Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Jun 1.
Published in final edited form as: Immunity. 2008 Jun;28(6):847–858. doi: 10.1016/j.immuni.2008.04.018

Selective CD4 T cell help for antibody responses to a large viral pathogen: deterministic linkage of specificities

Alessandro Sette 1, Magdalini Moutaftsi 1, Juan Moyron-Quiroz 1, Megan M McCausland 1, D Huw Davies 2, Robert J Johnston 1, Bjoern Peters 1, Rafii Mohammed Benhnia 1, Julia Hoffmann 3, Hua-Poo Su 4, Kavita Singh 4, David N Garboczi 4, Steven Head 3, Howard Grey 1, Philip L Felgner 2, Shane Crotty 1,✉
PMCID: PMC2504733  NIHMSID: NIHMS55832  PMID: 18549802

Summary

Antibody responses are critical components of protective immune responses to many pathogens, but it remains unclear what parameters determine which pathogen proteins are targeted by the host antibody response. Peptide vaccination of mice with individual MHC II restricted vaccinia virus epitopes revealed that after vaccinia (smallpox vaccine) infection CD4 T cell help to B cells was surprisingly non-transferable to B cells specific for different vaccinia virus proteins, including other virion surface or core proteins. Furthermore, large scale analysis of the antibody and CD4 T cell response specificities to vaccinia virus revealed that many of the CD4 T cell responses identified in an unbiased screen targeted viral virion proteins that were also targets of antibody responses, consistent with a deterministic linkage between the specificities (P < 0.0009). We then tested the deterministic linkage model by using this knowledge to efficiently predict new vaccinia virus MHC II epitopes (830% increase in identification efficacy). In contrast to the standard model, these data indicate that individual proteins are the primary unit of immunological recognition for a large virus, and therefore MHC restriction at the protein level is a key selective event for the antiviral antibody response and protective immunity, which is likely of great relevance for vaccine development to large pathogens.

Introduction

Vaccines are one of the most cost-effective medical treatments in modern civilization (Rappuoli et al., 2002). Vaccinia virus (VACV) is the viral species used as the human smallpox vaccine. The smallpox vaccine has been extraordinarily effective, having brought about the worldwide eradication of smallpox disease (Fenner, 1988). The smallpox vaccine is generally considered the gold standard of vaccines, and elucidating the immunobiology underlying the protection provided by the smallpox vaccine will continue to reveal vaccinology principles that can be applied to future vaccine development against other infectious scourges. However, identifying and analyzing the fine specificities of the adaptive immune response to a large pathogen—such as a poxvirus—is confounded by a number of factors, not least of which is the stark magnitude of the potential protein and peptide targets of the antibody and T cell responses. As a result of these challenges, we possess only a piecemeal understanding of the fine specificities of T cell and antibody responses to any large pathogen and therefore have a thin understanding of the roles of each fine specificity in protective immunity, limiting our ability to rationally design new vaccines against large and complex pathogens.

While neutralizing antibodies are of primary importance in the protection from smallpox provided by the smallpox vaccine in animal models (Belyakov et al., 2003; Edghill-Smith et al., 2005; Galmiche et al., 1999; Lustig et al., 2005) and humans (Amanna et al., 2006; Demkowicz et al., 1992), CD4 T cells and CD8 T cells are also of great value (Amanna et al., 2006; Fang and Sigal, 2005; Tscharke et al., 2005; Xu et al., 2004). Here, we have focused on understanding the relationship between antibody and CD4 T cell responses to vaccinia virus in mice, as part of a strategy to elucidate the value of individual fine specificities, potential interrelationships between those specificities, and the underlying immunobiological and virological parameters that determine the emergence of protective immune responses to a small subset of all possible specificities.

Results

Exquisitely selective antigen-specific T cell help

Infection of mice with VACVWR results is an acute infection characterized by several days of high viral replication and viral loads of > 108 infectious virions, followed by a strong adaptive immune response and viral clearance in 1–2 weeks (Amanna et al., 2006; Harrington et al., 2002; Xu et al., 2004). IgG responses to VACV are fully dependent on CD4 T cell help (Fig. 1A and ref. (Xu et al., 2004)). We recently identified 14 VACV MHCII epitopes after VACV infection of B6 mice (Moutaftsi et al., 2007). CD4 T cells of each specificity expressed CD40L after stimulation with cognate peptide, indicating their competence to provide B cell help (Fig. 1B). In an effort to boost the antiviral antibody responses to VACV infection, we increased the available CD4 T cell help by immunizing mice with the VACV I121–35 MHC II epitope, then infecting the mice with VACV, and finally monitoring the subsequent antiviral antibody responses. Vaccinating with I121–35 MHC II epitope resulted in a strong 10-fold increase in the total anti-VACV antibody response, as measured by a standard VACV ELISA (Fig. 2A). Unexpectedly, virus neutralizing antibody titers were unimproved in VACV infected mice preimmunized with I121–35 when compared to unprimed mice (Fig. 2B). While I1 is a viral virion core protein and therefore not itself a neutralizing antibody target, Il-specific CD4 T cells were expected to provide intermolecular help to all B cells specific for VACV viral particle proteins and thereby boost neutralizing antibody titers (Janeway et al., 2005). Surprisingly, when serum samples were probed for the detailed antigen specificities of the antibody response using vaccinia protein microarrays, we found the increased antibody response was exclusively against I1 and not other VACV proteins (Fig. 2C–H). IgG specific for virion core protein I1 was increased 1930% (P < 0.0004)(Fig. 2E). A10 is a second major core protein present along with I1 in VACV virions, and A10 is a target of the antiviral IgG response in infected B6 mice (Fig. 5A–D). Nevertheless the strength of the anti-A10 IgG response was unaltered in I1 MHC II epitope vaccinated mice (Fig. 2F). The IgG responses to two vaccinia surface membrane protein virion components, D8 and H3, also exhibited no improvement (Fig. 2G–H). Selective increase in anti-I1 IgG was still observed at a memory timepoint (Fig. 2I).

Figure 1. Vaccinia Virus CD4+ helper T cells and helper T cell dependent antibodies.

Figure 1

Open in a new tab

(A) Quantitative ELISA of anti-VACV IgG (µg/ml), day 30 post-infection. Anti-VACV IgG is absent in MHCII−/− mice. “WT”, wildtype B6 mice. (B) Splenocytes from day 10 VACVWR infected mice were incubated with CD11c+ dendritic cells (DCs) pulsed with VACV peptides. Cells were incubated for 6 hrs and then stained for flow cytometry. Gated CD4+ CD62Llo lymphocytes are shown, stained for intracellular IFNγ and CD40L. DCs infected with VACVWR (MOI=5) for 2h prior addition of splenocytes were used to identify the total anti-VACV CD4+ T cell response (bottom right, “VV+ DCs”). Background levels were determined using uninfected DCs (“neg”). Low frequency A28-specific response was only detectable by ELISPOT (not shown).

Figure 2. Selective protein-specific CD4 T cell help to B cells after VACV infection.

