- Email: [email protected]
Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine
G Model
ARTICLE IN PRESS
JVAC-17154; No. of Pages 7
Vaccine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Review
Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine Ryan C. McComb, Mikhail Martchenko ∗ Keck Graduate Institute, School of Applied Life Science, 535 Watson Dr., Claremont, CA, United States
a r t i c l e
i n f o
Article history: Received 7 August 2015 Received in revised form 5 October 2015 Accepted 9 November 2015 Available online xxx Keywords: Anthrax Bacillus anthracis Protective antigen Neutralizing antibody mapping Functional mapping Rational vaccine design
a b s t r a c t Anthrax is defined by the Centers for Disease Control and Prevention as a Category A pathogen for its potential use as a bioweapon. Current prevention treatments include Anthrax Vaccine Adsorbed (AVA). AVA is an undefined formulation of Bacillus anthracis culture supernatant adsorbed to aluminum hydroxide. It has an onerous vaccination schedule, is slow and cumbersome to produce and is slightly reactogenic. Next-generation vaccines are focused on producing recombinant forms of anthrax toxin in a well-defined formulation but these vaccines have been shown to lose potency as they are stored. In addition, studies have shown that a proportion of the antibody response against these vaccines is focused on non-functional, non-neutralizing regions of the anthrax toxin while some essential functional regions are shielded from eliciting an antibody response. Rational vaccinology is a developing field that focuses on designing vaccine antigens based on structural information provided by neutralizing antibody epitope mapping, crystal structure analysis, and functional mapping through amino acid mutations. This information provides an opportunity to design antigens that target only functionally important and conserved regions of a pathogen in order to make a more optimal vaccine product. This review provides an overview of the literature related to functional and neutralizing antibody epitope mapping of the Protective Antigen (PA) component of anthrax toxin. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction to rationally designed vaccines Developments in neutralizing antibody epitope mapping and structural determination of functional regions of pathogens and virulence factors are enhancing the capabilities for designing vaccine antigens that elicit targeted neutralizing antibody responses. This approach has been termed “rational” vaccination since the starting points for antigen design begin with thoroughly understanding the mechanism, by which virulence is blocked by the immune system [1,2] and is currently being explored for a number of intractable diseases such as HIV [3], Hepatitis C [4] and Influenza [5]. The strength in this approach lies in the fact that native antigens are not adapted for stable storage, optimal production, or elicitation of neutralizing antibodies to conserved functional regions leading to subsequent immunity [2]. Guided antigen design, by utilizing structural information and neutralizing antibody epitopes, provides an opportunity for developing potentially more effective
∗ Corresponding author. Tel.: +1 909 6070038. E-mail address: Mikhail [email protected] (M. Martchenko).
immunotherapeutics and vaccines that are better suited for production and storage in addition to being optimized for inducing a targeted immune response that may lead to long-lasting immunity. 1.1. Anthrax pathogenesis and biowarfare Anthrax is caused by the Gram-positive bacterium Bacillus anthracis, which has afflicted humans and their livestock for centuries. B. anthracis forms endospores and infection occurs when an animal inhales, ingests, or has contact with the spores through broken skin [6]. Spores germinate into vegetative bacteria and secrete the proteins Protective Antigen (PA), Edema Factor (EF) and Lethal Factor (LF). Anthrax PA, EF, and LF are the major pathogenesis factors of B. anthracis, as these toxins have been shown to inhibit phagocytosis process of myeloid cells. LF–PA causes host lethality through targeting the cardiomyocytes and vascular smooth-muscle cells, whereas EF–PA induces lethality mostly by targeting hepatocytes [7]. High mortality rates [6,8] require special actions for public health preparedness against an attack with B. anthracis [9]. Anthrax Vaccine Adsorbed (AVA) is the only vaccine approved to prevent anthrax infection. AVA consists of B. anthracis culture
http://dx.doi.org/10.1016/j.vaccine.2015.11.025 0264-410X/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
2
Fig. 1. Anthrax toxin cellular pathogenesis; PA83 or PA63 binds to cell surface receptors CMG2 and TEM8. PA83 undergoes cleavage by cellular furin releasing the 20 kD fragment. PA63 monomers bound to cellular receptors form an oligomer that is capable of binding three LF or EF molecules. Endocytosis of the PA63 oligomer bound to LF or EF, and acidification of the endosome, result in translocation of LF or EF to the cytosol where they act by cleaving Mitogen Activated Protein Kinase Kinases and Nlrp1 and causing cell death or converting ATP to cAMP and inhibiting phagocytosis.
