Smallpox infection is lethal in 10–30% of humans and leaves disfiguring scars and occasionally blindness in survivors1,2. Edward Jenner's observation more than 200 years ago that milkmaids who had undergone cowpox infection were protected against smallpox led to smallpox eradication by large-scale vaccination1,2. The only currently available vaccine (Dryvax, manufactured by Wyeth) may cause serious or even fatal complications in people with atopic dermatitis and acquired cell-mediated immune defects resulting from HIV-1 infection, aggressive chemotherapy for cancer, and organ transplantation3,4. Indeed, widespread vaccination was halted in the US in the early 1970s because its side effects exceeded its benefits in the absence of a smallpox threat.

Recently, the perceived risk of a deliberate use of smallpox5 led the US to begin vaccinating a portion of its population, but the side effects have necessitated excluding large numbers of persons with eczema or immune deficiency from vaccination. Safer vaccines are clearly needed, but a major limitation in the development of new vaccines will be the inability to assess their efficacy in phase 3 clinical trials because endemic areas of smallpox no longer exist. Thus, strategies in the development of new smallpox vaccines will necessitate a clear understanding of the mode of protection afforded by Dryvax.

Here, we designed a study in a Rhesus macaque model of infection with monkeypox virus, which causes a disease similar to human smallpox, and is currently being used to test new candidate smallpox vaccines6,7,8,9. Because Dryvax vaccination protects macaques against monkeypox virus infection and disease after an intravenous challenge6,7, it is an excellent model in which to dissect correlates of protection. We therefore investigated the relative contributions of B cell–mediated and CD4+ and CD8+ T cell–mediated responses to protection by performing lymphocyte-depletion experiments in macaques either 1 month or 6 months (memory phase) after immunization with Dryvax.


B cell and CD8+ T-cell depletion in immunized macaques

At first, we vaccinated a cohort of 12 naive juvenile macaques, divided into three groups of four animals each, with Dryvax by scarification with a bifurcated needle10 and challenged them 4 weeks later with monkeypox virus (Fig. 1a). We depleted vaccinia-specific CD8+ T cells using a chimeric monoclonal antibody to CD8 (cM-T807)11 administered to animals in group 1, 1 d before, at the time of, and 1, 3 and 6 d after challenge exposure to monkeypox virus (Fig. 1a). To minimize the development of humoral responses after vaccination, group 2 received the α-CD20 B cell–depleting antibody Rituxan12,13 8 d and 1 d before and 7 and 15 d after vaccination (Fig. 1a). Animals 237M and 241M in the control group 3 received the irrelevant α-respiratory syncytial virus (RSV) monoclonal antibody (known as Synagis) at times and doses identical to those received by animals in groups 1 and 2. We treated the two remaining macaques in group 3 (macaques 245M and 246M) with saline.

Figure 1: Depletion of B cells and CD8+ T lymphocytes in rhesus macaques.
figure 1

(a) Study design. Animals 237M and 241M received control antibody to RSV (Synagis). We gave antibody to CD20 (Rituxan) to animals in group 2 at the indicated times. We administered antibody to CD8 (cM-T807) to macaques in group 1 as indicated. At day 26 after Dryvax immunization, we performed monkeypox virus challenge. Absolute counts of CD3+ CD4+ T cells (b), CD3+ CD8+ T cells (c), and CD19+ B cells (d) in blood.

We monitored total white blood cells and absolute counts of CD3+ CD4+ T cells, CD3+ CD8+ T cells and CD19+ B cells before monkeypox virus challenge. Treatment with antibodies to CD8 or CD20 did not alter the absolute number of CD3+ CD4+ T cells, which fluctuated within a normal range in all animals (Fig. 1b). In contrast, the number of CD3+ CD8+ T cells in the blood markedly decreased after treatment with the CD8–depleting antibody in all macaques in group 1 (Fig. 1c), and α-CD20 treatment substantially decreased CD19+ B cells in all animals in group 2 (Fig. 1d). In one animal (238M), however, the number of CD19+ B cells began to rebound by day 18 (Fig. 1d).