Figure 2

Open in a new tab

Mice peptide vaccinated with VACV I121–35 MHCII epitope were then infected with vaccinia virus. (A) Vaccinia-specific IgG responses in VACV infected mice primed with adjuvant alone (“mock prime”, open circles), I1-primed mice subsequently infected with VACV (squares), I1-primed only mice (X symbol), and uninfected control mice (+ symbol) measured by ELISA. n = 4/group. Error bars ± SEM. (B) Virus neutralizing antibody titers (PRNT50) were unimproved in I1-primed mice (P ≫ 0.05). I121–35 MHCII peptide primed mice (“+”, n = 4) and mock primed mice (“—”, adjuvant only prime. n = 4), tested after VACVWR infection. (C–D) Sets of VACV proteins were synthesized and printed in microarray format to generate VACV proteome arrays that could be probed with serum samples (see Methods). VACV proteome microarrays were probed with serum day 7 post-VACV infection from representative mice (C) primed with I121–35 or (D) mock primed. Star indicates anti-I1 signal. Error bar indicates range of replicates. (E–H) Antibody responses to individual vaccinia virus protein determinants after VACVWR infection in I121–35 MHCII peptide primed mice (“+”, n = 4) and mock primed mice (“—”, adjuvant only prime. n = 4). (E) Quantitation of anti-I1 IgG revealed 19.3-fold increase in primed mice (P < 0.0004). (F) Anti-A10 IgG (P ≫ 0.05), (G) Anti-D8 IgG (P ≫ 0.05), and (H) Anti-H3 IgG (P ≫ 0.05) levels were unchanged. (I) IgG responses in I1 primed and not primed mice at day 30 after VACV infection. Anti-I1 IgG was selectively increased (P < 0.02), while anti-A10 and anti-D8 responses were unchanged (P ≫ 0.05). *, P < 0.05. ** P < 0.01. ***, P < 0.001. Data is representative of multiple independent experiments.

Figure 5. Interrelationship between anti-VACV CD4 T cell and antibody responses in virus infected mice.

Figure 5

Open in a new tab

(A–C) Quantitative analysis of IgG antibody responses to individual VACV proteins in two representative VACV-infected B6 mice (A–B) and one uninfected mouse (C), as measured by proteomic microarray (RU, relative fluorescence units). Stringent limit of detection is indicated by dashed line. Representative of > 20 animals. (D) Representative immunofluorescence microarray scan of VACV protein microarray probed with sera from an uninfected mouse versus a VACV infected mouse. Each VACV protein is present as duplicate spots. (E) Tabulation of interrelationship between the antiviral CD4 T cell targets (columns) and antibody targets (rows). Matched CD4 T cell and antibody specificities are indicated in red. Specificities are ranked roughly in descending order based on immunodominance / strength of response (T cell targets, left to right. Antibody targets, top to bottom). 17 IgG targets were identified in the majority of infected mice (shown), and variable IgG responses were also seen to minor antigens F9, I3, A56, A17, A13, and WR149 in some infected mice (not shown, but included in the statistical analysis to be conservative). B cell specificities subsequently selected for prediction of CD4 T cell responses are highlighted in yellow.

Antibody response selectivity driven by epitope-specific CD4 T cells

To directly establish the role of epitope-specific CD4 T cells in the selective targeting of individual viral virion proteins for antibody responses, we performed CD4 T cell adoptive transfer experiments. CD4 T cells were purified from donor mice immunized with I121–35 MHC II epitope, and cells were transferred into unimmunized recipient mice (Fig. 3A). Recipient mice were infected with VACV and examined for anti-VACV antibody responses. Mice receiving primed CD4 T cells responded with a dramatic increase in anti-I1 IgG (1210% enhancement, P < 0.0001)(Fig. 3B–D), while the antibody responses to other viral virion protein specificities remained unchanged (Fig. 3B–C, E–F), starkly ignorant of the CD4 T cells transferred from the VACV I121–35 vaccinated mice.

Figure 3. CD4 T cell dependent MHCII restricted help for VACV antibody response.

Figure 3

Open in a new tab

(A) CD4 T cell adoptive transfers, experiment schematic. Mice were immunized with a VACV MHCII-binding peptide, I121–35. CD4 T cells were purified from donor mice and transferred to unimmunized mice. Recipient mice were then infected with VACV. (B–C) Quantitative analysis of IgG antibody responses to individual VACV proteins in representative mice that (B) received I121–35 primed CD4 T cells, or (C) did not received primed CD4 T cells. Star indicates anti-I1 IgG. (D) Anti-I1 IgG response was increased 1210% in I121–35 MHCII peptide primed CD4 T cell recipient mice (“+”. P < 0.0001, n = 3. Control mice “—”, n = 4), while the (E) anti-A10 and (F) anti-D8 IgG responses were unimproved. Data are representative of two experiments.

This observation of paired targeting by the CD4 T cell and antibody response indicates that CD4 T cell help is preferentially provided to B cells with the identical protein specificity, even though the viral particle exists as a solid physical structure assembled from greater than 75 distinct viral protein components (Chung et al., 2006; Condit et al., 2006; Moss, 2001; Resch et al., 2006), which is expected to function as a unified immunological target for B cells and B cell – T cell interaction (Janeway et al., 2005; Milich et al., 1987; Russell and Liew, 1979; Scherle and Gerhard, 1986). Naked unwrapped viral nucleoprotein core particles are released from dying infected cells and would be expected to function in the same manner.

To determine whether these I1 core protein results were generalizable, we tested the properties of several additional VACV CD4 T cell responses shown in Figure 1. We chose MHC II epitopes from H3 and D8, as these major virion surface transmembrane proteins are key IgG targets (Amanna et al., 2006; Davies et al., 2005b) and are ostensibly recognized by B cells as prominent components of the surface of whole virions. In addition, we tested the vaccinia L4176–190 MHC II epitope, representing a second major viral particle core component to compare with the responses to I1 core protein. Antiviral IgG responses were enhanced after VACV infection of mice primed with any one of the three new VACV MHC II epitopes (Fig. 4A). VACV proteome microarray serological analysis again revealed a remarkable selectivity of the antibody response in peptide vaccinated animals. In H3272–286 CD4 epitope primed animals, anti-H3 IgG was increased dramatically (48-fold increase, P < 0.0001)(Fig. 4B). H3 is a known target of neutralizing antibodies (Davies et al., 2005b; Lin et al., 2000), and virus neutralizing antibody titers were selectively increased in mice with H3-specific CD4 T cells (Fig. 4C, P < 0.02). Anti-A10 IgG levels were unaltered (Fig. 4B), confirming and expanding the results first obtained for I1 (Fig. 2). Anti-H3 IgM levels were also selectively increased (Fig. 4G–H). Particularly noteworthy, since H3 is a surface virion protein, was the observation that IgG responses against a second transmembrane surface virion protein, D8, were unaltered in H3 primed animals (Fig. 4D). In contrast, anti-D8 IgG was increased substantially in D8238–252 CD4 epitope vaccinated mice (P < 0.03)(Fig. 4E). The increased antibody response was again selective, as anti-A10 IgG levels were unaltered (Fig. 4E). The fourth target tested, L4, confirmed and extended the results seen above for I1, H3, and D8. Anti-L4 IgG levels were boosted in mice vaccinated with L4176–190 MHC II peptide, but the remainder of the anti-vaccinia antibody response was unaltered (Fig. 4F).

Figure 4. Highly selective CD4 T cell help to B cells specific for VACV virion components.

Figure 4

Open in a new tab

Antiviral antibody responses in mice immunized with vaccinia virus H3272–286, D8238–252, or L4176–190 peptide MHC II epitopes and then infected with VACV. (A) Vaccinia-specific IgG responses, measured by ELISA, in VACV infected mice primed with adjuvant alone (“mock primed,” open circles), or H3 primed (closed circles), or D8 primed (closed squares), or L4 primed (closed diamonds) mice subsequently infected with VACV. Peptide primed only mice (X symbol) and untreated uninfected mice (+ symbol) served as controls. n = 4/group. Error bars ± SEM. (B) Left panel: Anti-H3 IgG response after VACV infection was increased 48-fold in H3272–286 MHCII peptide primed mice (“+”. P < 0.0001, n = 11. Adjuvant only “mock primed” mice, “—”. n = 12. Composite data from three independent experiments). Right panel: The anti-A10 response was unchanged (P ≫ 0.05). (C) Virus neutralizing antibody titers (PRNT50) were increased in H3-primed mice. (D) Anti-D8 IgG response was unchanged in H3272–286 primed mice (P ≫ 0.05). (E) Anti-D8 IgG response was increased 2.2-fold in D8238–252 MHCII peptide primed mice (primed, “+”. P < 0.04, n = 4. Adjuvant only “mock primed” mice, “—”. n = 4), while (right panel) the anti-A10 response was unchanged (P ≫ 0.05). (F) Anti-L4 IgG response was increased 3.2-fold in L4176–190 MHCII peptide primed mice (primed, “+”, n = 4. Adjuvant only “mock primed” mice, “—”. n = 4), while (right panel) the anti-A10 response was unchanged. (G) Vaccinia-specific IgM responses, measured by ELISA in VACV infected mice (open circles), or H3 primed (closed circles) or I1 primed (closed squares) or D8 primed (closed triangles) or L4 primed (closed diamonds) mice subsequently infected with VACV. CFA primed only mice (+ symbol) served as controls. n = 4/group. (H) Anti-H3 IgM response (left panel) was increased in H3272–286 primed mice (P < 0.0019), while (right panel) the anti-A10 IgM response was unchanged (P ≫ 0.05). *, P < 0.05. ** P < 0.01. ***, P < 0.001. Data is representative of multiple independent experiments.

