Development of a Highly Efficacious Vaccinia-Based Dual Vaccine

Home , Smallpox vaccine

Development of a highly efﬁcacious vaccinia-based dual vaccine against smallpox and anthrax, two important bioterror entities

Tod J. Merkela,1, Pin-Yu Pererab,1, Vanessa K. Kellya, Anita Vermaa, Zara N. Llewellync, Thomas A. Waldmannd,2, Joseph D. Moscae, and Liyanage P. Pererad,2

aCenter for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, MD 20892; bVeterans Affairs Medical Center, Washington, DC 20422; cSouthern Research Institute, Birmingham, AL 35255; dMetabolism Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; and eJDM Technologies, Ellicott City, MD 21042

Contributed by Thomas A. Waldmann, September 2, 2010 (sent for review May 26, 2010)

Bioterrorism poses a daunting challenge to global security and smallpox vaccinations in the civilian population ceased over 30 y public health in the 21st century. Variola major virus, the etiological ago, and we are now faced with an ever-increasing naive pop- agent of smallpox, and Bacillus anthracis, the bacterial pathogen ulation, along with declining immunity in the older generation responsible for anthrax, remain at the apex of potential pathogens who were vaccinated in their childhood. Although the currently that could be used in a bioterror attack to inflict mass casualties. available licensed vaccine [ACAM 2000 (Acambis, Inc.), a cell- Although licensed vaccines are available for both smallpox and an- culture-grown derivative (7) of the earlier Dryvax vaccine used in thrax, because of inadequacies associated with each of these vac- the smallpox eradication campaigns] is highly effective, it can cines, serious concerns remain as to the deployability of these vac- cause serious complications in some vaccinated individuals, in- cines, especially in the aftermath of a bioterror attack involving cluding encephalitis, progressive vaccinia, myocarditis, or myo- pericarditis. Additionally, its use is contraindicated in patients these pathogens. We have developed a single vaccine (Wyeth/IL- with atopic skin diseases and immune deficiencies and during 15/PA) using the licensed Wyeth smallpox vaccine strain that is ef- – ≈ fi pregnancy. Unlike 3 4 decades ago, 25% of the individuals in cacious against both smallpox and anthrax due to the integration contemporary populations are estimated to represent “at-risk of immune-enhancing cytokine IL-15 and the protective antigen (PA) ” B. anthracis groups for whom live ACAM 2000 vaccine is contraindicated (7). of into the Wyeth vaccinia virus. Integration of IL-15 Thus, a safer vaccine that can match the efficacy of Dryvax/ fi renders Wyeth vaccinia avirulent in immunode cient mice and en- ACAM 2000 vaccine but that is devoid of its residual virulence is hances anti-vaccinia immune responses. Wyeth/IL-15/PA conferred urgently needed for contemporary populations. sterile protection against a lethal challenge of B. anthracis Ames Anthrax caused by the bacterial pathogen B. anthracis clinically strain spores in rabbits. A single dose of Wyeth/IL-15/PA protected manifests itself in cutaneous, gastrointestinal, and inhalational 33% of the vaccinated A/J mice against a lethal spore challenge 72 h forms, all of which can be fatal, although the inhalational form later whereas a single dose of licensed anthrax vaccine protected causes the highest mortality (8). A number of antibiotics such as only 10%. Our dual vaccine Wyeth/IL-15/PA remedies the inadequa- ciprofloxacin and doxycycline are very effective in eliminating cies associated with the licensed vaccines, and the inherent ability vegetative bacilli, but the ability of the antibiotic-resistant spores to of Wyeth vaccinia virus to be lyophilized without loss of potency remain dormant in the tissues (9) and subsequently reactivate to makes it cold-chain independent, thus simplifying the logistics of cause fatal disease increases the difficulty of treating this disease storage, stockpiling, and field delivery in the event of a bioterror with antibiotics alone. In fact, even with antibiotic coverage, attack involving smallpox or anthrax. mortality could be as high as 60–70% especially with the inhalation form. The pathogenesis of anthrax is driven by plasmid-borne γ biodefense | IL-15 | bioterrorism virulence factors that include the poly- -D-glutamic acid capsule and the tripartite exotoxin that consists of protective antigen (PA), lethal factor (LF), and edema factor (EF). PA plays a central role he 2001 terror attacks in the United States have heightened in virulence by mediating the entry of LF and EF into cells. Col- Tthe potential threat of bioterrorism in the United States and lectively, these virulence factors promote the overwhelming sys- around the globe. Because of the purported past use of Variola temic pathophysiology associated with B. anthracis infection, often major and Bacillus anthracis in biowarfare and the reported pos- leading to a fatal outcome. The available evidence supports the sible existence of military programs for weaponization of these notion that induction of antibodies solely against PA can be pro- agents by some nations and rogue regimes, smallpox and anthrax tective against anthrax (10). PA is also the major component of represent the most deadly bioterror entities (1–3) that are a threat

the anthrax vaccine licensed for human use in the United States as MEDICAL SCIENCES to our national security and public health. Therefore, the de- an aluminum hydroxide-adsorbed cell-free filtrate of an acapsular velopment of medical countermeasures against their potential strain (V770-NP1-R) of B. anthracis [Biothrax or anthrax vaccine use remains a national priority (4). adsorbed (AVA)]. In addition to concerns regarding adverse Smallpox is a highly communicable, often fatal disease with effects associated with this vaccine, its undefined composition, lot- mortality rates approaching 30% in susceptible individuals (5). to-lot variation, and the requirement for multiple doses over a Currently there is no approved treatment for smallpox although protracted period to achieve adequate levels of protective immu- vaccination within the first few days of exposure may provide some nity make this vaccine less than optimal for use in response to level of protection against the disease (6) (http://www.bt.cdc.gov/ agent/smallpox/vaccination/facts.asp). Variola virus, the etiologi-