B-cell depletion impairs humoral responses to vaccinia

We measured vaccinia-neutralizing antibodies14 in all macaques before and 3 weeks after (days 21–22) Dryvax vaccination. All animals from groups 1 and 3 developed neutralizing antibody titers to vaccinia that ranged between 120 and 3,500, whereas animals in group 2 did not develop equivalent levels of neutralizing antibodies to vaccinia (Fig. 2a), showing, as we intended, that B-cell depletion affected the development of this humoral response to Dryvax.

Figure 2: Humoral immune response in the treated macaques.
figure 2

We collected sera from animals in groups 1–4 before (day 0) and 3 weeks after vaccination with Dryvax and tested them for vaccinia-specific neutralizing antibodies (a). Immunoprecipitation assay using radiolabeled lysates from COS cells transfected with plasmids expressing the vaccinia virus A27L, A33R, B5R or empty vector using sera from animals in group 1 (b), group 2 (c) and group 3 (d). Molecular mass markers (kDa) are shown on the left and the position of immunoprecipitated vaccinia proteins are shown on the right. 0, sera obtained at time of immunization; 3, sera obtained at 3 weeks.

To characterize further the antibody response to Dryvax that developed in T and/or B cell–depleted macaques, we measured the production of antibodies specific to surface proteins of two infectious forms of poxviruses: A27L, associated with intracellular mature virions (IMV), and A33R and B5R, specific to extracellular enveloped virions (EEV). DNA immunization with plasmids encoding these proteins has recently been shown to protect macaques against monkeypox7. We performed immunoprecipitation assays on sera collected on the day of vaccination (baseline) and 3 weeks later using extracts from metabolically labeled cells transfected with plasmids expressing A27L, A33R or B5R. Whereas sera from all animals in groups 1 and 3 precipitated the three vaccinia proteins, sera from the four macaques treated with antibody to CD20 did not recognize any of the vaccine proteins tested (Fig. 2b–d).

As we noticed in macaque 238M that B cells had rebounded at day 18 (Fig. 1d), we further investigated this animal as well as the other macaques to discern the degree of humoral response to all vaccinia proteins by performing enzyme-linked immunosorbent assay (ELISA) on cell extracts of vaccinia-infected cells. Sera from animals in groups 1 and 3 had titers of binding antibodies that ranged from 800 to ≥12,800 and, notably, among the sera from animals of group 2,238M serum had a titer of 1,600 whereas all the remaining sera were negative, indicating that the more transient depletion of B cells in this macaque may have accounted for its ability to mount a humoral response to vaccinia.

B- but not CD8+ T-cell depletion abrogates vaccine protection

To assess the relative contribution of CD8+ T cells and B cells to protection, we challenged all animals with 5 × 107 plaque-forming units (p.f.u.) of monkeypox virus (Zaire 79 strain) intravenously at day 26 after Dryvax vaccination (Fig. 1a). We assessed the virological and clinical sequelae of monkeypox virus challenge by measuring the monkeypox virus genome DNA copies in whole blood with a quantitative real-time PCR DNA assay15 and by monitoring the number of skin lesions and the health status9. At 3 min after inoculation with monkeypox virus, we detected more than 105 copies/ml of viral DNA genomes in the blood of all animals, showing exposure after intravenous challenge with monkeypox virus (Table 1). Thereafter, we found no monkeypox viral DNA in the blood of all control animals in group 3, consistent with the solid protection induced by Dryvax in macaques6,7. Notably, all animals that had sustained depletion of CD8+ T cells (group 1) had no viral genomes in blood and did not develop skin lesions, suggesting a dispensable role for CD8+ T cells. In contrast, of the four animals that received antibody to CD20 (group 2), three had monkeypox virus genomes in blood, developed skin lesions, became severely ill and died by day 13 from challenge (Table 1). A notable exception was animal 238M that was protected from lethal challenge. This macaque experienced B-cell rebound by day 18 and mounted an ELISA antibody response to vaccinia.