Matched immunodominant antibody and CD4 T cell responses during poxvirus infection

The results above suggest that each antibody response need be accompanied by a matched CD4 T cell response to the same protein, as if the virion were perceived as a collection of individual protein specificities, in contrast to the expectation that the virion behaves as a unified target with promiscuous CD4 T cell help. The standard assumption has been that no direct linkage is required between B and CD4 T cell responses to pathogens, since the physical body of the pathogen (the virion or bacterium) is expected to act as a unit, a unified target for B cell binding and subsequent processing and presentation to CD4 T cells (Janeway et al., 2005). This assumption can be succinctly summarized as an “any MHCII peptide is sufficient” model. That is, a B cell binds cognate Ag on the virion and then internalizes and processes the whole virion for Ag-presentation, resulting in the B cells presenting MHCII epitopes from many different virion proteins such that interaction with a CD4 T cell specific for an epitope from any of the virion proteins will result in a cognate interaction and appropriate CD4 T cell help to the B cell (termed intermolecular help), as shown for influenza and HBV (Janeway et al., 2005; Lake and Mitchison, 1976; Russell and Liew, 1979; Scherle and Gerhard, 1986). However, our data demonstrates an unexpectedly tight linkage between the protein specificity(ies) of the CD4 T cells and the protein specificity(ies) of the anti-VACV antibody response (intramolecular help), indicating that individual protein identities are the primary unit of immunological recognition for a large pathogen. Indeed, recognition of the physical vaccinia viral particle as a unit per se appears to be irrelevant, since the recognition of I1 or H3 by CD4 T cells is predominantly non-transferable to other virion components in the context of B cell help: I1-specific CD4 T cells did not provide help to B cells of other virion protein specificities (Fig. 2–Fig. 4). This is of great relevance, because the vast majority of potential viral antibody targets are generally considered irrelevant for protective immunity, and only virion surface proteins are relevant targets for antibody neutralization activity. This was demonstrated by the failure to improve the neutralizing antibody response after vaccination generating I1-specific CD4 T cells (Fig. 2B), in contrast to the improved neutralizing antibody titers after vaccination generating H3-specific CD4 T cells (Fig. 4C).

This led us to examine whether such CD4 T cell – B cell linkage is observed in the context of the natural VACV immune response. Given the lack of predictive algorithms for identifying IAb binding peptides (Rammensee et al., 1999; Sette et al., 1993), we identified the 14 MHC II VACV epitopes (Fig. 1B) by screening an unbiased, effectively random, set of 2,146 15-mer peptides sampling 31.5% of the VACV genome ORF sequences (see Methods). These epitopes represent 26–33% of the total anti-VACV CD4 T cell response (Fig. 1B, sum of all epitopes compared to VACV-infected DC APCs), consistent with the random distribution of the 2,146 peptides screened (see Methods). Concurrent with our determination of the CD4 T cell response specificities, we identified the protein targets of the antiviral antibody response in VACV infected B6 mice by using a series of vaccinia proteome arrays covering a total of 191 distinct genes (Fig. 5A–D, and refs.(Davies et al., 2005a; Davies et al., 2005b)). The viral protein targets of the antibody response were determined by probing vaccinia proteome arrays with serum from > 20 VACVWR infected animals at day 30 post-infection. Definitive IgG responses were identified against 23 proteins, 17 of which were observed in the majority of infected animals (Fig. 5A–E).

Extensive overlap was observed between the antigens recognized by the CD4 T cells and the antibodies. Six of the VACV protein antigens exhibited matched antibody and CD4 T cell responses (Fig. 5E), representing 46% of the known CD4 T cell targets (6 of 13) and 26% of the antibody targets (6 of 23). Given that VACV contains greater than 200 genes, this concurrence was striking. Statistical analysis confirmed that this linkage was highly unlikely to occur by chance (P < 0.0009, Fisher Exact Test). In contrast, no statistically significant linkage was observed between the CD4 T cell and CD8 T cell targets, indicating that these processes occur by distinct mechanisms (Moutaftsi et al., 2007).

The six paired B cell - CD4 T cell targets represented all three major classes of viral proteins: virion surface proteins (H3, D8, A28), virion core proteins (I1, L4), and nonstructural proteins expressed within infected cells (D13). Also of note, the CD4 T cell target antigens not matched by an antibody response were almost all (6 of 7) viral transcription factors and replication factors (A18 transcription factor, A20 DNA polymerase cofactor, A24 RNA polymerase subunit, E1 polyA polymerase, J4 RNA polymerase subunit)(Fig. 5E). We speculate viral replication factors may be poor B cell antigens due to low concentration or intracellular expression. This is consistent with a model where an antigen-specific B cell response is dependent on a CD4 T cell response of matched specificity, but not vice versa.

Prediction of antiviral CD4 T cell responses based on antibody specificities

Prediction of CD4 T cell response epitope specificities is difficult due to Class II binding motif degeneracy and other considerations (Rammensee et al., 1999; Sette et al., 1993). Furthermore, those difficulties are exacerbated by the sheer size of large pathogens such as poxviruses. Since our data show that anti-VACV virion antibody responses are restricted by protein matched CD4 T cell help (Fig. 2– Fig. 4), our model suggests we should be able to predict additional VACV CD4 T cell targets based on observing viral antibody target specificities. To test this hypothesis, VACV IgG targets B5, A4, A27, B2, and A33 were selected on the basis that CD4 T cell responses had not been detected to these targets in the 1st round of random screening (Fig. 1B, Methods, and (Moutaftsi et al., 2007)) but we now predicted that, for each antibody target, the antibody response was predicated on a matched CD4 T cell response (Fig. 5E). As negative controls two virion proteins that are not IgG targets were also selected, A9 and D3. Overlapping peptides representing the entirety of each protein were screened using IFNγ ELISPOT to detect the presence of epitope-specific CD4 T cell responses in VACV infected B6 mice. Responses were then confirmed by intracellular cytokine staining of CD4 T cells from VACV infected mice (Fig. 6A). Impressively, CD4 T cell responses were identified against 4 of the 5 B cell targets tested: B5, B2, A4, and A33 (Fig. 6A), raising our total number of identified VACV CD4 T cell epitopes to 18. No CD4 T cell responses were detected to control ORFs A9 and D3. Strikingly, the response to B246–60 was the strongest CD4 T cell response of all 18 epitopes identified, and the B546–60 epitope was the third strongest response overall (Fig. 6A, compared with Fig. 1B. All peptides were tested concurrently to allow for direct quantitative comparisons, ranked in Fig. 6B). The four new epitopes accounted for 22% of the total VACV specific CD4 T cells in infected mice, compared with 26% accounted for by the initially identified 14 epitopes. Taking into consideration that 211 peptides were tested from the five anti-VACV target proteins in the new selective screen vs. 2146 random peptides in the original screen, identification of CD4 T cell targets via prediction of linkage based on serology represented an 830% increase in predictive power over random screening of peptides. This efficient strategy for CD4 T cell epitope identification has obvious practical benefits for studying other complex pathogens of interest.

