop after transient depletion of CD4+ T cells. These data demonstrate that functional vaccine-elicited antibody responses can be induced even if CD4+ T cell help is provided at a time markedly separated from the time of vaccination. IMPORTANCE CD4+ T cells have a critical role in providing positive help signals to B cells, which promote robust antibody responses. The paradigm is that helper signals must be provided immediately upon antigen exposure, and their absence results in tolerance against the antigen. Here we demonstrate that, in contrast to the current model that the absence of CD4+ T cell help at priming results in long-term antibody nonresponsiveness, antibody responses can be induced by adenovirus vector immunization or alum-adjuvanted protein immunization even if CD4+ T cell help is not provided until >1 month after immunization. These data demonstrate that the time when CD4+ T cell help signals must be provided is more dynamic and flexible than previously appreciated. These data suggest that augmentation of CD4+ T cell helper function even after the time of vaccination can enhance vaccine-elicited antibody responses and thereby potentially enhance the immunogenicity of vaccines in immunocompromised individuals.

C D4 ϩ T cells, also termed T helper cells, are critical positive regulators of antibody and cytotoxic CD8 ϩ T cell responses (1). In the context of antibody induction, the primary function of CD4 ϩ T cells is to promote and maintain B cell germinal center responses (2,3). The current model is that CD4 ϩ T cell help must be provided at the time of antigen exposure (by either infection, immunization, or exposure to self-antigen), as an absence of CD4 ϩ T cell help at the time of priming results in tolerance (3)(4)(5)(6)(7)(8)(9). For productive antibody responses to develop, engagement of the CD40 signaling pathway on B cells by CD4 ϩ T cells must occur (5,10,11). Studies using model antigens have demonstrated that the proper development of germinal center responses is a dynamic process where CD4 ϩ T cell help, via CD40, is provided for several days (8,(12)(13)(14). In the absence of these positive signals from CD4 ϩ T cells, tolerance is induced because activated B cells are rapidly deleted (2,10,15). However, in some, but not all, cases, an antibody response can be induced by readministration of the antigen after recovery of the CD4 ϩ T cell population (5,8,11,13). Thus, our current understanding is that CD4 ϩ T cell help is required immediately at the time of antigen exposure for the development of functional antibody responses.
Adenovirus (Ad) vectors have primarily been pursued as vac-cine platforms due to their ability to induce strong CD8 ϩ T cell responses and antibody responses (16)(17)(18)(19)(20)(21)(22). We have recently described that following Ad vector immunization of mice, CD4 ϩ T cell help is required immediately upon antigen exposure to prevent immediate and irreversible dysfunction of vaccine-elicited Additionally, this CD4 ϩ T cell help is required for a month postimmunization to properly induce CD8 ϩ T cell responses (23). However, a role for CD4 ϩ T cells in regulating transgene-specific antibody responses following Ad vector immunization has not been previously demonstrated. Thus, we sought to identify a role for CD4 ϩ T cells in the promotion of transgene-specific antibody responses following Ad vector immunization of C57BL/6 mice. Furthermore, we sought to determine whether Ad vector vaccine-elicited antibody responses are also immediately and irreversibly dysfunctional if CD4 ϩ T cell help is not provided at the time of immunization. In this study, we identified that following Ad vector immunization, CD4 ϩ T cell help is required for between 10 and 14 days after immunization to induce optimal antigen-specific antibody titers. Unexpectedly, we also observed that CD4 ϩ T cell depletion prior to immunization does not result in a permanent ablation of antigen-specific antibody responses. Instead, the induction of antibody responses is simply delayed until the time at which the CD4 ϩ T cells begin to recover, and these responses developed without the readministration of antigen. These delayed antibody responses exhibit no apparent functional defects, and the development of these responses coincides with the development of delayed antigen-specific CD4 ϩ T cell responses. Delayed antigenspecific antibody responses are also observed by using a protein immunogen formulated in an alum-based adjuvant. Thus, in contrast to the current model of an immediate requirement for CD4 ϩ T cell help, we demonstrate that functional antibody responses can be induced at a time separate from the time of immunization with Ad vector-or protein-based vaccines in the context of transient CD4 ϩ T cell depletion.
