|Home | About | Journals | Submit | Contact Us | Français|
The antibody response elicited after immunization with vaccinia virus (VacV) is known to be sufficient to confer host protection against VacV or smallpox. In humans it has been shown that such anti-VacV antibody production can be sustained for decades. Nevertheless, little is known about the kinetics and the role in protection of the early antibody response after vaccination. In this study we identify VacV neutralizing IgM antibodies as early as four days after infection of C57BL/6 mice. Most of this IgM production is T cell dependent and predominantly independent of the germinal center reaction (SAP / SH2D1A independent). Importantly, the IgM neutralized both infectious forms of VacV: the intracellular mature virion (MV, IMV) and the extracellular enveloped virion (EV, EEV). Moreover, in mice primed with MHCII restricted peptides, an increase in the total VacV neutralizing antibody titers was seen, a large component of which was neutralizing IgM against the same protein from which the priming peptide was derived. To further demonstrate the biological relevance of this early neutralizing response, we examined anti-VacV antibodies in humans after vaccination. Human subjects could be divided into two groups early after immunization: IgGhi and IgGlo. VacV IgM neutralizing antibodies were detected in the IgGlo group. Taken together these results indicate that both in a small animal model and in humans an early neutralizing IgM response after VacV immunization is present and likely contributes to control of the infection prior to the development of a robust IgG response.
Immunization with the smallpox vaccine results in long term protection against VacV itself and other orthopoxviruses, most importantly variola (smallpox). The protective activity is mainly due to anti-VacV neutralizing antibodies , which are continuously produced for decades after vaccination [2-5]. Many studies have described roles of different innate and adaptive immune response cell types in the control of a primary VacV infection or vaccination [1, 6, 7]. However, while B cells and antibodies are crucial for the long term protection, the roles of B cells and antibodies in the primary response to the smallpox vaccine remain less clear [1, 8].
Mice deficient of B cells (μMT/μMT) have an increased susceptibility to VacV infection, and higher viral titers are found in μMT mice compared to wild-type (WT) controls , suggesting that the early antibody production has an important role in the control of the infection. Furthermore, antibodies are essential for control and clearance of the related poxvirus, ectromelia (mousepox) [9-11], while T cell responses are also critical for control of a primary ectromelia infection [9, 12]. VacV has two different infectious forms, the intracellular mature virion (IMV, MV) thought to be primarily responsible from host to host infection, and the extracellular enveloped virion (EEV, EV), responsible for much of the viral dissemination within the host . Studies have demonstrated that antibodies directed against either virion form are protective in mice [14-17], and we have recently demonstrated that B5-specific anti-EV antibodies can efficiently clear virally infected cells, illustrating that the protective value of anti-VacV antibodies is not limited to free virions . Moreover, in the rhesus macaque model it has been shown that B cells but not CD4+ nor CD8+ T cells are essential for protection against a lethal challenge with monkeypox virus . All of these reports highlight the importance of antibodies in the control of VacV and related poxviruses.
While it is clear that neutralizing antibodies are a very potent mechanism for protection against VacV, monkeypox, and variola, the neutralizing IgG response is relatively slow to develop after a primary infection or immunization. In vaccinated humans, neutralizing IgG is reported at 2-3 weeks after vaccination [1, 20, 21], after the resolution of the primary infection and lesion. In mice, significant anti-VacV IgG titers have not been reported earlier than 8 days post-infection (i.p.), which is after the bulk of the infection has been cleared. Therefore, it appears that the neutralizing IgG response develops too late in humans and mice to be of significant value in controlling a primary VacV infection/immunization. Given that, is there any role for neutralizing antibodies in control of the primary VacV infection?
The study reported here demonstrates for the first time that there is an early induction of neutralizing IgM after VacV immunization of mice and humans, and implicates a protective role for IgM during the primary response.