Table 1 Lymphocyte depletion and monkeypox virus challenge

Thus, both the virological findings and the clinical outcome showed that depletion of CD8+ T cells before and at the time of monkeypox virus challenge did not interfere with protection, whereas depletion of B cells beginning at the time of and continuing throughout vaccination resulted in susceptibility to lethal infection. The crucial role of B cells is further supported by the fact that animal 238M that showed early rebound of B cells was also protected against monkeypox virus challenge.

Antibodies but not CD4+ T cells mediate protection

The results presented above suggested that Dryvax-elicited antibodies to vaccinia are necessary for protection. But because depletion of B cells may also considerably affect the development of virus-specific CD4+ T-cell responses that, in turn, may participate in protection, we investigated (i) whether depletion of CD4+ T cells would abrogate protection and (ii) whether antibodies to vaccinia alone were sufficient to protect from a lethal monkeypox virus challenge. To this end, we designed a study that included 11 poxvirus-naive juvenile macaques (Fig. 3a). We vaccinated one group of three macaques (group 4) with Dryvax and treated this group with a humanized monoclonal antibody to CD4 (OKT4A-hulgG1)16 at day 25 with a dose of 10 mg/kg and at days 29, 34 and 44 after immunization (Fig. 3a). The α-CD4 antibody treatment effectively depleted CD4+ T cells in all animals whose levels remained low at the time of and after monkeypox virus challenge (Fig. 3b). Group 5 received human immunoglobulin prepared from plasma of recently vaccinated human volunteers (vaccinia immune globulin intravenous (human); VIGIV). This immunoglobulin preparation had a vaccinia-neutralizing titer of 1:4,166. We administered VIGIV to the four naive macaques in group 5 at a dose of either 1 g/kg (macaques 321M, 334M) or 0.1 g/kg (macaques 329M, 337M) 4 d before monkeypox virus challenge to allow for adequate tissue distribution of the antibodies. We treated two additional macaques (group 6) with a commercial preparation of human immunoglobulin (immune globulin intravenous (human); IGIV) that had not been obtained from plasma of recent vaccinees but nevertheless had a vaccinia-neutralizing titer of 1:637. Macaque 327M received 1 g/kg and macaque 333M received 0.1 g/kg of IGIV. We inoculated the two control animals in group 7 (macaques 322M, 325M) with saline at the same time point. We found vaccinia-neutralizing antibody titers in macaques depleted of CD4+ T cells, as expected, and in macaques treated with high and low doses of VIGIV. Both animals treated with IGIV had low levels of neutralizing antibodies, whereas none were detected in control animals from group 7 (Table 2). We challenged all macaques with the same stock of monkeypox virus used in the experiments outlined in Figure 1a at day 29 (Fig. 3a). We documented viral exposure by measuring viral DNA genomes in blood 3 min after exposure (Table 2). At day 3 after exposure and thereafter, all macaques depleted of CD4+ T cells were negative for viral genomes in blood and did not develop skin pocks. Macaques that had received VIGIV developed skin lesions that resolved and all survived lethal challenge. We measured viral genomes in blood after day 3 in all four animals. The number of pox lesions was higher in animals that received the lower dose of VIGIV (Table 2). In contrast, both macaques treated with IGIV and both macaques treated with saline became severely ill and developed numerous pocks; the macaque treated with 1 g/kg of IGIV survived infection, whereas the others became moribund and were killed.