Monoclonal antibodies. The monoclonal anti-CD4 antibody (GK1.5; BioXcell) was administered by two intraperitoneal injections of 500 g on consecutive days. To maintain depletion of CD4 ϩ T cells, where applicable, GK1.5 was readministered in two 500-g doses every 14 to 21 days.
Serum collection and tissue processing. Blood was collected, and serum was clarified by collection of the supernatant following centrifugation at 10,000 rpm for 5 min. Two centrifugation steps were performed, and following clarification, serum was stored at Ϫ20°C for further analysis. Splenic and iliac lymph node (LN) mononuclear cells were harvested as previously described (26). Tissues were ground through a 70-m strainer (Fisher Scientific), red blood cells were lysed by ammoniumchloride-potassium (ACK) treatment for 3 min, and debris was removed by subsequent filtering through a 30-m filter (Miltenyi Biotec).
Endpoint ELISA. Endpoint enzyme-link immunosorbent assays (ELISAs) were performed as previously described (27). Briefly, ELISA plates were coated overnight with 1 g/ml of SIV mac239 gp140 Env. Plates were blocked for 4 h with a solution containing phosphate-buffered saline (PBS), 2% bovine serum albumin (BSA), and 0.05% Tween 20. Mouse serum was serially diluted, added to the ELISA plate, and incubated for 1 h. Bound serum was detected by 1 h of incubation with peroxidase-conjugated, affinity-purified rabbit anti-mouse secondary antibody diluted 1:2,000 (Jackson ImmunoResearch Laboratories). Plates were developed and read on a SpectraMax Plus ELISA plate reader using SoftMax Pro 4.7.1 software (Molecular Devices). Positive titers were defined as the greatest serum dilution with an optical density (OD) Ͼ2-fold above the naive negative-control serum OD.
Isotype and urea disruption ELISAs. Isotype and urea disruption ELISAs were performed as previously described (28,29). A semiquantitative immunoglobulin ELISA protocol described previously was followed. Briefly, ELISA plates coated with 0.5 g/ml SIV mac239 gp140 Env were blocked with 1% BSA-0.05% Tween in PBS (PBS-T). After washing, diluted samples were added to the plates for 1 h before washing and the addition of a 1:4,000 dilution of either anti-mouse IgG conjugated to horseradish peroxidase (HRP), IgG1-HRP, or IgG2a-HRP (Southern Biotech). Standards consisted of coating ELISA plate wells with anti-mouse kappa (1:3,200) and lambda (1:3,200) light chains (Serotec, United Kingdom), blocking, washing, and then adding a standard 5-fold dilution series of purified IgG or IgA (Southern Biotech, United Kingdom) starting at 1,000 ng/ml. Samples and standards were developed by using 3,3=,5,5=tetramethylbenzidine (TMB), and the reaction was stopped after 5 min with Stop solution (Insight Biotechnologies, United Kingdom). The absorbance was read on a SpectraMax Plus ELISA plate reader (Molecular Devices) with SoftMax Pro 4.7.1 software.
The avidity indices of serum samples were determined by their antibody-antigen binding resistance to 8 M urea. Serum samples were prediluted to give an OD at 450 nm (OD 450 ) readout of between 1.0 and 1.5 in an ELISA and were added to plates coated with SIV mac239 gp140 Env. Plates were then washed three times with either PBS-T or 8 M urea in PBS-T, before incubation with anti-mouse IgG-HRP. Samples were developed with TMB as described above. The avidity index was calculated as the percentage of the OD 450 of urea-treated samples/OD 450 of PBS-Ttreated samples.
SIV neutralization assay. SIV-specific neutralization assays were performed as previously described, using the TZM.bl cell neutralization assay (30). Briefly, mouse serum was serially diluted and incubated for 1 h with the tier 1A neutralization-sensitive pseudotype virus SIVmac251.TCLA. 15 and was subsequently added to TZM.bl cells. A pseudotyped virus expressing the envelope gene of murine leukemia virus (MuLV) was used as a negative control. After 48 h, cells were lysed, and relative luminescence units were quantified. The 50% infective dose (ID 50 ) was calculated as the serum concentration that reduced relative luminescence units by 50% relative to the values for a no-serum control well. Pseudotyped viruses were prepared as previously described (31).