Mice were infected with the vaccinia virus Western Reserve strain (VacV) by bilateral intraperitoneal (i.p.) injection of 2×106 PFU total with standard purified VacV stocks. For peptide immunizations, 30 μg of peptide was emulsified in complete Freund adjuvant (CFA) and injected subcutaneously between the scapula, and 11–13 days after peptide immunization were infected with VacV. Serum was obtained by retro-orbital bleed at determined time points post-infection. To determine the kinetics of the appearance of neutralizing antibodies, two groups of WT mice were infected with VacV, one group was bled on odd days and the other was bled on even days. All mice were maintained in an accredited facility at LIAI, and all the experiments were conducted in accordance with approved animal protocols.
A cohort of 14 normal healthy volunteer donors were immunized for the first time with the smallpox vaccine (Dryvax) and blood samples were obtained at the indicated time post-vaccination. The gender distribution of the cohort was□□~50:50, within an age range from 23-60 years.
Plasma samples from the human subjects and sera from VacV infected mice was stored as aliquots at −80°C. To eliminate IgM antibodies, aliquots were treated with an equal volume of 0.1M 2-mercaptoethanol (2-ME) in PBS for 1 h at room temperature, as described [24-26].
VacV MV stocks were produced as described in . Briefly, MV were in grown on HeLa cells in D-10 (Dulbecco's modified Eagle medium [DMEM] plus 10% heat inactivated fetal calf serum [FCS] plus penicillin/streptomycin/glutamine) in T175 flasks (Falcon; Becton Dickinson), infecting at a multiplicity of infection (MOI) of 0.1 to 0.5. Cells were harvested at 2.5 to 3 days, and virus was isolated by rapidly freeze-thawing the cell pellet three times in a volume of 2.3 ml DMEM or RPMI supplemented with 1% heat inactivated FCS. Cell debris was removed by centrifugation (700×g, 8 min). Clarified supernatant was frozen at −80°C as virus stock. Titers of VacV stocks were determined with VeroE6 cells (~2×108 PFU/ml). Purified VacV stocks were made by sonication of the VacV stock (40 s) using a water sonicator (Branson Ultrasonics, CT) and layered over 36% sucrose in TM buffer (10 mM Tris-HCl [pH 7.4], 5 mM MgCl2). VacV was centrifuged (SW28 rotor) at 13,500 rpm (33,000×g) for 80 min at 4°C. The VacV pellet was resuspended in 1 ml TM buffer and then brought up to 10 ml with DMEM medium supplemented with 1% of heat inactivated FCS. Purified VacV was stored at −80°C.
VacV EV was prepared as described elsewhere . Briefly, HeLa cells were cultured in D-10 in T75 flasks (Falcon, Becton Dickinson) at 90% confluence and infected with VacV at a MOI of 0.5. The medium containing EV was harvested at 2 days, and virus was isolated by centrifugating twice (450 × g, 8 min) to remove cells and debris. Clarified supernatant was used immediately or stored at 4 °C for a maximum of 3 weeks. EV VacV stocks were titrated on VeroE6 cells (~5 × 105 PFU/ml).
Titration of VacV MV neutralizing antibodies, was performed according to Newman et al. . Briefly, VeroE6 cells were seeded at 2×105 cells/well into 24-well Costar plates (Corning, Inc., Corning, NY) and used the following day (75 to 90% confluence). Total or IgM depleted plasma or serum was incubated overnight with 50 μl of freshly sonicated VacV (104 PFU/ml) at 37°C with 5% CO2. Plasma from non-vaccinated human individuals or sera from naive mice were treated under the same conditions and used as negative controls. Multiple wells of VacV-alone controls were always used. Medium was aspirated, and the samples were added and allowed to adsorb for 60 min at 37°C. Then the cells were rinsed with warm phosphate-buffered saline (PBS) and 1 ml of D-10 medium was added. Plates were incubated for 40 to 48 h and then fixed and stained with 0.1% crystal violet in 25% reagent alcohol (90% ethanol, 5% methanol, 5% isopropanol) to count viral plaques.