cal agent of smallpox, is a member of the Orthopoxvirus subfamily Author contributions: T.J.M., P.-Y.P., T.A.W., J.D.M., and L.P.P. designed research; T.J.M., of viruses, which also includes the vaccinia virus that induces P.-Y.P., V.K.K., A.V., Z.N.L., and L.P.P. performed research; Z.N.L. supervised rabbit immu- potent cross-protective immunity against Variola virus. The medi- nization and challenge studies under contract; T.J.M., P.-Y.P., A.V., T.A.W., J.D.M., and cal triumph of eradicating smallpox as a natural disease in 1977 L.P.P. analyzed data; and L.P.P. wrote the paper. was achieved with the use of a calf-lymph–derived live vaccinia The authors declare no conﬂict of interest. virus vaccine (Dryvax containing the Wyeth strain in the United 1T.J.M. and P-Y.P. contributed equally to this work. ﬁ 2 States), thus attesting to the ef cacy and immunogenicity of To whom correspondence may be addressed. E-mail: [email protected] or tawald@ vaccinia viruses against Variola virus. In the United States, helix.nih.gov.

www.pnas.org/cgi/doi/10.1073/pnas.1013083107 PNAS | October 19, 2010 | vol. 107 | no. 42 | 18091–18096 Downloaded by guest on October 2, 2021 a bioterrorism incident. Furthermore, the limited shelf life of ≈4y of Biothrax/AVA results in the need for periodic replenishment of vaccine in the strategic national stockpile. Therefore, a compelling need exists for a better vaccine against B. anthracis that can confer rapid immunity with an abbreviated immunization schedule that can be stored long term and deployed quickly in the event of a bioterror event. We recently reported that the integration of IL-15, a cytokine with pleiotropic immune modulatory activities, into the Wyeth strain of vaccina virus derived from the licensed smallpox vaccine resulted in the development of a smallpox vaccine candidate (Wyeth/IL-15) with superior efficacy and immunogenicity. This smallpox vaccine candidate out-performed the parental vaccine in Fig. 1. The expression of glycosylated B. anthracis PA by recombinant two animal models that sufficiently recapitulate the pathophysi- vaccinia viruses with an integrated IL-15 gene. BHK-21 cells were infected ology of smallpox in humans; mice were protected when lethally with Wyeth/IL-15/PA and MVA/IL-15/PA for 48 h, and the infected BHK-21 challenged with a neurotropic strain of vaccinia virus intranasally cell lysates were subjected to SDS/PAGE on a 10% acrylamide gel. Western and cynomolgus monkeys were protected from a lethal i.v. chal- blot analyses were performed using a polyclonal goat PA-specific antibody lenge of monkeypox virus (11, 12). Moreover, unlike the parental (List Biological Laboratories). The glycosylation of vaccinia virus-expressed virus, our Wyeth/IL-15 is nonlethal in T-cell–deficient athymic PA polypeptide in BHK21 cells was assessed by the treatment of vaccinia- nude mice at a dose of 107 plaque-forming units (pfu) and un- infected cell lysates with PNGase F enzyme (New England Biolabs) before dergoes enhanced clearance in these mice, attesting to its safety SDS/PAGE. The positions of molecular weight markers are on shown on left. even in immunocompromised hosts (12). Having demonstrated the superiority in immunogenicity and attenuation of virulence of fi – Immunogenicity and Ef cacy of Wyeth/IL-15/PA in Rabbits. To assess IL-15 integrated Wyeth vaccinia, we exploited our Wyeth/IL-15 the immunogenicity and efficacy of Wyeth/IL-15/PA in compar- platform to generate a dual vaccine effective against both smallpox ison with the licensed Biothrax/AVA, groups of female New and anthrax by integrating the PA gene of B. anthracis (Wyeth/IL- Zealand White rabbits (each group consisting of nine animals) 15/PA) to overcome the problems of poor immunogenicity and were vaccinated with the respective vaccines. As a negative con- apparent lack of immunological memory associated with the li- trol, a group of six animals were vaccinated with Wyeth/IL-15 (see censed Biothrax/AVA vaccine. We demonstrate that Wyeth/IL- Table 1). When we measured the anti-PA ELISA antibody titers 15/PA confers superior protection against inhalation anthrax in in the serum samples from blood collected 28 d post primary both mice and rabbits and make a direct comparison with the li- vaccination, immediately before administering the booster dose censed Biothrax/AVA vaccine. We believe our dual vaccine, of the vaccine the mean PA-binding antibody titer for the Wyeth/ Wyeth/IL-15/PA, which is effective against two of the most deadly IL-15/PA group was 46,220 whereas for the group of rabbits that pathogens, will help consolidate and simplify our national bio- received the Biothrax/AVA the mean titer was 16,220 and the terror counterefforts by streamlining the manufacture, stockpiling, difference was statistically significant (P = 0.038). The same trend and swift deployment of such vaccines should the need arise. was noted when we examined the serum samples collected 21 d after the booster vaccination; immediately before animals were Results subjected to an inhalation challenge with a mean dose of 262 Validation of Dual Vaccine Agent for the Expression of Molecular LD50 of B. anthracis Ames strain spores, we noted a mean anti- Adjuvant IL-15 and PA of B. anthracis. The integration of human IL- PA ELISA titer of 515,600 for the group vaccinated with Wyeth/ 15 and B. anthracis PA genes in a head-to-tail configuration into IL-15/PA in comparison with a mean titer of 275,600 for the the hemagglutinin locus of the licensed smallpox vaccine strain Biothrax/AVA group (P = 0.012). These results indicate that (Wyeth) of vaccinia resulted in a stable recombinant virus (Wyeth/ Wyeth/IL-15/PA displays superior immunogenicity in rabbits in IL-15/PA) that displayed no phenotypic or growth characteristic a direct comparison with the licensed Biothrax/AVA when ad- differences in comparison with the parental virus. The presence of ministered in a single-dose vaccination regimen or a two-dose human IL-15 in the culture supernatants of Wyeth/IL-15/PA– vaccination regimen. However, surprisingly, when we determined infected cells was confirmed by a human IL-15–specific ELISA, the toxin-neutralizing antibody titer (TNA) in the serum samples and a CV-1 cell monolayer infected at a multiplicity of infection collected 21 d after the booster vaccination for the two groups of 10 yielded 2–4 ng/mL of IL-15 in the culture supernatant 3 d (Table 1), we noted a discordance in which the mean TNA titer postinfection, which is consistent with the levels that we previously for the Wyeth/IL-15/PA group was 4.5-fold lower than that of the reported for Wyeth/IL-15 (11). Because there is considerable in- Biothrax/AVA group (478 versus 2153), which was statistically terest in nonreplicative vaccinia viruses such as the modified vac- significant (P = 0.009). This observation is important because, for cinia virus Ankara (MVA) as an alternate smallpox vaccine that almost all reported PA-based vaccine candidates, the vaccine- can be safely administered to individuals for whom the use of the induced TNA titers paralleled those of PA-binding ELISA titers licensed Wyeth strain vaccinia is contraindicated, and because quantitatively (16–18). Our results may reflect the possibility that we have shown earlier that IL-15–integrated MVA (MVA/IL-15) vaccinia-expressed, glycosylated PA induces a qualitatively dif- is an effective vaccine in two animal models that recapitulate ferent humoral immune response in vaccinated rabbits. Mirroring smallpox in humans (11, 12), we generated a PA-expressing our observation, in a recent study Livingston et al. (19) reported recombinant virus in the MVA/IL-15 backbone as well (MVA/IL- that, whereas electroporation-mediated delivery of a DNA-based 15/PA). As shown in Fig. 1, an abundant expression of PA was PA vaccine in rhesus monkeys elicited a higher PA-binding confirmed by immunoblot analyses of Wyeth/IL-15/PA or MVA/ ELISA titer than Biothrax/AVA, the TNA titers induced by the IL-15/PA–infected BHK-21 cell lysates. It should be noted that the DNA vaccine were fourfold less than the titers induced by Bio- apparent molecular weight of the vaccinia-expressed PA in BHK- thrax/AVA vaccine. It is important to note that, in the present 21 cells is higher than the predicted 83 kDa. The primary amino study, the protection of vaccinated rabbits from a lethal intranasal acid sequence of the PA polypeptide contains multiple potential Ames spore challenge was achieved without any demonstrable N-linked and O-linked glycosylation sites that could add carbo- bacterimia. Thus, it is likely that protection occurs at a stage hydrate moieties as has been reported previously (13, 14). Thus we before vegetative bacterial multiplication and copious toxin pro- were able to reduce the molecular weight of vaccinia-expressed PA duction with the implication that the anti-spore activity of PA- to be closer to the predicted size of 83 kDa by treating the PA with specific antibodies, and not necessarily the toxin-neutralizing PNGase F, which cleaves sugars from asparagine residues (Fig. 1). activity of anti-PA antibodies, is the dominant component in