Statistics. Mann-Whitney U tests were performed with a significance cutoff of a P value of Ͻ0.05. Correlation analysis was performed by using the Spearman test.

RESULTS
Ad vector-elicited antibody responses are delayed following depletion of CD4 ؉ T cells at immunization. We first sought to determine the time at which CD4 ϩ T cell help is required for the induction of transgene antigen-specific antibody responses by Ad vector vaccination. To accomplish this, C57BL/6 mice were immunized i.m. with 10 10 vp of Ad26-SIV Env and depleted of CD4 ϩ T cells by the administration of anti-CD4 antibody GK1.5 on either day Ϫ1, day 3, day 7, day 10, day 14, day 21, or day 28 postimmunization or left untreated, as a control (Fig. 1A). In untreated control mice, robust SIV Env-specific serum antibody titers were detected on day 14 postimmunization and were maintained for at least 90 days (Fig. 1B). Depletion of CD4 ϩ T cells prior to immunization or on day 3 postimmunization resulted in nearly undetectable Env-specific antibody responses at days 14 and 30 postimmunization (Fig. 1B). Depletion of CD4 ϩ T cells on day 7 or 10 postimmunization had progressively less of an impact on Envspecific antibody titers measured on day 30, and depletion of CD4 ϩ T cells on or after day 14 postimmunization had no impact on antibody titers measured on day 30 (Fig. 1B). As expected, the decrease in antibody titers observed following the depletion of CD4 ϩ T cells was highly correlated with reductions in the frequencies of germinal center B cells in the iliac (draining) lymph nodes (LNs) on day 14 postimmunization (r ϭ 0.76 by Spearman test; P ϭ 0.001) ( Fig. 1C to E). Signaling by CD4 ϩ T cells through CD40 is a well-described mechanism by which CD4 ϩ T cells promote antibody responses (13). We confirmed that CD40 was an important signaling pathway for antigen-specific antibody responses following Ad vector immunization, as the absence of CD40L or CD40 resulted in a complete abolishment of Env-specific antibody responses on day 30 postimmunization (Fig. 1F). Thus, following Ad vector immunization, induction of antigenspecific antibody responses requires CD4 ϩ T cell help to promote germinal center responses via CD40-derived signals, consistent with data obtained by using other experimental systems (34).
We next sought to determine whether CD4 ϩ T cells regulated the maintenance of Ad vector-induced antibody responses. Unexpectedly, when Env-specific antibody responses were measured on days 60 and 90 postimmunization, we observed no significant difference in antibody titers between mice that had been previously depleted of CD4 ϩ T cells and untreated control animals (Fig.  1B, bottom). Despite the lack of detectable serum antibody responses on day 30 postimmunization in mice depleted of CD4 ϩ T cells prior to immunization, these same animals developed Envspecific antibody titers of a magnitude equivalent to that in untreated control mice by day 60 postimmunization. We define the antibody responses that develop on day 60 postimmunization after depletion of CD4 ϩ T cells as "delayed antibody responses." Similar delayed Env-specific antibody responses were also observed in mice depleted of CD4 ϩ T cells at between days 3 and 10 postimmunization. Thus, depletion of CD4 ϩ T cells immediately prior to, or after, immunization resulted in a delay but did not prevent the induction of Env-specific antibody responses.
Delayed antibody responses following CD4 ؉ T cell depletion are functional. We sought to determine whether the delayed antibody responses observed after day 60 in mice depleted of CD4 ϩ T cells prior to immunization are functionally normal. We first assessed if the delayed antibody responses exhibited alterations in their isotype proportions. The concentration of IgG2a isotype Env-specific antibodies increased from days 30 to 90 postimmunization in anti-CD4-treated mice, and by day 90, only a statistical trend for differences in antibody concentrations in anti-CD4treated versus untreated mice was observed (P ϭ 0.06) ( Fig. 2A). Similarly, IgG1 isotype Env-specific antibody concentrations increased in anti-CD4-treated mice between days 30 and 90 and reached antibody concentrations equivalent to those in untreated control mice by day 60 (untreated median of 1,924 ng/ml and anti-CD4 median of 1,543 ng/ml; P ϭ 0.2) (Fig. 2B). On both days 60 and 90 postimmunization, the ratio of IgG2a to IgG1 isotype antibodies was equivalent between the anti-CD4-treated and untreated groups (P ϭ 0.3) (Fig. 2C). Finally, we sought to determine if the delayed antibody response that developed following the depletion of CD4 ϩ T cells would have reduced antigen-binding avidity, as measured by using a urea disruption assay (28). Env-specific antibodies in anti-CD4treated and untreated mice on day 60 or 90 postimmunization displayed no significant differences in Env-binding avidity as determined by urea disruption ELISAs (Fig. 2D). Thus, the delayed antibody responses exhibited no significant abnormalities in their isotypes or binding avidities.