supernatant adsorbed to aluminum hydroxide and requires a long initial immunization time frame of intramuscular injections at 0, 1 and 6 months. Vaccine recipients are considered protected after completing this initial series. Following the initial series, booster doses are administered at 12 and 18 months and annually thereafter [10]. This schedule is not ideal for use in emergency scenarios or for military personnel with inflexible deployment schedules. AVA consists of undefined components and causes local (e.g., erythema, swelling, pain or tenderness, itching, and nodules) and systemic (e.g., fever, chills, myalgia, arthralgia, and malaise) reactions in some individuals, which have fostered a perception that the vaccine is unsafe. In response to these shortcomings, a Department of Defense commissioned study in 2001 outlined goals for future anthrax vaccine development [11]. According to this report, an ideal anthrax vaccine would consist of defined components that elicit sufficient immunity within 30 days after 2–3 doses and protect against aerosolized anthrax for at least 1 year after initial immunization. Furthermore, this vaccine should remain stable for a long period of time and have minimal local and systemic adverse reactions. Finally, this vaccine should be easily scaled up to ensure product consistency [11]. 1.2. Next generation anthrax vaccines Most of the focus on next-generation anthrax vaccines has been on recombinantly produced PA (rPA) from various expression systems. Recombinant PA vaccines that are furthest along in development are those that are produced in prokaryotic systems such as avirulent, nontoxigenic and sporulation deficient strains of B. anthracis, Escherichia coli, or Pseudomonas fluorescens [12–16]. Several of these rPA vaccines have been evaluated in animal studies and human phase 1 clinical trials and show that rPA with Alhydrogel® adjuvant produces comparable Total Neutralizing Antibody titers and anti-PA IgG titers to AVA. Production of rPA in prokaryotic cells has several advantages including expression and secretion of authentically folded protein into the culture medium for easier purification. Also, the development of avirulent, non-toxigenic and
sporulation deficient strains of B. anthracis improves safety during manufacturing [17]. Although rPA vaccines are a step forward, limitations still exist. For example, rPA adsorbed to aluminum hydroxide adjuvant has been shown to lose its ability to induce toxin neutralizing antibodies as it is stored [18,19]. Furthermore, recent studies have also found that the human antibody response induced by AVA is biased toward non-neutralizing epitopes [20–22]. Also, important functional regions of PA have been shown to be shielded from the immune response induced by AVA [23]. These limitations provide justification for a rationally designed vaccine that targets functionally important and neutralizing epitopes on PA toxin, which can be optimized for stability, safety, production, and a reduced vaccine schedule. This article provides an overview of the critical regions of PA that might be targeted for production of such a vaccine. 1.3. Neutralizing antibody and functional epitope mapping of PA The PA molecule is an 83 kD protein comprised of four domains. After binding to cellular receptors [24,25], PA83 undergoes cleavage by a cellular furin to release a 20 kD fragment from domain I [26]. Furin cleavage exposes the ligand (LF/EF) binding sites and allows for receptor bound PA63 to oligomerize with other PA63 molecules (Fig. 1) [27]. Domain I is thus broken into PA20, which has no known function at this point in time, and Domain I , which remains attached to the PA63 molecule. Amino acids and sequences necessary for the function of PA have been discovered through amino acid mutagenesis and crystal structure determination (Fig. 2). In addition, monoclonal neutralizing antibodies have been used to elucidate vulnerable epitopes of PA (Fig. 3). These PA epitopes are located on all four domains of PA, and some of them, when blocked, are sufficient by themselves for toxin neutralization. 1.3.1. PA20 Although the PA20 region is not known to have a functionally important role in anthrax toxin pathogenesis, neutralizing antibodies targeting this domain have been identified. The monoclonal
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
3
Fig. 2. Amino acids associated with functionally important processes in cellular pathogenesis; Location and process of functionally important amino acids (A); location of residues described in table on PA63 bound to cellular receptor CMG2 (B) (Picture, PD:1T6B [39] visualized with Cn3D software); Residues not shown: 1–174 (PA20), 276–287, 304–319, 457–476, and 596–607.