18092 | www.pnas.org/cgi/doi/10.1073/pnas.1013083107 Merkel et al. Downloaded by guest on October 2, 2021 Table 1. Anthrax vaccine study in rabbits with two doses of vaccine Anti-PA ELISA Anti-vaccinia † titer titer Identiﬁcation Postchallenge Postchallenge Postchallenge Experimental group tag no. Prime Boost TNA titer* Prime Boost survival anti-PA ELISA titer bacteremia

Wyeth/IL-15/PA 38 32,000 640,000 350.8 100 800 Yes 2560,000 Negative 34 128,000 320,000 397.7 400 800 Yes 640,000 Negative 54 64,000 640,000 840.6 100 1,600 Yes 1,280,000 Negative 48 64,000 640.000 448.4 400 1,600 Yes 640,000 Negative 55 16,000 640,000 106.1 <100 800 Yes 2,560,000 Negative 27 16,000 320,000 291.8 <100 400 Yes 1,280,000 Negative 51 32,000 640,000 1,326.5 <100 800 Yes 640,000 Negative 40 32,000 160,000 128.7 100 800 Yes 640,000 Negative 47 32,000 640,000 1,628.8 100 800 Yes 2,560,000 Negative

Biothrax/AVA 41 8,000 160,000 1,186.8 ND ND Yes 640,000 Negative 44 <2,000 80,000 593.1 ND ND Yes 1,280,000 Negative 42 8,000 320,000 4,519.0 ND ND Yes 1,280,000 Negative 25 64,000 320,000 2,103.0 ND ND Yes 640,000 Negative 37 16,000 160,000 1,405.8 ND ND Yes 320,000 Negative 33 16,000 160,000 2,360.0 ND ND Yes 640,000 Negative 56 8,000 320,000 1,074.5 ND ND Yes 1,280,000 Negative 39 16,000 320,000 1,985.8 ND ND Yes 1,280,000 Negative 52 8,000 640,000 2,306.3 ND ND Yes 2,560,000 Positive