Figure 3: Vaccinia antibodies and CD4+ T cells in protection from monkeypox virus.
figure 3

(a) Study design. (b) Level of CD3+ CD4+ T cells in blood of animals in group 4 after treatment with the α-CD4 antibody OKT4A-IgG1. Role of CD4+ CD8+ memory responses in protection from monkeypox virus. (c) Study design. (d) Kinetics of neutralizing antibodies to vaccinia within 6 months after immunization with Dryvax. Blood level of CD3+ CD4+ (e) and CD3+ CD8+ (f) T cells after treatment with the antibody to CD4 (OKT4A-IgG1) and the antibody to CD8 (cM-T807), respectively.

Table 2 Outcome of monkeypox virus challenge following CD4+ T cell depletion or prophylactic administration of VIGIV

CD4+ and CD8+ memory T cells in protection

Collectively, the data presented above indicated that the major mode of protection from monkeypox virus is mediated by antibodies. As we modeled depletion of CD4+ and CD8+ T cells at the time when neutralizing antibodies induced by vaccination are high (4 weeks after vaccination), the contribution of CD4+ and CD8+ T cells were possibly underestimated. To assess whether CD4+ and CD8+ T cells could be important during the memory phase of the immune response to Dryvax, we immunized with Dryvax two groups of four macaques each and, 6 months later, depleted them of either CD4+ T cells (group 8) or CD8+ T cells (group 9) before challenging them with monkeypox virus (Fig. 3c). The kinetics of neutralizing antibody responses measured after immunization were comparable in animals in groups 8 and 9, and by 6 months, neutralizing antibody titers ranged between 1:52 and 1:242 (Fig. 3d). We obtained effective depletion of CD4+ and CD8+ T cells in all macaques from groups 8 and 9, respectively (Fig. 3e,f). Challenge exposure to monkeypox virus of all these macaques showed that, whereas the control animal 485M became infected, developed innumerable skin pocks and died from disease at day 9, all animals from groups 8 and 9 survived (Table 3). Notably, the number of skin lesions was lower in animals in groups 8 and 9 than animals treated with passive transfer of VIGIV (group 5) in which CD4+ and CD8+ memory T cells were absent. In addition, several of the lesions in animals in groups 8 and 9 did not evolve from papular to vesicular pocks. These data, show that CD4+ and CD8+ T-cell responses are dispensable for, and confirm the importance of the memory B-cell response in protection from monkeypox virus. The level of neutralizing antibody titers present in protected macaques mirrors the level of protective titers measured in humans vaccinated with Dryvax17,18.

Table 3 Treatment


Monkeypox virus infection of healthy macaques is a suitable model for studying protective immune responses induced by vaccination as DNA genomes can be quantified in blood as well as through dissemination to tissues and skin, similar to smallpox infection of humans. The data presented here show that antibodies are necessary and sufficient to protect Rhesus macaques from a lethal monkeypox virus infection. In contrast, depletion of CD4+ and CD8+ T cells at 1 month and 6 months from vaccination did not abrogate protection from severe disease. This represents the first definitive demonstration in a nonhuman primate model that anti-vaccinia antibodies alone can protect against death from monkeypox virus infection. The importance of antibodies is further supported by our recent finding that the failure of Dryvax immunization to afford protection from monkeypox virus in immunodeficient macaques is related to the lack of maturation of B-cell response and IgM-to-IgG isotype switching in the condition of defective CD4+ T-cell help19. Our findings differ from one study in a mouse model of vaccinia-induced disease20 but are consistent with a previous study in mice21, which shows that B cells contribute to protection against vaccinia-induced weight loss and death.

The demonstration that monkeypox virus infection is mainly controlled by antibodies allows comparisons with other viral diseases such as influenza and poliomyelitis. The latter is blocked by antibodies that can be induced by vaccination with killed virus vaccines22, suggesting that sufficient amounts of antibodies at the portal of entry (influenza) or in the serum (poliomyelitis) are sufficient to prevent disease.