Figure 6. Utilization of antibody specificities to predict new vaccinia virus protein targets of CD4 T cell responses, and test of protective immunity induction in vivo.

Figure 6

Figure 6

Figure 6

Open in a new tab

(A) IFNγ and CD40L intracellular staining for VACV B246–60, B546–60, A33116–130, and A466–80 -specific CD4 T cells at day 10 after VACV infection. Gated on CD4+ CD62Llo lymphocytes. “Neg” = no peptide control. As negative controls, two virion proteins that are not IgG targets were also selected, A9 and D3. No CD4 T cell responses were detected to control ORFs A9 and D3 (data not shown). (B) Tabulation of the full set of discovered interrelationships between the antiviral CD4 T cell targets (columns) and antibody targets (rows). 11 of 18 CD4 T cell responses (highlighted in pink) are matched by IgG responses to the same smallpox vaccine virus protein. Specificities are ranked and marked as described above. (C) Antibody responses to B5 (left) and VACV MV (right) proteins after VACVWR infection of B546–60 MHCII peptide primed mice (“+”) and mock primed mice (“—”, adjuvant only prime). (D) Weight loss in B6 mice infected intranasally with VACVWR. Groups were primed to generate B5-specific CD4 T cells (closed circles, n = 8), H3- specific CD4 T cells (squares, n = 8), or primed with CFA adjuvant alone (n = 8) prior to challenge. (E) Weight loss in B cell deficient mice (µMT) infected intranasally with VACVWR. Experiment done concurrently with that of panel D. N = 4 per group. (F) Survival curves in B5 CD4 T cell primed (left) and H3 CD4 T cell primed (right) C57BL/6 (“WT”) and B cell deficient mice after VACVWR challenge.

Role in protective immunity in vivo

Having predicted and identified a CD4 T cell response to B5, we then showed that B5-specific CD4 T cells provide selective help to B5 specific B cells. This is of interest not only to confirm the prediction, but also because B5 is a important membrane protein and neutralizing antibody target on the second form of VACV viral particle, extracellular enveloped virion (EV), which has a second outer membrane layer, with distinct membrane proteins, covering an inner membrane and virion core that is identical to the much more abundant MV (mature virion) viral particle form (which includes H3 and D8 as surface membrane proteins). Mice primed to generate a B5-specific CD4 T cell response and then infected with VACV develop an enhanced anti-B5 antibody response (Fig. 6C, 4-fold increase, P < 0.04). Normal VACV antigen preparations are dominated by MV virion antigens and contain only trace levels of B5 (determined using an anti-B5 monoclonal antibody, unpublished data), and therefore VACV MV antigen ELISA was an efficient test to demonstrate that the B5 CD4 T cells only help B5 specific B cells, as the overall anti-VACV MV IgG response was not enhanced in mice with pre-existing B5-specific CD4 T cells (Fig. 6C).

Knowing that B5 and H3 are both targets of protective neutralizing antibody responses, we could test the biological relevance of the T-B linkage in vivo in a lethal poxvirus challenge system. Mice were immunized with H3272–286 or B546–60 MHC II peptide following the standard immunizations done previously in this study, such that they possessed B5-specific CD4s or H3-specific CD4s prior to intranasal challenge with 1 × 104 PFU VACVWR (~1 LD50). Primed mice possessing B5 CD4s or H3 CD4s were significantly protected from morbidity (weight loss) after VACVWR infection compared to control adjuvant-primed mice (Fig. 6D, P < 0.007, P < 0.02), and primed mice possessing I1 CD4s did not exhibit protection (P ≫ 0.05, data not shown). This protection was dependent on the enhanced neutralizing antibody response, since the protection mediated by the B5 and H3 specific CD4s was completely lost in the absence of B cells, measured either by morbidity (Fig. 6E) or mortality (Fig. 6F. B5, P < 0.02; H3; P < 0.0001) after VACVWR intranasal challenge.

Discussion

Three distinct lines of evidence demonstrate the deterministic linkage of B cell and CD4 T cell specificities to this large virus: 1) MHC II peptide vaccination results in highly selective CD4 T cell help to matched B cells specific for the same viral virion protein, 2) a strong overall correlation was found between CD4 T cell targets and antibody targets in vivo (P < 0.0009), and 3) new MHC II epitopes recognized by antiviral CD4 T cells were efficiently predicted based on this model (830% increase in predictive power). In total, 11 of the 18 CD4 T cell responses we identified were matched by a paired antibody response to the same viral protein (Fig. 6B), including all of the top 5 viral protein IgG targets (Fig. 5, 6B). Our results reveal an unexpectedly tight linkage between the CD4 T cell and antibody response specificities for VACV—the first large viral pathogen examined in this manner. Furthermore, we show that this is important for the generation of neutralizing antibodies (Fig. 4C vs. Fig 2B) and protective immunity (Fig. 6C–F). Intermolecular help has been a well accepted viral immunology model for 20 years, based on data from influenza virus studies (Russell and Liew, 1979; Scherle and Gerhard, 1986), and corroborating evidence from hepatitis B virus (HBV)(Milich et al., 1987). However, putative whole VACV viral particle uptake and antigen presentation is not detectable, as measured by in vivo CD4 T cell help (Fig. 2– Fig. 4).

We consider that there are two plausible reasons for why whole VACV viral particle uptake and presentation is not observed, and we consider that there are three feasible mechanisms for how the virus-specific B cells are acquiring antigen and T cell help. The first plausible reason for the failure of whole VACV viral particle uptake and antigen presentation is a size restriction. There is a substantial size difference between the experimental pathogens being considered. VACV virions are large (360nm diameter)(Cyrklaff et al., 2005; Roos et al., 1996), while flu virions are small (80nm (Knipe and Howley, 2001)), and HBV particles are smaller still (20–45nm (Knipe and Howley, 2001)). Antigen-specific uptake by B cells is via BCR mediated endocytosis, and endocytic vesicles are only 50–150nm in diameter (Goldstein et al., 1979; Lodish et al., 2003; West et al., 1994). This may result in size exclusion at the level of cellular uptake, and would be a restriction generally applicable to large pathogens, including large viruses, bacteria, and parasites. A second plausible reason for failure to observe whole VACV viral particle uptake and presentation could be that viral particles are not a significant source of antigen, either due to limiting numbers of virions produced or limited accessibility to B cells. It may be that VACV infection, while quite robust, does not reach the very high levels of viral particle production obtained during some other infections (for example, HBV infection results in systemic serum virion levels in excess of 109 per ml, whereas VACV is normally undetectable (<102 per ml) in serum (Briody, 1959; Hollinger and Liang, 2001)). This scenario would then be applicable to a wide range of infections that result in mid grade physiological concentrations of pathogen. Alternatively, the absolute number of viral particles may not be the relevant parameter, but instead the accessibility of that antigen to B cells. This issue has been highlighted by several recent studies examining the impact of lymph node architecture on antigen acquisition by B cells. Large multimolecular particles such as viruses (Hickman et al., 2008; Junt et al., 2007), bacteria (Carrasco and Batista, 2007), or antigens or comparable size (Carrasco and Batista, 2007; Phan et al., 2007) are excluded from diffusion into lymph nodes. Viruses and other particulate antigens traffic through lymph and are captured by macrophages that bridge the subcapsular sinus and the underlying B cell follicle. The particles are then transferred to B cells which migrate deep into the follicle to the T-B boundary (Carrasco and Batista, 2007; Junt et al., 2007; Phan et al., 2007), where the B cells presumably initiate interactions with CD4 T cells. However, while VACV is excluded from freely diffusing through follicles (Hickman et al., 2008; Junt et al., 2007), and VACV replication is highest in ovaries (Briody, 1959; Xu et al., 2004), it is also known that VACV directly infects lymph node and spleen cells (Briody, 1959; Hickman et al., 2008) and has considerable access to cells in the lymph node medulla (Junt et al., 2007). Furthermore, DCs can retain intact microbes (Balazs et al., 2002; Kwon et al., 2002; Macpherson and Uhr, 2004) and transfer antigens to B cells (Qi et al., 2006; Wykes et al., 1998), suggesting multiple different routes by which whole virions may encounter B cells.