Control vaccinia 31 ND ND <100 800 No ND Positive 24 ND ND 100 800 No ND Positive 49 ND ND <100 1,600 No ND Positive 30 ND ND 100 800 No ND Positive 36 ND ND 100 800 No ND Positive 29 ND ND <100 800 No ND Positive

Animals were vaccinated twice 4 wk apart. Recombinant vaccinia vaccines were given s.c. at a dose of 1 x 107 pfu. Biothrax/AVA vaccine was given at a dose of 50 μL/animal intramuscularly. For serological assays, sera were collected 4 wk after primary vaccination (“Prime”), 3 wk after booster vaccination (“Boost”),

or 14 d after spore challenge. Inhalation challenge of B. anthracis (Ames) spores (∼200 LD50) was done 3 wk after booster vaccination. Postchallenge survival was monitored for 14 d. Postchallenge bacteremia was assessed 6 d after spore challenge. ND. not detected. *Serum samples were collected 3 wk after booster vaccination, and LT toxin-neutralizing antibody titers were determined using J774.1 as described previously (15). †Vaccinia virus-neutralizing antibody titers were determined by a plaque reduction assay where reduction of 80% of the in-put virus was taken as the anti- vaccinia titer.

mediating the sterile protection seen in our study, which is similar whereas one rabbit in the Biothrax/AVA group was bacteremic to the observations of Welkos et al. (20) and Stepanov et al. (21). with B. anthracis. However, despite the lack of culture evidence for As a prerequisite for being an effective dual vaccine against anthrax infection, when we assessed the serum samples collected both anthrax and smallpox, we confirmed the ability of Wyeth/ 14 d after the spore challenge from the rabbits vaccinated with IL-15/PA to elicit robust vaccinia virus-neutralizing antibodies Wyeth/IL-15/PA and Biothrax/AVA (Table 1), we noted a two- to that were comparable to the levels generated by Wyeth/IL-15, fourfold rise in PA-specific ELISA antibody titers for most ani- a superior smallpox vaccine that we reported previously (11). mals, although one cannot interpret this to indicate an active in- Notably, the induction of vaccinia-neutralizing antibodies with fection. Perhaps, as has been purported, PA polypeptide is present the first dose of Wyeth/IL-15/PA did not have a negative impact in the spore coat or exosporium (22, 23) and could have triggered on the booster vaccination given 4 wk later in enhancing an anamnestic response in these animals, leading to a rise in anti- vaccinia-specific neutralizing antibodies and the heterologous PA titer. In contrast, all six animals vaccinated with Wyeth/IL-15 PA antigen-directed antibodies, contrary to the commonly held died of anthrax disease with a mean time to death of 3.5 d. The –

view that circulating anti-vaccinia neutralizing antibodies could blood collected at the moribund stage revealed the presence of MEDICAL SCIENCES render subsequent revaccinations ineffective. B. anthracis in all six animals. To ascertain the efficacy of our Wyeth/IL-15/PA vaccine, vaccinated rabbits were challenged 21 d after the second dose of Further Characterization of Vaccinia-Based PA Vaccines in A/J Mice. vaccine. Although our intended aim was to achieve a dose of 200 Having confirmed the potent efficacy of our Wyeth/IL-15/PA LD50 [1 LD50 = 105,000 colony-forming units (cfu)] of spores vaccine against a lethal inhalation spore challenge with the highly from the Ames strain of B. anthracis, the inhalation dosimetry virulent Ames strain of B. anthracis in rabbits, a recognized ani- revealed that the group vaccinated with Wyeth/IL-15/PA received mal model of human disease, we set out to further characterize 7 adoseof240LD50 (2.52 × 10 cfu) whereas the group vaccinated the immunogenicity of our vaccines including the MVA de- 7 with Biothrax/AVA received a dose of 255 LD50 (2.68 × 10 cfu). rivative, MVA/IL-15/PA, in a more economical model using A/J The control group vaccinated with Wyeth/IL-15 received a dose of mice (24). As shown in Fig. 2, when we tested the sera of vacci- 7 290 LD50 (3.05 × 10 cfu). Throughout a 14-d observation period nated mice 3 wk after vaccination, as we demonstrated earlier in after the spore challenge, with daily measurements of body weights rabbits, both Wyeth/IL-15/PA and MVA/IL-15/PA induced ro- and three-times-a-day measurement of body temperature, none of bust anti-PA antibodies in our PA-specific ELISA. In fact, MVA/ the animals vaccinated with Wyeth/IL-15/PA or Biothrax/AVA IL-15/PA appeared consistently slightly more immunogenic than showed any appreciable changes in these parameters. Moreover, the replicative Wyeth/IL-15/PA. Concordant with the ability to blood samples taken 6 d after challenge were culture negative for induce a robust immune response, when we subjected the vacci- B. anthracis in all nine rabbits vaccinated with Wyeth/IL-15/PA, nated mice to an inhalation spore challenge 21 d after the booster