antibody (mAb) 47F12, which was isolated from a human donor vaccinated with AVA, showed the ability to neutralize LF–PA (LT) in vitro [28]. Experiments with radiolabeled PA83 showed that 47F12 neutralizes through inhibition of furin cleavage. Analysis of 47F12 binding against a yeast library displaying randomly mutated PA20 on its surface found that binding was dependent on the fidelity of the amino acids corresponding to E95 , N98 , A100 , N104 , and I106 of PA20. In a separate study, a screen of AVA vaccinated donor serum tested against overlapping peptides of PA identified N102 –Q115 as an epitope, which bound 5/6 donor samples associated with high neutralization of LT as determined in an in vitro cellular assay with Raw264.7 macrophages [29]. This shows that
antibodies directed against this region contribute to neutralization, but it is unknown whether they are sufficient on their own for providing protection against anthrax toxin. The sequence R164 KKR167 that spans the interface between PA20 and PA63 is the furin cleavage site [30]. Variants of PA with residues 163–168 deleted are insensitive to furin cleavage and are non-toxic when administered with LF in cellular assays and in rats [31]. The mAb 7.5G was isolated from mice immunized with PA83 and recognizes the sequence between L156 and 170 S, which overlaps this site [32,33]. However, it is unclear from the data whether this antibody can inhibit furin cleavage [33]. When purified and administered to mice 24 h prior to LT challenge, it was found that 1 mg of 7.5G
Fig. 3. mAbs with defined epitopes that bind PA63; mAb epitope location, inhibited function, and neutralization capacity (A); Location of epitopes described in table on PA83 bound to cellular receptor CMG2 (B) (Picture, PD: 1T6B [39] visualized with Cn3D software); Residues not shown: 1–15, 159–174, 276–287, 302–319, 457–476, and 596–607.
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7 4
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
was needed to prolong survival of mice challenged with 100 g of LT by 3 days. When the peptide epitope of 7.5G was synthesized as a Multiple Antigenic Peptide (MAP) and administered to mice in Complete Freund’s Adjuvant (CFA), it was claimed that sera samples from vaccinated animals “moderately” protected J774.A macrophages from LT challenge [32]. Since the mice were not challenged with toxin, however, it is impossible to know if the antibody species and titers obtained post-vaccination were sufficient for in vivo neutralization. In another study, human antibodies directed against the sequence between N162 and N180 of PA were affinity purified from AVA vaccinated donors and tested for their ability to neutralize LT in vitro and in vivo [29]. This study showed that furin cleavage site directed antibodies protected 70% of Raw264.7 macrophages after treatment with LT but only 1 of 10 A/J mice survived challenged with 3 × LD50 LT after treatment with 30 g of affinity purified antibodies [29]. It appears that, although furin cleavage is an attractive target for focused therapies from a functional standpoint, there are limited levels of protection that can be achieved with antibodies directed against this epitope. 1.3.2. Domain I Domain I is the remainder of domain I after PA20 is cleaved by furin. A number of antibody epitopes have been mapped to this domain and correspond with functionally important amino acids sequences. For example, point mutations introduced into domain I showed that replacement of any of seven residues (R178 , K197 , R200 , P205 , I207 , I210 and K214 ) with alanine almost completely eliminated LF/EF binding to PA [34]. Analysis of the location of these seven residues in their 3-dimensional conformation on the crystal structure of the PA oligomer reveals that the LF/EF binding site spans across two PA molecules. Two murine antibodies, 19D9 and 20G7, bind the peptide sequence from V196 to I210 that include many of these vulnerable amino acids associated in LF/EF ligand binding [32]. Interestingly, 19D9 (IgG) is neutralizing at 1 g/ml against LT in cellular assays while 20G7 (IgM) is not neutralizing at concentrations up to 50 g/ml. The peptide sequence was synthesized as a MAP and used to immunize mice in the same manner as discussed above with the 7.5G peptide epitope. Neutralizing titers of 1:800 were observed from serum of