To further investigate the functionality of these delayed antibody responses, we assessed whether these responses could expand after a boosting immunization. C57BL/6 mice were immunized i.m. with 10 10 vp of Ad26-SIV Env and depleted of CD4 ϩ T cells on day Ϫ1 or left untreated as a control (Fig. 3A). Four months after the primary immunization, mice were boosted i.m. with 10 10 vp of Ad5-SIV Env (Fig. 3A). One month after the boosting immunization, the anti-CD4 antibody-treated and untreated mice had equivalent median SIV Env-specific endpoint titers (Fig.  3B), which reflected equivalent mean fold expansions of 54-fold and 43-fold, respectively (P ϭ 0.8) (Fig. 3C). As a final measure of functional capacity, we assessed the ability of these delayed antibodies to neutralize an SIV Env-expressing pseudovirus. No SIVspecific neutralizing antibodies in anti-CD4-treated mice were detected at 1 month postimmunization (Fig. 3D), which is consistent with the lack of Env-specific binding antibodies at this time point (Fig. 1B). However, by 4 months postimmunization, anti-CD4-treated mice had median neutralizing antibody titers that were not significantly different from those in undepleted vaccinated mice (P ϭ 0.9) (Fig. 3D). In both groups, neutralizing antibody titers increased following the boosting immunization, and again, no differences between the two groups were observed (Fig.  3D). Collectively, these data demonstrate that the delayed antibody responses that developed following depletion of CD4 ϩ T cells at the time of primary Ad vector immunization have no detectable defects in boosting capacity, the ability to acquire func-tional neutralization capacity, isotype proportions, or antigenbinding avidity.
Continuous absence of CD4 ؉ T cells prevents development of antibody responses. We hypothesized that the development of delayed antibody responses following CD4 ϩ T cell depletion reflected the transient nature of CD4 ϩ T cell depletion following a single regimen of anti-CD4 antibody treatment. To test this, C57BL/6 mice were immunized i.m. with 10 10 vp of Ad26-SIV Env and divided into three experimental groups: (i) mice treated with anti-CD4 antibody on day Ϫ1 (anti-CD4 at prime), (ii) mice treated with anti-CD4 antibody on day Ϫ1 and again every 14 to 21 days (anti-CD4 repeated), and (iii) untreated controls (Fig.  4A). Administration of anti-CD4 antibody at prime resulted in the complete depletion of CD4 ϩ T cells for at least 30 days, but the CD4 ϩ T cell population had largely recovered by day 60 (Fig. 4B). The recovery of the CD4 ϩ T cell compartment coincided with the development of SIV Env-specific antibody responses (r ϭ 0.68 by Spearman test; P ϭ 0.0009) (data not shown and Fig. 4C). In contrast, repeated administration of anti-CD4 antibody maintained the depletion of CD4 ϩ T cells for at least 60 days (Fig. 4B). The repeated administration of anti-CD4 antibody prevented the development of SIV Env-specific antibody responses in 18 of 20 mice by day 60 postimmunization (P Ͻ 0.001) (Fig. 4C). Consistent with the observation that repeated administration of anti-CD4 antibody prevented the development of a delayed antibody response, MHC-II KO mice, which permanently lack CD4 ϩ T cells, had no SIV Env-specific antibody responses on day 60 postimmunization (Fig. 4D). Thus, the recovery of CD4 ϩ T cells following anti-CD4 antibody treatment is necessary for the development of these delayed antibody responses.