Merkel et al. PNAS | October 19, 2010 | vol. 107 | no. 42 | 18093 Downloaded by guest on October 2, 2021 capable of inducing a protective immune response rapidly. When we assessed the kinetics of anti-PA antibody induction after a single dose of our vaccinia-based dual vaccines in A/J mice (Fig. 4), a demonstrable antibody response was evident by the third day post vaccination, whereas in mice vaccinated with Biothrax/AVA, the appearance of demonstrable anti-PA antibodies required more than a week. As shown in Fig. 5, the induction kinetics of vaccinia-neutralizing antibodies were equally rapid with our dual vaccine with measurable neutralizing antibodies appearing within 6 days of vaccination. More importantly, and consistent with the superior induction kinetics of PA-specific antibodies with our vaccinia-based dual vaccines, when we lethally challenged mice (20 LD50 dose of spores) 3 d after a single dose of vaccine, ≈33% of the mice that were vaccinated with either MVA/IL-15/PA or Wyeth/IL-15/PA survived whereas only Fig. 2. Assessment of PA-specific antibodies in A/J mice vaccinated with 10% (2 of 20) of the mice vaccinated with Biothrax/AVA with- vaccinia-based dual vaccines. Mice were vaccinated twice with Wyeth/IL-15/ stood the lethal spore challenge (Table 3). This difference is PA, MVA/IL-15/PA (1 × 107 pfu/mouse), or Biothrax/AVA (25 μL/mouse) s.c. 4 statistically significant (P = 0.0251). Similarly, in a 6-d post- wk apart, and 3 wk after the second dose of vaccine, animals were bled. vaccination challenge, Wyeth/IL-15/PA again protected 65% of Serial dilutions of pooled sera from each group (n = 6) vaccinated with re- the mice whereas Biothrax/AVA lagged behind with a 35% spective vaccinia recombinant vaccines or Biothrax/AVA were then analyzed survival of vaccinated mice. This reflects the superior efficacy by a PA-specific ELISA to determine the levels of PA-specific antibody in- and rapidity of immune response of our vaccinia-based dual duced by vaccination. vaccines in protecting against inhalation anthrax. Discussion vaccination (20 LD50 dose of spores from an acapsular, toxigenic A US congressional commission on the prevention of weapons of strain of B. anthracis strain 7702), both Wyeth/IL-15/PA and mass destruction, proliferation, and terrorism unanimously con- MVA/IL-15/PA conferred efficient protection that was compa- cluded that, of all means of mass destruction, bioterrorism poses rable to the licensed vaccine Biothrax/AVA (Table 2). An im- the greatest threat to the global community in the coming years portant attribute of a vaccine intended for individuals with (25). In a sobering assessment early this year, the same commis- a potential risk of exposure to anthrax is the durability and the sion assigned a failing grade for our nation’s efforts to enhance the persistence of the vaccine-induced immune response. Therefore, medical countermeasures to rapidly mitigate or prevent mass ca- we evaluated the levels of PA antibodies in A/J mice 6 mo after sualties arising from a biological attack (http://www.preventwmd. vaccination. As shown in Fig. 3, in an ELISA that captures PA- gov/static/docs/report-card.pdf). specfic total IgG levels, the MVA/IL-15/PA–induced antibody Because of their prior use in biowarfare/bioterrorism or the levels were comparable to those induced by Biothrax/AVA in purported past/present existence of military programs for both duration and magnitude. However, when we examined the weaponization, V. major virus and B. anthracis remain at the apex subclass specificity of this PA-specfic IgG response, the immune of potential pathogens that could be used in a bioterror attack to response induced by MVA/IL-15/PA was predominantly of the inflict mass casualties. Although licensed vaccines are available Th-1 type with IgG2a and IgG2b levels that exceeded the levels for both smallpox caused by V. major (ACAM 2000) and anthrax of these IgG subclasses induced by Biothrax/AVA, which pre- caused by B. anthracis (Biothrax/AVA), serious concerns remain dominantly induced a Th-2 type immune response. It is note- as to the deployability of these vaccines in large-scale vaccination worthy that anti-PA antibodies can confer protection against efforts, especially in the aftermath of a bioterror attack involving a lethal spore challenge without the appearance of vegetative these pathogens. For example, although the smallpox vaccine has bacteria in the blood stream, for example, as we have shown been administered to over one billion people in the past and its above in our rabbits vaccinated with Wyeth/IL-15/PA or Biothrax/ stellar efficacy against smallpox is considered a gold standard for AVA (Table 1). Because of the purported presence of PA on the infectious disease vaccines, the residual virulence of this vaccine spore coat surface, it is possible that PA-bound antibodies, es- can cause devastating complications, including death in immu- fi pecially of the Th1 type, are able to initiate powerful effector node cient individuals or people with atopic skin disease who activities such as complement fixation and antibody-dependent account for a sizable swath of the contemporary population (7). cellular cytotoxicity to prevent germination of spores and lead to On the other hand, although a committee convened by the In- their destruction through sporucidal or germination inhibitory stitute of Medicine of the National Academies concluded in mechanisms as has been suggested previously (22). October 2000 that Biothrax/AVA vaccine is effective against all From a biodefense perspective, it is most critical to generate forms of anthrax, the need for a better vaccine was strongly a vaccine that can be used in a postevent situation and that is advocated (26). In developing Wyeth/IL-15/PA, a dual vaccine that is efficacious against two leading deadly pathogens with high bioterror Table 2. Anthrax vaccine study in A/J mice with two doses of potential, we integrated cytokine IL-15 into a licensed smallpox vaccine vaccine with the disruption of the single vaccinia gene hemagglutinin—which does not play an appreciable role in viral path- Vaccine No. of survivors/total (%) ogenesis or replication (27)—primarily to attenuate the residual virulence of the Wyeth strain of vaccinia. IL-15 is a powerful MVA/IL-15/PA 10/10 (100) immunostimulatory cytokine with a wide range of biological ac- Wyeth/IL-15/PA 9/10 (90) tivities (28, 29). It is involved in the activation, proliferation, and Biothrax/AVA 10/10 (100) differentiation of CD8+ T cells and NK cells and in the main- Unvaccinated mice 0/10 (0.0) tenance of CD8+ memory T cells in addition to supporting the Mice were vaccinated twice 4 wk apart s.c.. The dose of recombinant survival of mature dendritic cells. IL-15 has also been implicated vaccinia vaccines was 1 x 107 pfu/mouse and Biothrax/AVA was 25 μL/mouse. in B-cell proliferation and differentiation as well as in antibody