VacV EV neutralization assay was described elsewhere . Briefly, all the samples and FCS were heat inactivated (56°C, 30-60 min) prior to use to eliminate complement function. VeroE6 cells were prepared in 24-well Costar plates as described. The samples were incubated for 30 min at 37°C with an equal volume (50 μl) of EV stock (1:100 - 1:400 dilution) supplemented with 10% (final concentration) sterile baby rabbit complement (Cedarlane Laboratories, Ontario, Canada) and rabbit anti-L1 (1:25-1:100 final). Anti-L1 was used to neutralize the MV present in the EV stock [18, 29]. EV supplemented with anti-L1 antibody alone was regularly used in each assay +/− baby rabbit complement as negative controls. Medium from 24-well plate wells was aspirated and samples were added and allowed to adsorb for 45 minutes at 37°C. After this step the plates were treated as in the MV neutralization assay.
In both neutralization assays the PRNT50 was defined as the reciprocal of the last dilution of the plasma that reduced the average number of plaques by 50% compared to the mean number of VacV-alone plaques.
Production and use of protein microarrays is described elsewhere [17, 27, 30]. Briefly, VacV proteome arrays were probed with the sera or plasma samples and bound antibodies were detected with a Cy3-conjugated goat anti-mouse IgG (heavy and light chains) secondary antibody (Jackson ImmunoResearch). The arrays were examined in a GSI Lumonics ScanArray 4000 confocal glass slide scanner and intensities were quantified using QuantArray software. Results were quantified as relative fluorescent units (RU) over background. Signals from three negative control spots were averaged, and that background signal was subtracted from all spots on the array to give RU.
For whole VacV enzyme-linked immunosorbent assay (ELISA), the antigen preparation was made by incubating unpurified VacV (108 PFU/ml) in 0.1% bovine serum albumin with 10 μg/ml trioxsalen-psoralen (4’ aminomethyl-trioxsalen HCl; Calbiochem) for 10 min at RT . VacV was then UV inactivated with 2.25 J/cm2 (Stratalinker 1800; Stratagene, CA). This resulted in a >108-fold reduction in PFU. The UV-inactivated virus was then used at a 1:25 dilution in PBS to coat Nunc Polysorp flat-bottomed 96-well plates. Plates were washed and samples were added in PBS plus 0.05% Tween-20 plus 10% FCS (PBS-T+FCS). Horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (Caltag) or biotinylated anti-human IgG (BD-Pharmingen) followed by HRP-strepatvidin (Vector labs) diluted in PBS-T+FCS were used for detection of murine and human IgG respectively. For human IgM detection, biotinylated anti-human IgM (BD-Pharmingen) followed by HRP-strepatvidin diluted in PBS-T+FCS was used. The plates were developed using o-phenylenediamine and the optical density (OD) at 490 nm was read on a SpectraMax 250 (Molecular Devices).
For B5 ELISA, the NUNC Polysorp flat-bottomed 96-well plates were coated overnight with 50 μl of a 2 μg/ml solution of recombinant B5 protein in PBS and processed as described.
Surface staining of mouse splenocytes was done as previously described . Intracellular staining was done as previously described [32, 33]. Briefly, DCs were generated by subcutaneous implantation of Flt3L-producing B16 cells and harvesting CD11c+ DCs from spleen at day 12 after implantation with CD11c paramagnetic beads (MACS Miltenyi). Splenocytes from B546-60-immunized mice (11-13 days after immunization) were incubated with peptide-pulsed CD11c+ DC for 1 hr prior to addition of brefeldin A. After 5 hr, cells were surface stained for CD62L and CD4, followed by intracellular staining for CD40L and IFN-γ or CD40L and IL-2. Cells expressing CD40L and producing IFN-γ or IL-2 were determined by gating lymphocytes on FSC/SSC and then gating on CD62Llo/CD4+ T cells. All antibodies for flow cytometry were purchased from eBiosciences or BD PharMingen.