The identity of the vaccinia immunogens necessary to elicit protective antibodies and the mechanism of antibody-mediated protection remain to be determined. Recent findings that a single IMV immunogen (L1R) could confer some degree of protection against a lethal monkeypox virus challenge and that a combination of two IMV and two EEV immunogens (L1R, A27L, A33R and B5R) could protect against severe monkeypox disease7 suggest that it might be possible to achieve complete protection with antibodies raised against a relatively small subset of the more than 200 proteins in the poxvirus proteome. In addition to the use of antigens that are targets of IMV neutralizing antibodies (e.g., A27L, L1R), EEV surface proteins may also be desirable components of the next generation of safe and effective vaccines for smallpox and monkeypox23,24,25,26,27.

An effective smallpox vaccine with a better safety profile would serve the public health not only in the event of bioterrorism but also in case of a monkeypox outbreak28,29,30,31. The undesired side effects of Dryvax prompted the development of plaque-purified vaccinia strains8 or highly attenuated poxviruses such as NYVAC32 and modified vaccinia virus Ankara33 as alternative vaccines. Because our studies show that neither CD4+ nor CD8+ T cells are essential for disease protection, vaccine modalities able to induce high-affinity antibodies rather than major histocompatibility class I– or class II–restricted immune responses may be efficacious. Thus, vaccine modalities (including DNA, DNA plus proteins or proteins alone) able to induce high-titer, long-lasting, high-affinity antibody responses may be protective and have a better safety profile than the live-virus smallpox vaccine currently available.


Animals, antibody treatment and monkeypox virus challenge.

We obtained all 32 colony-bred Rhesus macaques (Macaca mulatta) used in our studies from Covance Research Products and housed and handled them in accordance with the standards of the American Association for the Accreditation of Laboratory Animal Care, per US National Institutes of Health Animal Care and Use Committee guidelines.

We immunized animals with Dryvax by scarification, as previously described10. Every 2 d we measured and photographed Dryvax lesions. We intravenously administered to group 2 animals the monoclonal mouse-human chimeric antibody to human CD20 (Rituxan, Genentech and IDEC Pharmaceuticals) at days −8, −1, 7 and 15 at a dose of 20 mg/kg (Fig. 1). We intravenously administered the mouse-human chimeric monoclonal antibody to CD8 (cM-T807)11 at days 25, 26, 27, 29 and 32 at a dose of 5 mg/kg (Fig. 1a). We intravenously administered to the animals described in Figure 3a the humanized monoclonal antibody to CD4 (OKT4A-hulgG1)16 at an initial dose of 10 mg/kg at day 24 (5 d before monkeypox virus challenge) and at 5 mg/kg on the day of challenge (day 29) and thereafter (days 34, 39). We gave to all animals in group 8 a single dose of 50 mg/kg 7 d before monkeypox virus challenge (Fig. 3c). We intravenously administered the control α-RSV monoclonal antibody (Synagis, MedImmune) at weeks −8, −1, 0, 7 and 15 to animal 237 at a dose of 20 mg/kg and at weeks 25, 26, 27, 29 and 32 to animal 241 at a dose of 5 mg/kg. Group 5 animals received intravenous administration of VIGIV (Cangene). Group 6 animals received intravenous administration of IGIV (Gamimune, Bayer Biological Products). We intravenously administered monkeypox virus (Zaire 79 strain) at a dose of 5 × 107 p.f.u. to all animals studied here.

CD4+, CD8+ and CD19+ cell counts.