Vaccinated mice were challenged via inhalation route with a dose of 20 LD50 synthesis and secretion along with a role in maintaining sero- using spores of toxigenic acapsular B. anthracis strain 7702 (pXO1+; pXO2−) logical memory (30). It is likely that the effects of IL-15 on NK and monitored for 14 d. cells and innate immune cells facilitate the rapid clearance of

18094 | www.pnas.org/cgi/doi/10.1073/pnas.1013083107 Merkel et al. Downloaded by guest on October 2, 2021 Fig. 3. Long-term persistence and Ig subclass specificities of PA-specific antibodies induced by vaccinia-based dual vaccines. Groups of A/J mice (n = 6) were vaccinated with two doses of either MVA/IL-15/PA (1 × 107 pfu/mouse) or Biothrax/AVA (25 μL/mouse) s.c. 4 wk apart, and their serum samples were collected 6 mo after the second vaccination. Pooled serum samples from each group were then tested in a PA-specific ELISA to determine the total anti-PA IgG levels as well as anti-PA

IgG2a and anti-PA IgG2b levels using either total or subclass- speciﬁc anti-mouse IgG secondary antibodies in the ELISA.

vaccinia virus even in immune-deficient hosts, thereby markedly iation associated with Biothrax/AVA. Wyeth/IL-15/PA, however, reducing the residual virulence associated with the Wyeth strain. is genetically stable and amenable to defined cell culture-based Our previous work has shown that the incorporation of IL-15 production within the existing vaccine-manufacturing capabilities. into vaccinia-derived vaccines induces high avidity, long-lived Moreover, the inherent ability of Wyeth vaccinia virus to be ly- antigen-specific memory cytotoxic T lymphocytes and persistent ophilized without loss of potency makes our Wyeth/IL-15/PA antigen-specific antibody responses, thereby conferring durable vaccine cold-chain independent, thus simplifying the logistics of effective immunity against vaccine antigens (31, 32). It is indeed storage, stockpiling, and field delivery in the event of a bioterror remarkable that Wyeth/IL-15/PA induces a PA-specific pro- attack involving smallpox or anthrax. We believe that these fea- tective antibody response within 72 h post vaccination as we have tures make Wyeth/IL-15/PA a preferred choice for integrating into shown in this study, highlighting the potential utility of this our national biodefense preparedness agenda to protect the nation vaccine in a postevent scenario. The superiority of our vaccinia- by enhancing the capabilities of rapidly responding to and re- based dual vaccine against anthrax as opposed to the poorly covering from a devastating attack involving these bioweapons of protective WR recombinant described by Iacono-Connors (33) is mass destruction. reflective of the use of a codon-optimized synthetic PA gene along with the incorporation of IL-15 as a molecular adjuvant in Materials and Methods our vaccine that enhances the kinetics, magnitude, and longevity Construction of Vaccinia-Based Vaccine Candidates That Express PA Gene of of vaccine-induced antibody responses. B. anthracis. The construction of smallpox vaccine candidates Wyeth/IL-15 It is important to emphasize that the vaccinia-based dual vac- and MVA/IL-15 has been described earlier (11, 34). A synthetic PA gene was cine with integrated IL-15 not only is superior in immunogenicity manufactured de novo that encodes a polypeptide identical to the deduced and efficacy in comparison with the currently licensed vaccines amino acid sequence of the mature PA polypeptide described by Welkos et al. (35) (GenBank accession no. AAA22637.1) with preferred human against smallpox and anthrax, but also remedies the inadequacies codons with a signal peptide derived from the tissue plasminogen activator associated with such licensed vaccines. These inadequacies include polypeptide along with vaccinia expression sequence elements embedded 5′ the residual virulence seen with smallpox vaccines in immune- for proper expression. In constructing Wyeth/IL-15/PA and MVA/IL-15/PA, the deficient hosts and the undefined composition and lot-to-lot var- synthetic PA gene was integrated into the hemagglutinin locus such that IL- 15 and PA genes were in a head-to-tail configuration at the hemagglutinin locus of the respective recombinant viral genomes. Wyeth/IL-15 and Wyeth/ IL-15/PA viral stocks were made and titrated in CV-1 monkey kidney cells, whereas MVA/IL-15/PA stocks were made and titrated in BHK-21 cells.

Measurement of Antibody Titers in Serum Samples from Vaccinated Animals. An ELISA was used to determine the titer of anti-PA–binding antibodies. MEDICAL SCIENCES

Fig. 4. Superior induction kinetics of PA-specific antibodies after vaccination Fig. 5. Induction kinetics of vaccinia-neutralizing antibodies following with vaccinia-based dual vaccines. A group of A/J mice (n = 6) were given vaccination with vaccinia-based dual vaccines. Mice were vaccinated with a single dose of MVA/IL-15/PA vaccine (1 × 107 pfu/mouse) or Biothrax/AVA (25 a single dose of MVA/IL-15/PA vaccine (1 × 107 pfu/mouse), and longitudinal μL/mouse). Animals were bled longitudinally at indicated time points following serum samples were collected as described in the legend to Fig. 4. The in- vaccination, and serial dilutions of pooled sera from each group were tested in duction kinetics of vaccinia-neutralizing antibodies and the quantitation of aPA-specific ELISA to determine the kinetics of PA-specific antibody induction such antibodies were determined using a plaque reduction assay as de- and to quantitate the levels of such PA-specfic antibodies in their sera. Pre- scribed in Materials and Methods. Prebleeds taken from each group before bleeds taken from each group before vaccination served as a baseline control. vaccination served as a baseline control.