We next tested the possibility that the development of delayed antibody responses is due to incomplete depletion of CD4 ϩ T cells following treatment, thus resulting in signals from residual primed antigen-specific CD4 ϩ T cells. To test this, adult C57BL/6 mice were thymectomized and depleted of CD4 ϩ T cells prior to immunization with Ad26-SIV Env. A single administration of the anti-CD4 antibody to thymectomized mice results in a permanent depletion of CD4 ϩ T cells, due to an inability to generate new CD4 ϩ T cells (data not shown). On day 60 postimmunization, no SIV Env-specific antibodies were detected in thymectomized mice treated at priming with anti-CD4 antibody (Fig. 4E). Thus, our data show that any residual CD4 ϩ T cells are insufficient for the  generation of a delayed antibody response. Moreover, our data further highlight that de novo generation of CD4 ϩ T cells by the thymus is required for the delayed antibody response to develop. Collectively, these data demonstrate that the delayed antibody response that develops following anti-CD4 antibody treatment requires thymus-driven reconstitution of the CD4 ϩ T cell compartment.
CD4 ؉ T cell rebound following transient depletion induces normal germinal center responses. As CD4 ϩ T cells were critical for promoting the formation of germinal center responses following Ad vector immunization (Fig. 1), we hypothesized that when CD4 ϩ T cells recovered following transient depletion, de novo germinal center responses would be induced. On day 30 postimmunization, no germinal center B cells were detected in the iliac lymph nodes of Ad26-SIV Env-immunized mice treated with anti-CD4 antibody at priming (Fig. 5A). However, by day 60 postimmunization, when the frequency of CD4 ϩ T cells had returned to near-basal levels in these animals, robust germinal center responses were observed. These responses were equivalent in frequency and absolute number of cells to the germinal center responses measured in untreated control mice (Fig. 5A to C). These germinal center B cells also displayed normal downregulation of IgM and IgD (Fig. 5D). Recovery of CD4 ϩ T cells was required for the induction of these delayed germinal center responses, as sustained depletion of CD4 ϩ T cells prevented germinal centers from developing ( Fig. 5A to C). Thus, recovery of CD4 ϩ T cells results in the generation and expansion of germinal centers and the development of the delayed antibody response.
Expansion of antigen-specific CD4 ؉ T cells occurs following transient CD4 ؉ T cell depletion. We next sought to characterize the recovering CD4 ϩ T cells to determine if any of the cells were specific for the SIV Env antigen. To test this, on day 60 postim-  munization, CD4 ϩ T cells from the spleen and iliac LNs were stimulated with an overlapping SIV Env peptide pool, and cytokine production was measured by intracellular cytokine staining (Fig. 6A). CD4 ϩ T cells from both untreated control mice and mice treated with anti-CD4 antibody at priming produced modest amounts of IL-21, IFN-␥, and IL-2 in response to SIV Env peptide stimulation (Fig. 6B). Antigen-specific CD4 ϩ T cells were detected in both the spleen and iliac LNs of mice from both groups. IL-21producing CD4 ϩ T cells were identified, and as CD4 ϩ T cellderived IL-21 is a critical cytokine for germinal center responses (35), it likely partially explains how delayed germinal center responses are induced. These data demonstrate that in the context of Ad vector immunization following transient CD4 ϩ T cell depletion, antigen-specific CD4 ϩ T cells expand upon recovery.
Development of delayed antibody responses occurs following soluble protein immunization. We sought to identify other immunization regimens where delayed antibody responses may occur following transient depletion of CD4 ϩ T cells. Therefore, we tested adjuvanted soluble protein immunogens as an alternate vaccine platform. C57BL/6 mice were immunized i.m. with 50 g of trimeric SIV Env gp140 protein formulated in 100 g of Adju-Phos adjuvant, and mice were either left untreated, depleted of CD4 ϩ T cells at priming, or repeatedly depleted of CD4 ϩ T cells (Fig. 7). This immunization regimen induced robust Env-specific titers at day 30 postimmunization in untreated control mice (Fig.  7), and titers were maintained to day 60. In mice transiently depleted of CD4 ϩ T cells, no Env-specific antibody responses were detected on day 30 postimmunization, but by day 60 postimmunization, these animals had titers equivalent to those in untreated control animals (Fig. 7). Continuous depletion of CD4 ϩ T cells by repeated administration of anti-CD4 antibody prevented the development of Env-specific antibodies on day 60 (Fig. 7). These data demonstrate that a delayed antibody response can occur in the context of multiple immunization regimens and is not a unique characteristic of immunization with replication-incompetent Ad vectors.