We determined T- and B-cell counts at specified intervals and analyzed them by flow cytometry. We stained 100 μl of whole blood from each macaque for 30 min at 21 °C with a 40 μl cocktail containing CD3-FITC (BD Biosciences), CD4-peridinin chlorophyll protein (BD Biosciences), CD8-phycoerythrin (DakoCytomation), and CD19-allophycocyanin (Beckman Coulter). We used antibody to CD19 in place of CD20 because of the receptor-blocking effect of the α-CD20 on B cells. In conjunction, we performed isotype controls using a 15 μl mixture containing each of the following antibodies: mouse IgG3-FITC, mouse IgG1-phycoerythrin, mouse IgG1-peridinin chlorophyll protein, and mouse IgG1-allophycocyanin. We lysed red blood cells using 2 ml of 1× FACS Lysis solution (BD Biosciences) and incubated them for 15 min at 21 °C in the dark. Then we centrifuged the samples at 2,500 r.p.m. for 5 min, washed them with 1× FACS Wash Buffer (1% FBS, 0.05% NaAzide in 1× PBS), resuspended them in 200 μl wash buffer, and acquired them by a FACSCalibur flow cytometer (BD Biosciences) within 24 h. We expressed the counts as absolute CD3+ CD4+, CD3+ CD8+ and CD19+ cell numbers.

Real-time PCR to detect monkeypox virus genomes.

We extracted DNA from frozen blood samples using the QIAGEN QIAamp DNA mini kit15. We selected the primers OPHA-F89 (5′-GATGATGCAACTCTATCATGTA-3′) and OPHA-R219 (5′-GTATAATTATCAAAATACAAGACGTC-3′) and the probe OPHA-p143S-MGB (5′-FAMAGTGCTTGGTATAAGGAG MGBNFQ-3′) from the hemagglutinin gene. Whereas Invitrogen synthesized the primers, PE Biosystems synthesized the TaqMan probe, and it contained 6-carboxyfluorescein (FAM) in the 5′ end and the nonfluorescent quencher and the minor groove binder in the 3′ end.

We performed the 5′ nuclease PCR reaction and amplification conditions using Platinum Taq DNA polymerase (Invitrogen)15. All reactions included at least one positive control that had 25 copies of cloned target DNA and one no-template control. The positive control for each run established the cycle threshold (Ct) value for positivity. We retested samples yielding Ct values that marginally exceeded the threshold value. If the Ct value was confirmed as exceeding the threshold after retesting, we considered the sample negative (contained less than 25 gene copies).

Immunoprecipitation of vaccinia proteins.

We transfected plasmid DNA (pWRG/A33R, pWRG/B5R, pWRG/A27L, empty vector pWRG7077) into COS cell (American Type Culture Collection) monolayers (60–80% confluent) using Fugene 6 reagent, as described by the manufacturer. After 24 h, we radiolabeled monolayers with Promix (200 mCi per T-25 flask, 35S-methionine and [35S]-cysteine; Amersham) for approximately 4 h and immunoprecipitated them, as described previously34. We lysed transfected cells on ice for 5 min with a modified RIPA buffer: 4% Zwittergent (Calbiochem), 0.5 M NaCl, 1 mM EDTA, 10 mM Tris, pH 8, and protease inhibitors (Complete; Boehringer Mannheim). We combined lysates with the indicated antibody (previously incubated 1 h with unlabeled COS cell lysate) and incubated them overnight at 4 °C. We combined lysate-antibody mixtures with protein A Sepharose (CL-4B; Sigma), incubated them at 4 °C for 30 min, and then washed them three times with lysis buffer and once with 10 mM Tris, pH 8.0. We added sample buffer (125 mM Tris (pH 8.0), 1% sodium dodecyl sulfate, 10% glycerol, 0.01% bromophenol blue containing 2% 2-mercaptoethanol) and boiled the samples for 2 min. We ran samples on 4–12% bis-Tris sodium dodecyl sulphate polyacrylamide gel electrophoresis gradient gels with 2-N (morpholino) ethane sulfonic acid running buffer (NuPAGE), at a 200 V constant voltage. We fixed gels in an equal mixture of methanol and acetic acid for 15 min, soaked them in Amplify (Amersham) for 15 min, dried them and placed them on phosphor screen overnight. We obtained digital images using a Cyclone (Packard) phosphoimager.