Tests were performed using Prism 5.0 (GraphPad, San Diego, CA). Statistics were done using two-tailed, unpaired T test with 95% confidence bounds. Error bars are ± SEM.
Intraperitoneal VacV infection of mice results in a potent adaptive immune response that leads to virus clearance from most tissues within seven days. Using VacV proteome microarrays, we were able to identify the first IgG specificities elicited by the virus infection [17, 33, 34]. While IgG responses were detected in most mice, a low percentage of mice exhibited very low or undetectable IgG titers at day 8 post-infection (Figure 1A), while all mice exhibited consistently high IgG titers by day 15 after infection. However, we observed that even day 8 serum samples from mice with very low anti-VacV IgG titers (Figure 1A) still consistently neutralized VacV in vitro (Figure 1B). This observation led us to hypothesize that a substantial early anti-VacV IgM neutralizing antibody response was generated after immunization with VacV, before an IgG response, and this early response contributed to host protection. To test our hypothesis, we then proceeded to determine the kinetics of the appearance of anti-VacV antibodies after infection and the effects of IgM depletion on VacV neutralization.
Two groups of B6 mice were infected i.p. with VacV and bled at different days to determine neutralizing antibodies in serum (Figure 2). Virus neutralization was first detected at day 4 post-infection; after that, the neutralizing antibody titers increased daily. IgM, but not IgG, can be fully inactivated by treatment with 0.1M 2-mercaptoethanol [24-26]. At days 4 and 5 post-infection all the VacV neutralization was due to IgM antibodies, as the elimination of IgM resulted in no detectable neutralizing activity (Day 4, P = 0.0153. Day 5, P = 0.0338. Figure 2). Day 6 was the earliest time point that VacV neutralizing IgG antibodies were detected (Figure 2). These results prove that the earliest anti-VacV neutralizing antibodies are of the IgM isotype.
Previously we have shown that almost all of the anti-VacV IgG response is T cell help dependent (TD) . Nevertheless, others have detected the presence of some T cell independent (TI) anti-VacV antibodies in the sera of infected mice [6, 26]. Therefore, we used CD4 T cell deficient mice (MHCII−/−) to evaluate the T cell help requirement for production of anti-VacV neutralizing IgM. No anti-VacV IgG was found in MHCII−/− mice (Figure 3A). In addition there was very low VacV neutralization capacity of sera from class II−/− mice compared to WT mice (Figure 3B). By depleting IgM we determined that 91% of the VacV neutralizing activity in the WT mice was due to IgM antibodies (P = 0.0039), similarly to what we observed before (Figure 2). The small amount of anti-VacV neutralizing antibody detected in the sera of MHCII−/− mice dropped below the limit of detection of the neutralization test after IgM depletion (Figure 3B), indicating that the few neutralizing antibodies generated in MHCII−/− mice were IgM.
To confirm that the treatment of the sera with 2-ME only inactivated the IgM, leaving the IgG antibodies intact, we measured VacV neutralization in sera from WT mice 64 days after VacV infection, with or without IgM inactivation (2-ME treated). At memory time points, high levels of anti-VacV IgG are present and no anti-VacV IgM is detectable (data not shown). As expected, the IgM inactivation resulted in no significant reduction in neutralizing antibodies (Figure 3C. P >>0.05). Altogether, these results show that the majority of the early anti-VacV IgM neutralizing antibody production, and all of the IgG neutralizing antibody production, is the result of cognate T cell help to VacV-specific B lymphocytes.