Given those possible reasons why whole VACV virion uptake and antigen presentation is not occurring at detectable levels, there are three feasible mechanisms for how the B cell antigen uptake and presentation is occurring during a VACV infection. The first and simplest model is that VACV virions are not involved (for reasons described above) and free individual viral proteins are the physiological B cell antigens. Data is available that supports this model. Soluble proteins can be rapidly trafficked through lymph and be acquired by lymph node resident B cells (Pape et al., 2007). This has also been observed in the context of a viral infection (VSV, (Junt et al., 2007)). This appeal of this model is only weakened by the consideration that monomeric soluble antigens would fail to induce BCR crosslinking and signaling. This lack of BCR signaling would fail to enrich for antibody responses to virion surface proteins presented in their appropriate conformation, which are generally the only physiologically relevant target antigens for neutralizing antibody responses and protective immunity, and which are efficiently induced by the highly multimeric nature of virions via extensive BCR crosslinking upon interaction with a virus-specific B cell (Bachmann and Zinkernagel, 1997). One could argue for an intermediate level of macromolecular structure (sub-virion fragments or particles) such that sufficient copies of antigen are available for BCR crosslinking (which may be occurring in the context of VSV infection, as observed in Sup. Fig. 6 of (Junt et al., 2007)), but this is an unlikely solution for large viruses with complex virion protein mixtures, as the macromolecular structures would have to be of single protein species to fit the observation that CD4 T cell help to VACV B cells is highly protein specific. However, a second potential mechanism would allow immunologic utilization of the information contained in the multimeric nature of the virion: if B cells bind to whole virions in vivo but are unable to internalize the large particle, the B cells may pinch off cognate antigen after forming an immunological synapse with the target membrane (in this case the virion surface membrane or infected cell). B cells are known to be able to extract antigen from target cell membranes (Batista et al., 2001) or planar membrane surfaces (Fleire et al., 2006) in a BCR mediated process and the acquired antigen induces B cell activation and is efficiently processed and presented to cognate T cells (Batista et al., 2001; Fleire et al., 2006). Concentration of membrane bound antigen by formation of an immunological synapse and subsequent pinching of that antigen would provide a mechanism both to enhance recognition of potential neutralizing antibody targets via increased BCR mediated stimulation, and a mechanism for selective CD4 T cell help. A third potential mechanism for selective B-T interactions would be preferential protection of cognate antigen by the BCR during transport to intracellular antigen processing compartments (Watts, 1997). Further studies are required to discriminate between these mechanisms. Initial attempts to elucidate molecular mechanism were inconclusive due to the low frequencies of antigen-specific B cells in vivo (data not shown), and therefore BCR transgenics will be required to address this issue. Panels of VACV hybridomas have now been generated and characterized (unpublished data), and BCR cloning and transgenic production is underway. It will also be important to test the prediction of B-T linkage for additional large pathogens, and the general power of serological analysis to greatly enhance prediction and identification of novel CD4 T cell epitopes for other large pathogens of interest. It is intriguing to speculate that these findings may also relate to the mechanisms underlying B-T linkage to surface membrane proteins in cancer immunity and autoimmunity (e.g. myasthenia gravis), where the antigen is present is the context of a whole cell.

The data presented herein demonstrate that the cellular mechanism driving the linkage of CD4 T cell and antibody specificities is a B cell requirement for intramolecular protein-specific CD4 T cell help. As such, MHC restriction at the protein level is a key event for humoral immune responses to VACV, as individual protein identities are the primary unit of immunological recognition. This is a powerful principle, as we have shown that it impacts protective immunity and can also be used to predict the presence of novel pathogen-specific CD4 T cell responses. These findings from the smallpox vaccine virus are relevant for understanding the nature of B cell antigen presentation to CD4 T cells and non-self recognition, the prediction of MHC II epitopes, and vaccine development against complex pathogens. Since the smallpox vaccine is the only vaccine to result in eradication of a disease from the human population, immunological results with this virus must be taken quite seriously.

Experimental Procedures

Protein arrays and analysis

16-pad nitrocellulose FAST slides (Whatman, Florham Park, New Jersey) were used for protein microarray printing. Proteins were printed at Scripps (TSRI) using a custom arrayer with 100µm pins built by Robotic Labware Design (RLD). Humidity was maintained at 40–60% during printing. Microarray slides were subsequently dried and stored in a desiccator at −80 °C. Vaccinia genes were cloned into pXi (pNHisCHA derived (Davies et al., 2005b)) and sequenced prior to protein synthesis using the Roche RTS E.col in vitro coupled transcription and translation expression system. RTS reactions without plasmid were used as negative controls. Expression was confirmed by dot blot, western blot, or microarray probing for His tags. Whole virus antigen was printed using 10-fold concentrated, PBS buffer-exchanged, psoralen inactivated (Davies et al., 2005b) standard VACVWR stock. Uninfected HeLa cell lysate processed identically served as negative control (Crotty et al., 2003). Purified His-tagged H3L protein (Davies et al., 2005b) was printed at 100 µg/ml. E. coli produced L1, F9, A21, and A28 were purified and refolded under oxidizing conditions to generate natively folded protein with appropriate disulfide bond formation (Su et al., 2005). Secreted B5R was produced using a baculovirus expression vector in S2 insect cells and subsequently purified utilizing an N-terminal His tag. All antigens were resuspended in 0.02% Tween-20 prior to printing.

Mouse serum samples were initially screened using a large protein microarray covering 185 VACV proteins (Davies et al., 2005b). As only a minority of VACV antigens elicit antibody responses in mice, humans, hyperimmune rabbits, or primates (ref. (Davies et al., 2005a; Davies et al., 2007) and DD, SC, PF unpublished data), the majority of experiments done in this study were done using a compact microarray (Fig. 5) consisting of 55 proteins identified as an antibody target in any species, as well as all known virion surface proteins (as of early 2005). This strategy allowed for an 8-fold increase in the numbers of replicate experiments (16 pad microarray slide vs. 2 pad microarray slide). In addition, after the identification of VACV CD4 T cell targets (Fig. 1), CD4 T cell target proteins without known IgG responses (J4, F15, A20, E1, E9, A24, and A18) were retested in a new set of microarrays to determine if any IgG targets had been missed, and no new IgG targets were detected. In total, using several generations of microarrays, 191 VACV proteins were tested for IgG responses, out of 218 total annotated VACVWR genes.

Proteome microarrays were used to detected antibodies by techniques comparable to a fluorescence ELISA on a microscale. After blocking arrays (Protein Array Blocking Solution, Whatman), microarrays were probed with 1:50 mouse serum diluted in array blocking solution preincubated/preadsorbed for 30min with 10% clarified E. coli lysate. (Clarified E. coli lysate was prepared using 100mg/ml DH5α in PBS, sonicated and then centrifuged at 6000g. Lysate was then stored at −80 °C.) After 2 hr incubation, extensive washing was done with PBS + 0.05% Tween-20, and secondary antibody (Cy3 labelled goat anti-mouse gamma chain Fc region specific immunoglobulin, Jackson Immunoresearch) at 1:50 in array blocking solution was added for 1 hr. Cy5 labeled goat anti-mouse IgM Fc specific F(ab)2 immunoglobulin was also used in some experiments. Arrays were then washed extensively with PBS + 0.05% Tween-20 and PBS alone, and then spun dry. Arrays were scanned on an Axon 400B GenePix (Molecular Dynamics) and data was acquired using GenePix Pro 5.1. Total 532nm fluorescence intensity (TFI532) of each spot was the signal strength. Background signal was subtracted out using relevant matched control samples (e.g. RTS translation reaction without plasmid), and background subtracted signal was converted to the final IgG relative units (RU) via 10−6 transformation. Data is plotted as the average of duplicate protein prints (spots), with the full range shown as the error bar (e.g. Fig. 5A–C). Stringent signal thresholds were established as 1 RU above background, which was greater than 10x the background observed in uninfected animals for individual protein antigens (Fig. 5). Using these experimental conditions, IgG signals detected by protein microarray are linear and correlate tightly with signals observed by conventional endpoint dilution ELISA.