Vaccinia-infected cell lysate ELISA.

We performed ELISAs essentially as described previously34. Briefly, we coated 96-well ELISA plates with vaccinia virus (strain Connaught)-infected Vero cell lysate antigen or mock-infected Vero cell lysate antigen and dried them overnight. We washed plates once with wash buffer (PBS plus 0.05% Tween-20). In 100 μl amounts per well, we added heat-inactivated (56 °C, 30 min) serum samples diluted in wash buffer containing 5% milk to antigen- or mock antigen–containing wells and incubated them for 1 h at 37 °C. We washed plates four times and added 100 μl per well of peroxidase-labeled goat antibody to monkey IgG (KPL). After a 30 min incubation at 37 °C, we washed the plates as before and incubated them for 10–30 min at 21 °C with 100 μl of 2,2′-azino bis (3-ethylbenzthiazoline-6-sulfonic acid) substrate. We stopped the reactions by adding 100 μl per well of 0.2 N phosphoric acid, and we determined the optical density (OD) at 405 nm by an ELISA plate reader. For each serum dilution series, we subtracted OD values from mock antigen wells from the experimental values to give specific OD values. Each specific OD value represented the average value of two independent dilution series. We determined end-point titers as the highest dilution with an absorbance value greater than the mean absorbance value from serum from animals vaccinated with a negative control plasmid plus three standard deviations.

Measurement of vaccinia-neutralizing antibodies.

We collected plasma samples from monkeys in groups 1–5 immediately before (day 0) and at 3 weeks after Dryvax immunization (day 21 or 22). We heat-inactivated (56 °C for 30 min) all plasma samples and evaluated them for the presence of vaccinia-neutralizing antibodies using a new assay based on expression of a reporter gene expressing β-galactosidase (β-Gal)14.

Briefly, we used a recombinant vaccinia virus vSC56, expressing β-Gal under the control of a synthetic early-late promoter35, to develop a neutralization assay based on a single-round infection of HeLa cells (CCL-2, American Type Culture Collection). This is a rapid (24 h), high-throughput assay that was shown to have similar sensitivity to the classical plaque reduction neutralization tests14. Each assay includes as a positive control US Food and Drug Administration interim standard VIGIV obtained from Dynport and vialed at the Center for Biologics Evaluation and Research. Negative controls included plasma from unvaccinated children and albumin. We preincubated four serial dilutions of each monkey plasma with vSC56 virus for 60 min at 37 °C and then dispensed them into 96-well round bottom plates containing 2 × 105 HeLa cells per well (four replicates per antibody dilution). We incubated plates for an additional 16 h at 37 °C in a humidified CO2 incubator. Then we lysed cells with the detergent IGEPAL CA630 (Sigma). In the second stage of the assay, we measured β-Gal enzymatic activity in each well using 96-well Immunlon 2 plates (Thermo Labsystems). Each plate included a β-Gal standard curve using a recombinant β-Gal enzyme (Roche Diagnostics Corporation). We added chlorophenol red beta-D-galactopyranoside monosodium salt (CPRG) substrate (Roche Diagnostics) to all wells for 30 min at 21 °C in the dark, and we stopped the enzymatic reaction with 1 M Na2CO3 solution. We determined OD at 575 nm by an ELISA reader. We transferred OD readings to Microsoft Excel for further analysis. We used the β-Gal standard curves to convert OD values into β-Gal activity per experimental or control group (in mU/ml). We expressed the β-Gal activity of each experimental group (virus mixed with a given dilution of test plasma) as percentage β-Gal activity in the virus-only control wells. We used Microsoft Excel to plot the percentage of control values for the serial dilutions of each plasma versus log dilutions. To calculate the 50% inhibitory dilution, we used the equation of each curve.

GenBank accession number.

The GenBank accession number for the Variola major virus (strain Bangladesh-1975) is L22579; open reading frame J7R contains the hemagglutinin gene.