Once the B lymphocytes receive T cell help in vivo there are two possible outcomes. One is a rapid B lymphocyte differentiation to short lived plasma cells at an extrafollicular site, and the other is the initiation of a germinal center (GC) reaction [35-38]. Given the kinetics, it was unlikely that the anti-VacV IgM was GC-dependent. To formally dissect the nature of the early anti-VacV IgM response, we utilized SAP−/− (sh2d1a−/−) mice. SAP-deficient mice lack germinal center CD4 T cell help and no germinal center reaction occurs [23, 39, 40], but extrafollilcular T-dependent antibody responses can still occur . When SAP−/− mice were infected with VacV, IgG antibodies were detected at day 8 post-infection. This IgG production faded with time (Figure 4A). In contrast, the anti-VacV IgG titers in the WT mice increased with time(Figure 4A). These results are in agreement with previous reports of SAP−/− mice infected with LCMV [39, 41], or immunized with SRBC . There was no statistical difference in VacV neutralizing antibodies between WT and SAP−/− mice at day 8 (Figure 4B). IgM depletion resulted in a pronounced diminution of the VacV neutralizing titers in both cases (~70%, WT, P = 0.0178. SAP−/−, P = 0.0388, Figure 4B). SAP−/− mice make no germinal centers after infection with VacV, in contrast to the robust germinal center response observed in WT mice (Fig. 4C-D. WT vs. SAP−/−, P < 0.0001). These results demonstrate that the VacV-neutralizing IgM response is CD4+ T cell dependent but precedes the germinal center reaction.
In addition to VacV neutralizing anti-MV IgG antibodies, immunized mice also make neutralizing anti-EV IgG antibodies. Having shown that mice make anti-MV VacV neutralizing IgM responses (Figures 2--4),4), we then tested whether immunized mice make anti-EV neutralizing IgM responses. Using a complement-dependent “physiological EV neutralization” assay , we detected a clear anti-EV VacV neutralizing antibody response at day 7 after VacV infection (Figure 5A), and the response was demonstrated to be predominantly IgM (Figure 5A).
Previously we have shown that VacV-specific CD4 T cells preferentially provide help to B cells of paired protein specificity . Since the anti-VacV IgM response is dependent on T cell help, we wanted to determine if priming the CD4 T cell response induced an increased anti-VacV IgM response. Mice were immunized subcutaneously with the class II restricted peptide B546-60, derived from the EV-specific protein B5. After 12 days, we observed a population of B5 specific CD4 T cells expressing CD40L, IFN-γ, and IL-2 (Figure 5B). Groups of peptide primed or mock primed mice were then infected i.p. with VacV and bled at day 7 post-infection. Sera was used to determine anti-B5 IgM and IgG antibodies by ELISA and to measure VacV neutralizing activity. The physiological EV neutralization assay was utilized . Mice with B546-60 specific CD4 T cells developed higher titers of anti-B5 IgG and IgM antibodies than the control group (Figure 5C, P < 0.0008 and P < 0.0002). Importantly, mice pre-immunized to develop B546-60 specific CD4 T cells showed a significant increase in VacV neutralizing antibody titers compared with the CFA only group (P = 0.0148, Figure 5D). This difference was maintained even after IgM inactivation (P = 0.0009, Figure 5D), demonstrating that both IgM and IgG neutralizing antibodies were boosted by preexisting VacV-specific CD4 T cells. The bulk of the EV neutralizing activity at day 8 was due to IgM antibodies, as IgM depletion resulted in >70% reduction in VacV neutralizing titers (B546-60 = 73%, CFA only = 74%). These results suggest that in the extrafollicular environment, as well as the germinal center, the virus-specific B cells preferentially interact with T cells of the exact same protein specificity.