DISCUSSION
In this study, we demonstrate that the induction of transgene antigen-specific antibody responses by vaccination can be temporally separated from the time of immunization by the transient depletion of CD4 ϩ T cells. CD4 ϩ T cell help is required for the induction of antibody responses. However, the transient absence of CD4 ϩ T cells at the time of Ad vector immunization delays the development of an antibody response only until the time at which the CD4 ϩ T cell population returns, and we identify no functional defects in these delayed antibody responses once they develop. The development of a delayed antibody response coincides with the expansion of antigen-specific CD4 ϩ T cells. This is unexpected because we previously demonstrated that the absence of CD4 ϩ T cells at the time of Ad vector immunization immediately and irreversibly impairs the development of functional CD8 ϩ T cell responses (23; Provine et al., submitted). Thus, while CD4 ϩ T cell help is critical for the induction of both antibody and CD8 ϩ T cell responses following Ad vector vaccination, the timing and mechanisms of CD4 ϩ T cell help are profoundly different. These findings have implications for understating the biology of antibody responses elicited by vaccination.
CD4 ϩ T cells have many well-established roles in regulating antibody responses against an array of antigens (3). One key aspect of this process is the sustained provision of positive signals to responding B cells by CD4 ϩ T cells (8,13,14). In those previous studies, which all utilized soluble protein antigens, the need for CD4 ϩ T cell help was absolute at the time of antigen exposure and waned gradually. Subsequent work directly visualized CD4 ϩ T cell-B cell contacts for Ͼ1 week postimmunization (12), which is in line with the timing determined by CD4 ϩ T cell depletion studies. Consistent with these findings, the induction of robust antibody responses by Ad vector immunization requires sustained CD4 ϩ T cell help for between 10 and 14 days following immunization (Fig. 1). Thus, prolonged, gradually waning CD4 ϩ T cell help to B cells appears to be a highly generalizable trait.
However, depletion of CD4 ϩ T cells at the time of antigen administration was initially suggested as a means to induce B cell tolerance, since antigen-specific antibody responses do not de-velop even after the CD4 ϩ T cell population recovers (4)(5)(6)(7)(8)(9). In contrast to those reports, we demonstrate that transient CD4 ϩ T cell depletion at the time of intramuscular immunization with an Ad vector or a protein formulated with alum did not induce tolerance to the vaccine antigen. Instead, once the CD4 ϩ T cell compartment is reconstituted, robust antigen-specific antibody responses develop without the need to experimentally readminister the antigen, and these responses appear functionally indistinguishable by our measures from the response induced if CD4 ϩ T cells are present at the time of immunization. Thus, while CD4 ϩ T cell help is required for the induction of antigen-specific antibody responses elicited by Ad vectors and soluble proteins formulated in an alum-based adjuvant, this help need not be provided at the time of immunization.
Several studies have demonstrated that, in principle, CD4 ϩ T cell help can be provided after initial antigen exposure (7,(36)(37)(38), but in those previous reports, additional experimental stimulation of CD4 ϩ T cells was required to induce antibody responses. Aubert and colleagues (7) demonstrated that adoptive transfer of naive lymphocytic choriomeningitis virus (LCMV)-specific CD4 ϩ T cells into mice transiently depleted of CD4 ϩ T cells at the time of LCMV clone 13 infection results in the generation of LCMV-specific antibody responses. The other three studies (36)(37)(38) all utilized self-antigen-specific B cell receptor (BCR) transgenic mouse models. In such a system, cognate B cell antigens are continuously present, and therefore, any provision of CD4 ϩ T cell help can be seen as being after the time of initial antigen exposure. In all three reports, experimental provision of CD4 ϩ T cell help induced the development of autoantibody responses. However, in all of those previous studies, antigen-specific CD4 ϩ T cells were adoptively transferred into recipient animals and were experimentally activated prior to or after transfer. From those studies, it remains unclear if CD4 ϩ T cell help can occur at a time separated from the time of initial antigen exposure without experimental manipulation and reactivation of the CD4 ϩ T cell compartment. In this study, we identify antigen-specific cells within the recovering CD4 ϩ T cell population without the need for experimental manipulation, and the presence of these cells likely predicates the development of the delayed antibody response. Thus, we demonstrate for the first time, to our knowledge, that endogenous CD4 ϩ T cell responses can provide help to B cells at a time separated from the time of initial antigen exposure.