Merkel et al. PNAS | October 19, 2010 | vol. 107 | no. 42 | 18095 Downloaded by guest on October 2, 2021 Table 3. Short-term anthrax vaccine study in A/J mice with one Animals, Immunizations, and B. anthracis Spore Challenges. Female New Zea- dose of vaccine land White rabbits ≈10 wk of age were purchased from Myrtle’s Rabbitry. Female A/J mice 6–8 wk of age were purchased from the National Cancer No. of survivors/total (%) Institute’s animal resources program. Rabbits were immunized twice, 4 wk apart, s.c., with the respective vaccinia virus recombinants at a dose of 1 × 107 3-d post vaccination 6-d post vaccination pfu in a volume of 0.5 mL Biothrax/AVA vaccine. This was administered in- Vaccine challenge challenge tramuscularly with the same schedule as for vaccinia recombinants; the dose consisted of 50 μL of Biothrax/AVA vaccine in 450 μL of PBS. Rabbits were chal- MVA/IL-15/PA 6/19 (31.6) 8/17 (47.1) lenged with a mean dose of 262 LD50 of B. anthracis (Ames strain) obtained Wyeth/IL-15/PA 6//18 (33.3)* 13/20 (65.0) from BEI Resources by oro-nasal inhalation under appropriate biosafety con- Biothrax/AVA 2/20 (10.0) 7/20 (35.0) tainment at the Southern Research Institute, Birmingham, AL. In immunizing Unvaccinated mice 0/17 (0.0) 0/18 (0.0) A/J mice, the vaccinia recombinant viruses were administered s.c. at a dose of 1 × 107 pfu in a 100-μL volume. Biothrax/AVA vaccine was administered s.c. in Mice were vaccinated once s.c.. The dose of recombinant vaccinia vaccines a 100-μL volume after 25 μL of Biothrax/AVA vaccine was diluted in 75 μLof was 1 x 107 pfu/mouse and Biothrax/AVA was 25 μL/mouse. Vaccinated mice PBS. Challenging of mice with the spores of toxigenic acapsular B. anthracis were challenged via inhalation route with a dose of 20 LD using spores of 50 strain 7702 (pXO1+; pXO2−) was done as described previously (24) using a six- toxigenic acapsular B. anthracis strain 7702 (pXO1+; pXO2−) and monitored jet Collision nebulizer with nose only exposure system to deliver a 20-LD dose for 14 d. 50 of spores. Housing and care of animals were carried out in accordance with *Statistically significant, P = 0.0251, compared with Biothrax/AVA. the American Association for Accreditation of Laboratory Animal Care stand- ards in accredited facilities. The experimental design of this study was approved by the Institutional Animal Care and Use Committees. Briefly, plates were coated with recombinant PA (List Biological Laborato- ries) at a concentration of 1 mcg/mL in carbonate coating buffer (pH 9.0) Statistical Analyses. Statistical analysis was by unpaired Student’s t test, one- overnight at 4 °C. Plates were then washed three times in washing buffer way ANOVA with pairwise comparisons, and two-way ANOVA with pairwise (PBS with 0.01% Tween 20) and blocked for 2 h at room temperature (RT) comparisons. P < 0.05 was considered significant. with blocking buffer [20 mM Tris, (pH 8.0); 150 mM NaCl; 0.01% Tween 20; 2% FCS; 0.1% BSA]. Serum samples diluted in blocking buffer were then ACKNOWLEDGMENTS. We thank Dr. Toyoko Hiroi for assistance with the added and incubated for 1 h at RT, washed three times with washing buffer graphics and Ms. Jean R. Decker for contractual administrative support. We and appropriate horseradish peroxidase-conjugated species-specificsec- gratefully acknowledge the receipt of reagents from the Biodefense and ondary antibody diluted in blocking buffer added, and incubated for 1 h at Emerging Infections Research Resources Repository (BEI Resources, National RT. Plates were then washed three times, and the bound antibodies were Institute of Allergy and Infectious Diseases) for this study. This work was supported in part by the Intramural Research Program of the National Can- detected by colorimetry. Lethal toxin-neutralizing antibody titers were cer Institute, Center for Cancer Research, National Institutes of Health, and determined essentially as described previously using J774.1 cells (15). Anti- by a 3-y competitive research funding award (to L.P.P.) from the Trans- vaccinia–neutralizing antibody titers were determined by a plaque reduction National Institutes of Health/Food and Drug Administration Intramural assay as previously reported (11). Biodefense Program.