Is an early anti-VacV neutralizing IgM response produced in humans after vaccination? To determine the time needed for the appearance of VacV neutralizing antibodies, and the role of IgM antibodies in such neutralization, a group of 14 healthy human subjects were immunized with Dryvax, the licensed US smallpox vaccine. Blood samples were obtained between days 12 to 18 post-vaccination. Based on anti-VacV IgG titers, we were able to divide the vaccinees in two groups. One group presented an anti-VacV IgG response (IgGhi), and consisted of five vaccinees, 4 that were bled at day 14 post-vaccination and one bled at day 18 post-vaccination (Figure 6A). The other group showed a poor or absent anti-VacV IgG response (IgGlo), and consisted of nine individuals bled primarily at earlier time points: days 12 (n = 1), 13 (n = 5), 14 (n = 2), or 15 (n = 1) post-vaccination (Figure 6A). The IgGhi group had 2/5 vaccinees that presented IgM titers above background, meanwhile, the IgGlo group had 7/9 vaccinees with an IgM response above background (Figure 6B). When the samples were tested for neutralizing antibodies, both groups (IgGhi and IgGlo) showed significant VacV-neutralizing titers. Importantly, the IgGhi group did not show a significant reduction in VacV neutralization titers when the plasma was IgM depleted (P >> 0.05. Figure 6C). In contrast, when the plasma of the IgGlo group was depleted of IgM, a statistically significant 77% decrease in neutralizing antibody titer was observed (P = 0.0095, Figure 6D), showing that IgM was a substantial contributor to the VacV neutralization antibody titer.
Most of the smallpox vaccine antibody research has been focused on IgG and particularly the long-lived IgG response, as these antibodies are responsible for the long-term protection against smallpox, VacV, and related orthopoxvirus infections . While antibodies after vaccination or passive transfer are very effective at protecting mice [14, 17, 18, 43, 44], primates [19, 45], and humans [46-50]  against subsequent poxvirus infections, the role of antibodies in control and clearance of a primary VacV immunization is less clear. VacV infection in mice is well under control by the time IgG is present at day 8. A human subject immunized with VacV will not develop circulating anti-VacV IgG until day 14 or later in our study (Fig. 6), with a strong IgG response present by day 30 [1, 20, 52, 53]. This is well past the point when the primary lesion has begun to be controlled, and in many vaccinees the lesion is completely resolved by this time point. Therefore, do neutralizing antibodies play a role in control of a primary vaccinia infection or immunization?
This study demonstrates the presence of early IgM antibodies after vaccination, capable of neutralizing VacV. Importantly, the IgM response was capable of neutralizing both infectious forms of the VacV, MV and EV. Anti-VacV neutralizing IgM was found in the sera of immunized mice and human subjects. This neutralizing IgM response is detectable as early as day four post-vaccination in mice (Figure 2), and suggests an important role for the first wave of anti-VacV IgM antibodies in the control of the infection. The IgM EV neutralization activity was complement dependent, consistent with our previous EV neutralization studies using IgG . Germinal center independence was consistent with a rapid T-cell dependent extrafollicular B cell differentiation to plasma cells (Figure 4). Extending our previous work , the linkage found in which CD4 T cells preferentially provide help to B cells of the same virion protein specificity is also seen in the extrafollicular IgM response (Figure 5).
IgM can be detected prior to IgG after monkeypox infection, as early as 5 days after rash formation (likely ~13 days after infection, based on known kinetics of vaccinia and smallpox infections in humans) . There are three reports in the literature that implicate neutralizing IgM in the control of human poxvirus infections. There is one report of a severe smallpox infection in a patient with selective IgM deficiency . Interestingly, this person had been previously vaccinated multiple times with VacV, and the only evidence of immundeficiency in the patient was the severe reduction in circulating IgM. Surprisingly, the patient was reported to have “adequate levels” of anti-VacV IgG. A second case involved generalized non-progressive vaccinia after vaccination of a child with IgM deficiency . A third case involved a 16 year old female with disseminated molluscum contagiosum . Again the only evident immunodeficiency in each case was IgM, all other serum immunoglobulin levels were normal. Human IgM deficiency is a rare and poorly understood condition without an identified genetic cause. As such, it is difficult to put IgM-deficient patients into a clear immunological context. Nevertheless, these reports of severe poxvirus infections in humans with IgM deficiency is consistent with our findings of very early neutralizing IgM to VacV in immunized mice and humans. We suggest that this neutralizing IgM response likely contributes to early control of viral spread.
This work was supported in part by NIH NIAID AI63107, NIH NIAID AI077953, NIH NIAID AI072543, a Pew Scholar Award, and a Cancer Research Institute Award to SC.
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.