CD4 T cell assays and MHC II epitope screening

Initial VACV epitope screening by IFNγ ELISPOT was described (Moutaftsi et al., 2007). The 2146 15mer peptides represented 31.5% of the VACV genome coding sequence. These peptides were synthesized for various purposes, without consideration of antibody targets or other potential biases, such that the peptide screening library randomly queried sequences throughout the VACV amino acid sequence space. The peptides represented 199 of the 218 annotated VACV genes (18 of 19 missed ORFs are less than 80 a.a. in length). 30% of total structural protein sequence and 31% of total nonstructural protein sequence was represented in the 2146 peptides (32,190 a.a.). The percentage of the anti-VACV CD4 T cell response identified using these peptides (26–33%, Fig. 1B) was consistent with the random distribution and overall VACV sequence coverage of the peptides (31.5%). In addition, we could retroactively examine the peptide library contents for bias in coverage of known antibody targets, and no bias is present (30.7% a.a. coverage of antibody targets vs. 30.4% coverage of other), independently confirming the unbiased nature of the peptide library used.

The overlapping peptides used in Fig. 6 were synthesized as 15mers overlapping by 5 a.a. on each end. Increased epitope identification predictive power via utilization of antibody targets (Fig. 6) was calculated by summing the CD4 T cell responses to each epitope of interest (Fig. 1B and Fig. 6A, experiments done at the same time) and then dividing by the total anti-VACV CD4 T cell response (Fig. 1B). This was then divided by the number of peptides screened in each group.

Intracellular cytokine staining of splenocytes from VACVWR infected mice (day 7 or 10 postinfection) was done by incubation of splenocytes with peptide-pulsed CD11c+ DC for 1h prior to addition of brefeldin A. After 5hr, cells were surface stained for CD62L and CD4 surface molecules, followed by intracellular staining for IFNγ and CD40L. IFNγ -producing/CD40L expressing cells were determined by gating lymphocytes on FSC/SSC followed by gating on CD62L low/CD4+ T-cells. CD11c+ DC infected with VACVWR (MOI=5) for 2hr prior to addition of splenocytes were used to identify the total anti-VACV CD4+ T cell response. DCs were generated by subcutaneous implantation of Flt3L producing B16 cells, followed by harvesting CD11c+ DCs from spleen at day 12 post-implantation, using CD11c paramagnetic beads (MACS Miltenyi). Background level was established using CD4 T cell values in the presence of uninfected CD11c+ DC. All antibodies were purchased from eBiosciences (San Diego, CA) or BD Pharmingen (San Diego, CA).

Mouse procedures and viral infections

C57BL/6J (B6), MHC class II−/− (C57BL/6J Iab-Ea−/−), and B cell deficient (C57BL/6J µMT) mice were purchased from the Jackson Laboratory and bred in house. VACVWR stocks were grown on HeLa cells, infecting at a multiplicity of infection of 0.5 (MOI = 0.5). Cells were harvested at 60hrs, and virus was isolated by rapid freeze-thawing the cell pellet 3x in a volume of 2.3 ml RPMI + 1% FCS per T175 flask. Cell debris was removed by centrifugation. Clarified supernatant was frozen at −80 °C as virus stock (Davies et al., 2005b). Stocks were titered on VeroE6 cells. For all experiments except Fig. 6D–F, mice were infected with VACVWR by bilateral intraperitoneal (i.p.) injections of 2 × 105 or 2 × 106 total PFU total using standard VACVWR stocks. Replicates of Figure 1, 2, 4, and 5 experiments were also performed using purified VACVWR for infections; no differences in the results were observed. Purified VACVWR stocks were made by centrifuging standard VACVWR stock through a 36% sucrose cushion in PBS plus 5 mM MgCl2 and resuspension of the virion pellet in RPMI + 1% FCS. For peptide immunizations, 30 µg peptide (or PBS control) was emulsified in CFA. Subcutaneous injections were done dorsal to the base of the tail; subcutaneous injections for some repeat experiments were done between the scapula. VACVWR i.p. infections were done 11–13 days after peptide priming. For adoptive transfers, untouched SMtg+ CD4 T cells were magnetically purified from spleen and lymph node preparations (MACS Miltenyi, Auburn, CA). 50 × 106 CD4 T cells were transferred per mouse, resulting in approximately one half mouse equivalent of donor CD4 T cells after take (e.g., 5 × 106 donor CD4 T cells/spleen). For viral challenge experiments (Fig. 6D–F), 8 week old female mice were infected intranasally with 1 ×104 PFU VACVWR. Virus was placed in a volume of 10 µl on the nares of lightly sedated mice, which was then inhaled by the mice. Mice were weighed daily to track disease progression. Dose titration experiments established 1 × 104 PFU VACVWR to be approximately 1 LD50 in 8 week old female C57BL/6 mice (unpublished data).

ELISA

VACV ELISAs were done as described (Davies et al., 2005b), with the additional use in Figure 1 of standard curves using mouse IgG (Southern Biotech, Birmingham, AL) in anti-mouse Ig (goat anti-mouse IgM+IgG+IgA, Caltag) coated wells. B5 ELISA used recombinant B5.

Neutralization assays

VeroE6 cells were seeded into 24-well Costar plates (Corning Inc, Corning, NY) and used within 24 hours of reaching 90% confluency. Mouse serum samples were collected by retro-orbital bleed from B6 mice at various time points postinfection. All serum samples were heat-inactivated prior to use (56°C, 30 min). Diluted sera were incubated in an equal volume of sonicated sucrose gradient purified VACVWR (104 PFU/ml) overnight at 37°C, 5 % CO2 (Newman et al., 2003). Negative control sera from untreated mice were also done at the same conditions. The medium from 24-well plate was aspirated and 100µl of virus/serum mixture added to each well and left to adsorb for 60 minutes at 37°C with periodic swirling. The infected VeroE6 cells were rinsed with warm PBS. One ml of complete D-10 medium (DMEM + 15% FBS + pen/strep + L-glut) was then added and the plates incubated for 40–50 hrs. Medium was then aspirated and cells fixed and stained in one step with 0.1% crystal violet in 20% ethanol, and plaques quantified over white light transillumination. 50% plaque reduction neutralization titer (PRNT50) was defined as the furthest dilution of each sample to neutralize > 50% the virus according to the formula: [(PN VACV – PN sample) / PN VACV], where “PN VACV” is the average number of plaques in wells infected with VACVWR alone, and “PN sample” is the average number of plaques for a serum sample at a given dilution.

Statistics

Tests were performed using Prism 4.0 (GraphPad, San Diego, CA) or VassarStats (Fisher Exact Test). Statistics were done using two-tailed, unpaired Student’s T test with 95% confidence bounds unless otherwise indicated. Bar graph error bars are ± one SEM, except for raw microarray data, which shows full range (e.g. Fig. 5A–C). Arithmetic means were used for all analyses. CD4 T cell – antibody linkage (Fig. 5E) was analyzed using Fisher’s Exact Test with the null hypothesis being no increased frequency of antibody targets among the CD4 T cell target proteins. Six CD4 T cell target proteins were IgG targets, 7 not. Twenty-three of 191 VACV proteins screened were IgG targets. Statistical analysis of mouse weight loss after intranasal infection with VACVWR (Fig. 6D–E) was done by tabulating the weight nadir (or 70% for dead mice, which was the maximum weight loss before euthanization) for each mouse and then comparing each experimental group by standard two-tailed, unpaired Student’s T test with 95% confidence bounds. Survival curves significance were calculated using Kaplan-Meier statistical analysis.