1. World Health Organization (1970) Health Aspects of Chemical and Biological Weap- 19. Livingston BD, Little SF, Luxembourg A, Ellefsen B, Hannaman D (2010) Comparative ons: A Report of a WHO Group of Consultants (World Health Organization, Geneva, performance of a licensed anthrax vaccine versus electroporation based delivery of Switzerland). a PA encoding DNA vaccine in rhesus macaques. Vaccine 28:1056–1061. 2. Cieslak TJ, et al. (2000) Immunization against potential biological warfare agents. Clin 20. Welkos S, Little S, Friedlander A, Fritz D, Fellows P (2001) The role of antibodies to – Infect Dis 30:843 850. Bacillus anthracis and anthrax toxin components in inhibiting the early stages of 3. Russell PK (2007) Project BioShield: What it is, why it is needed, and its infection by anthrax spores. Microbiology 147:1677–1685. – accomplishments so far. Clin Infect Dis 45(Suppl 1):S68 S72. 21. Stepanov AV, Marinin LI, Pomerantsev AP, Staritsin NA (1996) Development of novel 4. Mair M (2007) Recent progress in biodefence countermeasure development. Biosecur vaccines against anthrax in man. J Biotechnol 44:155–160. Bioterror 5:93–96. 22. Cote CK, et al. (2005) The detection of protective antigen (PA) associated with spores 5. Fenner F, Henderson DA, Arita I, Jecek Z, Ladnyi ID (1998) Smallpox and Its Eradica- of Bacillus anthracis and the effects of anti-PA antibodies on spore germination and tion History of International Public Health (World Health Organization, Geneva, macrophage interactions. Microb Pathog 38:209–225. Switzerland). 23. Schumacher WC, Storozuk CA, Dutta PK, Phipps AJ (2008) Identification and 6. Fenner F (1989) Risks and benefits of vaccinia vaccine use in the worldwide smallpox fl eradication campaign. Res Virol 140:465–466, discussion 487–491. characterization of Bacillus anthracis spores by multiparameter ow cytometry. Appl – 7. Greenberg RN, Kennedy JS (2008) ACAM2000: A newly licensed cell culture-based live Environ Microbiol 74:5220 5223. vaccinia smallpox vaccine. Expert Opin Investig Drugs 17:555–564. 24. Loving CL, Kennett M, Lee GM, Grippe VK, Merkel TJ (2007) Murine aerosol challenge 8. Frankel AE, et al. (2009) Pathophysiology of anthrax. Front Biosci 14:4516–4524. model of anthrax. Infect Immun 75:2689–2698. 9. Henderson DW, Peacock S, Belton FC (1956) Observations on the prophylaxis of 25. Graham B, et al. (2008) World at Risk: The Report on the Commission on the experimental pulmonary anthrax in the monkey. J Hyg (Lond) 54:28–36. Prevention of WMD Proliferation and Terrorism (Vintage Books, New York), 132pp. 10. Friedlander AM, Little SF (2009) Advances in the development of next-generation 26. IOM Committee to Assess the Safety and Efficacy of the Anthrax Vaccine (2002) The anthrax vaccines. Vaccine 27(Suppl 4):D28–D32. Anthrax Vaccine: Is It Safe? Does It Work? (National Academies Press, Washington, DC). 11. Perera LP, et al. (2007) Development of smallpox vaccine candidates with integrated 27. Buller RML, Palumbo GJ (1992) Safety and attenuation of vaccinia viruses. Recombinant interleukin-15 that demonstrate superior immunogenicity, efficacy, and safety in Poxviruses, eds Binns MM, Smith GL (CRC Press, Boca Raton, FL), pp 235–267. mice. J Virol 81:8774–8783. 28. Waldmann TA (2006) The biology of interleukin-2 and interleukin-15: Implications for 12. Zielinski RJ, et al. (2010) Smallpox vaccine with integrated IL-15 demonstrates cancer therapy and vaccine design. Nat Rev Immunol 6:595–601. enhanced in vivo viral clearance in immunodeficient mice and confers long term 29. Rochman Y, Spolski R, Leonard WJ (2009) New insights into the regulation of T cells protection against a lethal monkeypox challenge in cynomolgus monkeys. Vaccine, by gamma(c) family cytokines. Nat Rev Immunol 9:480–490. 10.1016/j.vaccine.2010.08.008. 30. Bernasconi NL, Traggiai E, Lanzavecchia A (2002) Maintenance of serological memory 13. Lee JS, Hadjipanayis AG, Welkos SL (2003) Venezuelan equine encephalitis virus-vectored by polyclonal activation of human memory B cells. Science 298:2199–2202. – vaccines protect mice against anthrax spore challenge. Infect Immun 71:1491 1496. 31. Oh S, Berzofsky JA, Burke DS, Waldmann TA, Perera LP (2003) Coadministration of 14. McConnell MJ, Hanna PC, Imperiale MJ (2006) Cytokine response and survival of mice HIV vaccine vectors with vaccinia viruses expressing IL-15 but not IL-2 induces long- immunized with an adenovirus expressing Bacillus anthracis protective antigen lasting cellular immunity. Proc Natl Acad Sci USA 100:3392–3397. domain 4. Infect Immun 74:1009–1015. 32. Oh S, Perera LP, Burke DS, Waldmann TA, Berzofsky JA (2004) IL-15/IL-15Ralpha- 15. Verma A, et al. (2008) Role of the N-terminal amino acid of Bacillus anthracis lethal mediated avidity maturation of memory CD8+ T cells. Proc Natl Acad Sci USA 101: factor in lethal toxin cytotoxicity and its effect on the lethal toxin neutralization – assay. Clin Vaccine Immunol 15:1737–1741. 15154 15159. 16. Pittman PR, et al. (2005) Protective antigen and toxin neutralization antibody 33. Iacono-Connors LC, Welkos SL, Ivins BE, Dalrymple JM (1991) Protection against patterns in anthrax vaccinees undergoing serial plasmapheresis. Clin Diagn Lab anthrax with recombinant virus-expressed protective antigen in experimental Immunol 12:713–721. animals. Infect Immun 59:1961–1965. 17. Little SF, et al. (2004) Defining a serological correlate of protection in rabbits for 34. Earl PL, Moss B, Wyatt LS, Carroll MW (2001) Generation of recombinant vaccinia a recombinant anthrax vaccine. Vaccine 22:422–430. viruses. Curr Protoc Protein Sci, 5.13.1–5.13.19. 18. Pitt MLM, et al. (2001) In vitro correlate of immunity in a rabbit model of inhalational 35. Welkos SL, et al. (1988) Sequence and analysis of the DNA encoding protective anthrax. Vaccine 19:4768–4773. antigen of Bacillus anthracis. Gene 69:287–300.

18096 | www.pnas.org/cgi/doi/10.1073/pnas.1013083107 Merkel et al. Downloaded by guest on October 2, 2021