vaccinated animals in a cellular LT cytotoxicity assay but no in vivo survival data was reported. Finally, Crowe and colleagues report that sera from AVA vaccinated human donors reacted with the peptide Y192 to P205 in a solid-phase epitope mapping experiment but no neutralization assays or toxin survival experiments in animals were performed [29]. These studies show that antibodies targeting the LF/EF binding site of PA are likely critical for toxin neutralization and support including this as a potential vaccine target. However, the fact that the LF/EF binding site spans two PA molecules in the oligomer may require an antigen mimic that closely resembles its native conformational structure. Another region with functional significance in domain I was found that included amino acids I210 , K225 , T240 , and K245 . Cysteine substitutions at any of these 4 amino acids resulted in the inhibition of PA activity by at least 100-fold [35]. Antibodies directed against the epitope P232 to V247 were found in 4 of 6 serum samples from AVA vaccinated human donors with high neutralizing serum titers [29]. Affinity purified antibodies directed against P232 to V247 were tested against LT and shown to provide 50% protection of Raw264.7 cells. Furthermore, 30 g of these affinity purified antibodies were shown to protect 3 of 10 mice challenged with 3 × LD50 LT [29]. Although this sequence is near the LF/EF binding site it is unclear what effect inhibition of this epitope has on the normal function of PA. The proximity of this sequence to the LF/EF binding site and its ability to partially protect cells and mice from LT warrants further investigation into its potential usefulness as a vaccine antigen target.
1.3.3. Domain II Neutralizing antibodies with epitopes that overlap known functional regions also exist in domain II of PA. The 22–23 loop (E302 to S325 ), which contains a chymotrypsin site (F313 to F314 ), has been shown to be involved in translocating LF/EF to the cytosol [36]. Crystal structure determination of PA and experimentation with cysteine substitutions along amino acids 302–325 provide strong evidence that 22–23 loops from individual PA molecules in the heptamer complex combine to form a 14-stranded transmembrane -barrel induced by the low-pH of the endosome [37–39]. Furthermore, when this loop was substituted with a homologous membrane-inserting loop from iota-b toxin of Clostridium perfringens and mixed with wild-type PA in equimolar ratios, a dominant-negative phenotype was observed, in which LT killing was completely inhibited in cellular studies and in rats [30]. These studies provide strong evidence that this region is necessary for LF/EF toxicity mediated through PA. Various studies report antibodies that bind PA at this location but differ in effectiveness for neutralizing LT. Murine monoclonal antibodies (5E12, 2A8 and 5E1) mapped to the chymotrypsin sequence (312–315) and were shown to inhibit chymotrypsin cleavage of PA into 47 kD and 37 kD fragments [40]. Although the neutralization capacity of antibodies 5E12, 2A8 and 5E1 is claimed, no percent survival values are reported from cellular assays. In another report, antibodies 2H9 and 16A12 mapped to the region between amino acids 312–326 and were not shown to be neutralizing [32]. However, Gubbins and colleagues report the discovery of three antibodies (F20G75, F20G76 and F20G77) that recognize the consensus sequence A311 SFFD315 within the 22–23 loop [41]. Cellular assays, in which dilutions of each antibody was incubated with LT toxin for 1 h at 37 ◦ C prior to addition onto J774.A macrophages, show that 90–100% protection was achieved for F20G75, F20G76 and F20G77 antibodies at concentrations as low as 12.5, 11.8 and 16.0 ng/ml, respectively. Interestingly, when the assay format was changed such that PA was allowed to incubate with cells prior to the addition of LF and antibodies, 90% protection was achieved only with higher concentrations (1–10 g/ml). This observation provides evidence that this epitope is inaccessible once PA has bound to the cellular membrane. Discordance between antibody neutralization capabilities targeting the 22–23 loop are possibly indicative of specific affinity/avidity, antibody class, or epitope fine tuning requirements that are necessary to achieve effective neutralization. Functional analysis and mAb