Acknowledgements

We thank Steve Granger, Lilia Koriazova, and Kirin Pharma for providing B5R protein. This work was supported in part by NIH grants, a Pew Scholar Award, and a Cancer Research Institute Award to SC and an NIH grant to AS.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Amanna IJ, Slifka MK, Crotty S. Immunity and immunological memory following smallpox vaccination. Immunol Rev. 2006;211:320–337. doi: 10.1111/j.0105-2896.2006.00392.x. [DOI] [PubMed] [Google Scholar]
  2. Bachmann MF, Zinkernagel RM. Neutralizing antiviral B cell responses. Annu Rev Immunol. 1997;15:235–270. doi: 10.1146/annurev.immunol.15.1.235. [DOI] [PubMed] [Google Scholar]
  3. Balazs M, Martin F, Zhou T, Kearney J. Blood dendritic cells interact with splenic marginal zone B cells to initiate T-independent immune responses. Immunity. 2002;17:341–352. doi: 10.1016/s1074-7613(02)00389-8. [DOI] [PubMed] [Google Scholar]
  4. Batista FD, Iber D, Neuberger MS. B cells acquire antigen from target cells after synapse formation. Nature. 2001;411:489–494. doi: 10.1038/35078099. [DOI] [PubMed] [Google Scholar]
  5. Belyakov IM, Earl P, Dzutsev A, Kuznetsov VA, Lemon M, Wyatt LS, Snyder JT, Ahlers JD, Franchini G, Moss B, Berzofsky JA. Shared modes of protection against poxvirus infection by attenuated and conventional smallpox vaccine viruses. Proc Natl Acad Sci U S A. 2003;100:9458–9463. doi: 10.1073/pnas.1233578100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Briody BA. Response of mice to ectromelia and vaccinia viruses. Bacteriol Rev. 1959;23:61–95. doi: 10.1128/br.23.2.61-95.1959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carrasco YR, Batista FD. B cells acquire particulate antigen in a macrophage-rich area at the boundary between the follicle and the subcapsular sinus of the lymph node. Immunity. 2007;27:160–171. doi: 10.1016/j.immuni.2007.06.007. [DOI] [PubMed] [Google Scholar]
  8. Chung CS, Chen CH, Ho MY, Huang CY, Liao CL, Chang W. Vaccinia virus proteome: identification of proteins in vaccinia virus intracellular mature virion particles. J Virol. 2006;80:2127–2140. doi: 10.1128/JVI.80.5.2127-2140.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Condit RC, Moussatche N, Traktman P. In a nutshell: structure and assembly of the vaccinia virion. Adv Virus Res. 2006;66:31–124. doi: 10.1016/S0065-3527(06)66002-8. [DOI] [PubMed] [Google Scholar]
  10. Crotty S, Felgner P, Davies H, Glidewell J, Villarreal L, Ahmed R. Cutting edge: long-term B cell memory in humans after smallpox vaccination. J Immunol. 2003;171:4969–4973. doi: 10.4049/jimmunol.171.10.4969. [DOI] [PubMed] [Google Scholar]
  11. Cyrklaff M, Risco C, Fernandez JJ, Jimenez MV, Esteban M, Baumeister W, Carrascosa JL. Cryo-electron tomography of vaccinia virus. Proc Natl Acad Sci U S A. 2005;102:2772–2777. doi: 10.1073/pnas.0409825102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Davies DH, Liang X, Hernandez JE, Randall A, Hirst S, Mu Y, Romero KM, Nguyen TT, Kalantari-Dehaghi M, Crotty S, et al. Profiling the humoral immune response to infection by using proteome microarrays: high-throughput vaccine and diagnostic antigen discovery. Proc Natl Acad Sci U S A. 2005a;102:547–552. doi: 10.1073/pnas.0408782102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Davies DH, McCausland MM, Valdez C, Huynh D, Hernandez JE, Mu Y, Hirst S, Villarreal L, Felgner PL, Crotty S. Vaccinia virus H3L envelope protein is a major target of neutralizing antibodies in humans and elicits protection against lethal challenge in mice. J Virol. 2005b;79:11724–11733. doi: 10.1128/JVI.79.18.11724-11733.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Davies DH, Molina DM, Wrammert J, Miller J, Hirst S, Mu Y, Pablo J, Unal B, Nakajima-Sasaki R, Liang X, et al. Proteome-wide analysis of the serological response to vaccinia and smallpox. Proteomics. 2007;7:1678–1686. doi: 10.1002/pmic.200600926. [DOI] [PubMed] [Google Scholar]
  15. Demkowicz WE, Maa JS, Esteban M. Identification and characterization of vaccinia virus genes encoding proteins that are highly antigenic in animals and are immunodominant in vaccinated humans. J Virol. 1992;66:386–398. doi: 10.1128/jvi.66.1.386-398.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Edghill-Smith Y, Golding H, Manischewitz J, King LR, Scott D, Bray M, Nalca A, Hooper JW, Whitehouse CA, Schmitz JE, et al. Smallpox vaccine-induced antibodies are necessary and sufficient for protection against monkeypox virus. Nat Med. 2005;11:740–747. doi: 10.1038/nm1261. [DOI] [PubMed] [Google Scholar]
  17. Fang M, Sigal LJ. Antibodies and CD8+ T cells are complementary and essential for natural resistance to a highly lethal cytopathic virus. J Immunol. 2005;175:6829–6836. doi: 10.4049/jimmunol.175.10.6829. [DOI] [PubMed] [Google Scholar]
  18. Fenner F, Henderson DA, Arita I, Jezek Z, Ladnyi ID. Geneva: World Health Organization; Smallpox and its eradication. 1988
  19. Fleire SJ, Goldman JP, Carrasco YR, Weber M, Bray D, Batista FD. B cell ligand discrimination through a spreading and contraction response. Science. 2006;312:738–741. doi: 10.1126/science.1123940. [DOI] [PubMed] [Google Scholar]
  20. Galmiche MC, Goenaga J, Wittek R, Rindisbacher L. Neutralizing and protective antibodies directed against vaccinia virus envelope antigens. Virology. 1999;254:71–80. doi: 10.1006/viro.1998.9516. [DOI] [PubMed] [Google Scholar]
  21. Goldstein JL, Anderson RG, Brown MS. Coated pits, coated vesicles, and receptor-mediated endocytosis. Nature. 1979;279:679–685. doi: 10.1038/279679a0. [DOI] [PubMed] [Google Scholar]
  22. Harrington LE, van der Most RG, Whitton JL, Ahmed R. Recombinant vaccinia virus-induced T-cell immunity: quantitation of the response to the virus vector and the foreign epitope. J Virol. 2002;76:3329–3337. doi: 10.1128/JVI.76.7.3329-3337.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hickman HD, Takeda K, Skon CN, Murray FR, Hensley SE, Loomis J, Barber GN, Bennink JR, Yewdell JW. Direct priming of antiviral CD8+ T cells in the peripheral interfollicular region of lymph nodes. Nat Immunol. 2008;9:155–165. doi: 10.1038/ni1557. [DOI] [PubMed] [Google Scholar]
  24. Hollinger FB, Liang TJ. Hepatitis B Virus. In: Knipe D, Howley P, editors. Fields Virology. Philadelphia: Lippincott Williams; 2001. pp. 2971–3036. [Google Scholar]
  25. Janeway CA, Travers P, Walport M, Shlomchik MJ. Immunobiology : the immune system in health and disease. 6th edn. New York: Garland Science; 2005. [Google Scholar]
  26. Junt T, Moseman EA, Iannacone M, Massberg S, Lang PA, Boes M, Fink K, Henrickson SE, Shayakhmetov DM, Di Paolo NC, et al. Subcapsular sinus macrophages in lymph nodes clear lymph-borne viruses and present them to antiviral B cells. Nature. 2007;450:110–114. doi: 10.1038/nature06287. [DOI] [PubMed] [Google Scholar]