Optimal priming of a T cell response requires T cell receptor (TCR) engagement, costimulation, and proinflammatory cytokines (1). The induction of a delayed antibody response requires de novo thymic production of naive CD4 ϩ T cells (Fig. 4), and it takes Ͼ30 days for these cells to recover following depletion. Thus, for priming of CD4 ϩ T cells to occur, Ad vector immunization must induce sufficient expression of the necessary priming pathways for at least 30 days postimmunization. It was previously reported that Ad vectors induce the necessary signals to induce naive CD8 ϩ T cell proliferation for at least 30 days postimmunization (39,40). We extend these findings by demonstrating not only that Ad vectors provide the necessary signals to induce T cell proliferation for more than a month postimmunization but also that Ad vectors induce all of the signals necessary to induce a functional CD4 ϩ T helper response for at least a month postimmunization. Thus, despite lacking the capacity to replicate, Ad vectors and appropriate soluble protein-based immunization regimens can adjuvanted soluble protein formulated in Adju-Phos but not following immunization with a replication-incompetent poxvirus vector. SIV Env-specific serum antibody titers in C57BL/6 mice immunized intramuscularly with 50 g of SIV Env gp140 plus Adju-Phos and depleted of CD4 ϩ T cells at the time of immunization (anti-CD4 at prime) or continuously depleted of CD4 ϩ T cells (anti-CD4 repeated) or mice left untreated were determined. Each dot represents an individual mouse, and the line is the median. The horizontal dotted line denotes the limit of detection for the assay (n ϭ 4 to 5/group from one experiment).
induce functional T cell responses at a time markedly separated from the time of immunization.
How B cell priming occurs with regard to a delayed antibody response remains unclear. Even in response to T cell-dependent antigens, the initial activation of naive B cells by cognate binding of antigen to the BCR occurs prior to the interaction with T cells (41)(42)(43). However, a failure to immediately receive T cell help results in the anergy or apoptosis of these activated B cells (44)(45)(46)(47). Thus, it is possible that the delayed antibody response that occurs following transient depletion of CD4 ϩ T cells is the result of these antigen-experienced B cells receiving the necessary activation signals following reconstitution of the CD4 ϩ T cell compartment. Alternatively, the delayed antibody response may instead involve B cells that recognize antigen and become activated at a time more contemporaneous with the recovery of CD4 ϩ T cells. In this study, the vaccine doses used (10 10 vp of the Ad vector or 50 g of protein) are on the high end of doses used in mice but are not outside accepted dose ranges for these immunogens in mice (13,(48)(49)(50)(51), and these high antigen levels might allow sufficient antigen persistence to prime B cells at a time coinciding with the recovery of CD4 ϩ T cells. Future experiments will be required to distinguish between these two possibilities and will enhance our understanding of this phenomenon.
In conclusion, we have shown the role of CD4 ϩ T cell responses in helping to elicit serum antigen-specific antibody responses following immunization with nonreplicating Ad vectors and adjuvanted soluble protein immunogens. Our data demonstrate that antibody responses can be induced at a time separated from the time of immunization by manipulation of CD4 ϩ T cells, and this may be a useful approach to specifically control the timing of development of antibody responses. These findings demonstrate that the provision of critical CD4 ϩ T cell-derived help signals to B cells can occur substantially after the time of initial antigen exposure. Thus, the temporal regulation of Ad vector-elicited antibody responses by CD4 ϩ T cells is completely distinct from the temporal regulation of Ad vector-elicited CD8 ϩ T cell responses (23; Provine et al., submitted). These data identify a previously unappreciated dynamic nature of the timing of CD4 ϩ T cell help for the induction of antibody responses.