neutralization studies give a rational basis behind targeting the 22–23 loop in a vaccine formulation. Based on this data, there have been reports of vaccines that target the epitope comprising amino acids G305 to S319 . The first attempt at targeting this epitope was performed by inserting this amino acid sequence onto the self-assembling Hepatitis B core protein scaffold [42,43]. Chimeric Hepatitis B core proteins carrying the 22–23 loop assembled to form virus like particles. Additionally, 50 g of virus particles administered without adjuvant was sufficient for protecting 4 of 7 guinea pigs challenged with B. anthracis spores (40 × LD50 ) compared to 2 of 3 animal survivors that received rPA vaccination with adjuvant [42]. The second attempt was made by creating a synthetic peptide constructed as a MAP on a lysine backbone. This vaccine was shown to induce protective antibodies in vitro and in vivo [44,45]. Seven New Zealand White Rabbits were immunized initially with 250 g of MAP comprising the 305–319 sequence from PA and a helper T-cell epitope from Plasmodium falciparum in CFA followed by four 125 g vaccine boosts in Incomplete Freund’s Adjuvant (IFA) at 2 week intervals. Vaccinated rabbits were challenged through the inhalational route with 200 × LD50 B. anthracis spores. Seven out of seven MAP-305319 vaccinated rabbits survived spore challenge compared to 7/7, 0/6 and 0/6 PA vaccinated, MAP control vaccinated or naïve control rabbits, respectively [45]. Antibodies targeting this region are
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
induced at low frequencies in humans, rabbits, and non-human primates by AVA or whole rPA vaccines [23]. The 22–23 loop has a necessary role in anthrax toxin pathogenesis and is an important target for next generation vaccines especially since current vaccines induce low antibody titers against this site. Mutational analysis and mAb neutralization has implicated a second region in domain II necessary for anthrax toxin pathogenesis. PA mutants with single amino acid substitutions in the region between amino acids 337 S to 458 N were found to reduce LT toxicity by at least 100-fold in a cellular toxicity assay [35]. More specifically, cysteine substitutions at I364 , T380 , S382 , T393 , N399 , Y411 , N422 and changes at D425 K, K397 D and F427 A exhibited a dominantnegative phenotype, in which PA containing any of these mutations mixed with wild-type PA inhibited death caused by LT in treated cells [35,46,47]. Further analysis showed that these PA mutants all retained the ability to bind cellular receptors, form heptamers, and bind LF but were deficient in endosomal pore formation and LF translocation to the cytosol [35]. Only one neutralizing mAb has mapped to the region between I364 and F427 and is the IgG1 mAb 48.3, which recognizes an epitope consisting of S412 to I419 [48]. Unexpectedly, the mechanism, by which 48.3 provides protection is through inhibition of the furin cleavage step of PA83. There is no obvious explanation for the ability of this antibody to inhibit furin cleavage based on its epitope location since analysis of the crystal structure shows that the furin cleavage site and amino acids 412–419 are not co-localized and are even located on opposite faces of the PA molecule. It could be that this site is required for furin recognition and stabilization for subsequent cleavage at R164 KKR167 site. Protective efficacy of mAb 48.3 was shown in mice challenged with a subcutaneous infection of B. anthracis spores. In this study, 6/8 and 2/8 mice treated with 100 g or 10 g, respectively, of antibody survived challenge of a 10 × LD50 dose of spores [48]. Further evidence for the importance of this region is found in the fact that human serum antibodies from 4/6 AVA vaccinated donors cross-reacted with peptides corresponding to amino acids 406–419 [29]. These studies suggest that amino acids located between I364 and F427 play a primary functional role in anthrax toxin pathogenesis and that antibody epitopes within this region are an important component of a protective immune response. This region, therefore, represents another potential vaccine target for epitope focused therapies. 