  27. Knipe D, Howley P. Fields Virology. 4 edn. Philadelphia, PA: Lippincott; 2001. [Google Scholar]
  28. Kwon DS, Gregorio G, Bitton N, Hendrickson WA, Littman DR. DC-SIGN-mediated internalization of HIV is required for trans-enhancement of T cell infection. Immunity. 2002;16:135–144. doi: 10.1016/s1074-7613(02)00259-5. [DOI] [PubMed] [Google Scholar]
  29. Lake P, Mitchison NA. Regulatory Mechanisms in Immune-Response to Cell-Surface Antigens. Cold Spring Harbor Symposia on Quantitative Biology. 1976;41:589–595. doi: 10.1101/sqb.1977.041.01.068. [DOI] [PubMed] [Google Scholar]
  30. Lin CL, Chung CS, Heine HG, Chang W. Vaccinia virus envelope H3L protein binds to cell surface heparan sulfate and is important for intracellular mature virion morphogenesis and virus infection in vitro and in vivo. J Virol. 2000;74:3353–3365. doi: 10.1128/jvi.74.7.3353-3365.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lodish H, Berk A, Matsudaira P, Kaiser C, Krieger M, Scott M, Zipursky L, Darnell J. Molecular Cell Biology. 5 edn. W.H. Freeman; 2003. [Google Scholar]
  32. Lustig S, Fogg C, Whitbeck JC, Eisenberg RJ, Cohen GH, Moss B. Combinations of polyclonal or monoclonal antibodies to proteins of the outer membranes of the two infectious forms of vaccinia virus protect mice against a lethal respiratory challenge. J Virol. 2005;79:13454–13462. doi: 10.1128/JVI.79.21.13454-13462.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Macpherson AJ, Uhr T. Induction of protective IgA by intestinal dendritic cells carrying commensal bacteria. Science. 2004;303:1662–1665. doi: 10.1126/science.1091334. [DOI] [PubMed] [Google Scholar]
  34. Milich DR, McLachlan A, Thornton GB, Hughes JL. Antibody production to the nucleocapsid and envelope of the hepatitis B virus primed by a single synthetic T cell site. Nature. 1987;329:547–549. doi: 10.1038/329547a0. [DOI] [PubMed] [Google Scholar]
  35. Moss B. Poxviridae: The Viruses and their Replication. In: Knipe DM, Howley P, editors. Fundamental Virology. 2001. [Google Scholar]
  36. Moutaftsi M, Bui HH, Peters B, Sidney J, Salek-Ardakani S, Oseroff C, Pasquetto V, Crotty S, Croft M, Lefkowitz EJ, et al. Vaccinia virus-specific CD4+ T cell responses target a set of antigens largely distinct from those targeted by CD8+ T cell responses. J Immunol. 2007;178:6814–6820. doi: 10.4049/jimmunol.178.11.6814. [DOI] [PubMed] [Google Scholar]
  37. Newman FK, Frey SE, Blevins TP, Mandava M, Bonifacio A, Jr, Yan L, Belshe RB. Improved assay to detect neutralizing antibody following vaccination with diluted or undiluted vaccinia (Dryvax) vaccine. J Clin Microbiol. 2003;41:3154–3157. doi: 10.1128/JCM.41.7.3154-3157.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pape KA, Catron DM, Itano AA, Jenkins MK. The humoral immune response is initiated in lymph nodes by B cells that acquire soluble antigen directly in the follicles. Immunity. 2007;26:491–502. doi: 10.1016/j.immuni.2007.02.011. [DOI] [PubMed] [Google Scholar]
  39. Phan TG, Grigorova I, Okada T, Cyster JG. Subcapsular encounter and complement-dependent transport of immune complexes by lymph node B cells. Nat Immunol. 2007;8:992–1000. doi: 10.1038/ni1494. [DOI] [PubMed] [Google Scholar]
  40. Qi H, Egen JG, Huang AY, Germain RN. Extrafollicular activation of lymph node B cells by antigen-bearing dendritic cells. Science. 2006;312:1672–1676. doi: 10.1126/science.1125703. [DOI] [PubMed] [Google Scholar]
  41. Rammensee H, Bachmann J, Emmerich NP, Bachor OA, Stevanovic S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics. 1999;50:213–219. doi: 10.1007/s002510050595. [DOI] [PubMed] [Google Scholar]
  42. Rappuoli R, Miller HI, Falkow S. Medicine. The intangible value of vaccination. Science. 2002;297:937–939. doi: 10.1126/science.1075173. [DOI] [PubMed] [Google Scholar]
  43. Resch W, Hixson KK, Moore RJ, Lipton MS, Moss B. Protein composition of the vaccinia virus mature virion. Virology. 2006 doi: 10.1016/j.virol.2006.08.025. [DOI] [PubMed] [Google Scholar]
  44. Roos N, Cyrklaff M, Cudmore S, Blasco R, Krijnse-Locker J, Griffiths G. A novel immunogold cryoelectron microscopic approach to investigate the structure of the intracellular and extracellular forms of vaccinia virus. Embo J. 1996;15:2343–2355. [PMC free article] [PubMed] [Google Scholar]
  45. Russell SM, Liew FY. T cells primed by influenza virion internal components can cooperate in the antibody response to haemagglutinin. Nature. 1979;280:147–148. doi: 10.1038/280147a0. [DOI] [PubMed] [Google Scholar]
  46. Scherle PA, Gerhard W. Functional analysis of influenza-specific helper T cell clones in vivo. T cells specific for internal viral proteins provide cognate help for B cell responses to hemagglutinin. J Exp Med. 1986;164:1114–1128. doi: 10.1084/jem.164.4.1114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sette A, Sidney J, Oseroff C, del Guercio MF, Southwood S, Arrhenius T, Powell MF, Colon SM, Gaeta FC, Grey HM. HLA DR4w4-binding motifs illustrate the biochemical basis of degeneracy and specificity in peptide-DR interactions. J Immunol. 1993;151:3163–3170. [PubMed] [Google Scholar]
  48. Su HP, Garman SC, Allison TJ, Fogg C, Moss B, Garboczi DN. The 1.51-Angstrom structure of the poxvirus L1 protein, a target of potent neutralizing antibodies. Proc Natl Acad Sci U S A. 2005;102:4240–4245. doi: 10.1073/pnas.0501103102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tscharke DC, Karupiah G, Zhou J, Palmore T, Irvine KR, Haeryfar SM, Williams S, Sidney J, Sette A, Bennink JR, Yewdell JW. Identification of poxvirus CD8+ T cell determinants to enable rational design and characterization of smallpox vaccines. J Exp Med. 2005;201:95–104. doi: 10.1084/jem.20041912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Watts C. Capture and processing of exogenous antigens for presentation on MHC molecules. Annu Rev Immunol. 1997;15:821–850. doi: 10.1146/annurev.immunol.15.1.821. [DOI] [PubMed] [Google Scholar]
  51. West MA, Lucocq JM, Watts C. Antigen processing and class II MHC peptide-loading compartments in human B-lymphoblastoid cells. Nature. 1994;369:147–151. doi: 10.1038/369147a0. [DOI] [PubMed] [Google Scholar]
  52. Wykes M, Pombo A, Jenkins C, MacPherson GG. Dendritic cells interact directly with naive B lymphocytes to transfer antigen and initiate class switching in a primary T-dependent response. J Immunol. 1998;161:1313–1319. [PubMed] [Google Scholar]
  53. Xu R, Johnson AJ, Liggitt D, Bevan MJ. Cellular and humoral immunity against vaccinia virus infection of mice. J Immunol. 2004;172:6265–6271. doi: 10.4049/jimmunol.172.10.6265. [DOI] [PubMed] [Google Scholar]

RESOURCES