1.3.4. Domain III From a functional standpoint, domain III appears to be involved with PA oligomer formation. Insertion of missense point mutations within domain III at amino acids D512 , L514 , and D520 blocked detection of PA63 oligomer assembly suggesting that these residues are important in this step anthrax toxin pathogenesis [49]. In a separate study, a point mutation at E515 inhibited the activity of PA by 100-fold in cellular toxicity assays [35]. From the crystal structure it is apparent that this region of PA forms a loop that is in close contact with amino acids P173 to A258 of domain I in a neighboring PA monomer stabilizing the oligomer complex [50]. Although functionally important, no neutralizing mAbs have been reported that bind this region of PA. It has been determined that antibodies directed against domain III contribute to neutralization of LT in vitro. However, none of these antibodies appear to bind regions, which are correlated with functional importance as described previously. For example, recombinantly expressed and purified domain III of PA reduces the ability of serum from PA vaccinated rabbits, non-human primates and humans to neutralize native PA [51]. This study did not identify epitopes on domain III, which were associated with this competitive inhibitory effect. In another study, antibodies from mouse hybridomas (2D3, 2D5 and 10D2) were discovered that bind between I581 and N601 , which spans the end of domain III and
5
the beginning of domain IV [52]. Interestingly, the data suggests that the mechanism, by which 2D3, 2D5 and 10D2 confer protection seems to be by inhibition of LF binding to PA suggesting that this region interacts closely with domain I . Although these mAbs showed neutralization of LT in J774.A cellular assay, they were not capable of protecting Fisher 344 Rats from death. Rats challenged with 13 × LD50 LT incubated with up to 1.7 mg IgG from ascites for 1 h only delayed death by 18 h (2D3 and 2D5) or 0 h (10D2 and IgG negative control) [52,53]. These studies show that mAbs targeting discrete epitopes on domain III do contribute to a neutralizing response against LT but are not sufficient on their own to neutralize LT in vivo. This fact is supported by a number of other studies that report mAbs that bind domain III but on their own are incapable of neutralizing LT. For example, antibody 2-A7, that was found to bind amino acids G532 to Q543 , did not protect mice against LT challenge [54]. Also, Fisher 344 Rats challenged with LT and administered domain III binding antibody 8A7 at a 1:1 (antibody: PA) molar ratio also all died [22]. However, when 9 g of 8A7 was administered in combination with 9 g of 2A6 (an antibody whose epitope is undetermined but also was unable to protect Fisher 344 Rats against LT challenge on its own) 5/5 rats survived a lethal dose of anthrax toxin. This data supports the conclusion that domain III antibodies may need to work in concert with antibodies against other regions on PA in order to provide sufficient protection against anthrax toxin. 1.3.5. Domain IV Domain IV contains important functional sequences for epitope targeted therapeutics. Numerous studies have shown that antibodies directed solely against domain IV are sufficient for neutralizing anthrax toxin. Crystal structure analysis of PA bound to cellular receptor shows that this interaction is primarily attributed to amino acids found in three loops comprising amino acid regions E654 –M662 , Y681 –Y688 , and E712 –G714 of domain IV and one loop from domain II [39,50]. Consistent with the prediction that these amino acids have important functional activity due to their role in binding cellular receptors is the observation that cysteine substitutions at residues I656 , N657 , I665 , N682 , D683 , and L687 inhibit PA activity by at least 100-fold [35]. An additional study shows that mutations in the loop containing amino acids K679 to N693 were most detrimental to PA toxicity, whereas mutations between E704 to K722 allowed PA to retain its toxicity [55]. Interestingly, it was found that an alanine substitution at D683 had the most significant effect (1000-fold reduction) on PA toxicity and binding [56]. Some of the most potent mAbs for neutralizing PA have the ability to block binding to cellular receptors. Neutralizing murine antibodies 3B6 (926 g) and 14B7 (23 g) protected 4/4 Fisher 344 Rats challenged with LT [53]. These antibodies were subsequently shown to bind between amino acids D671 and I721 [52]. Further studies with alanine substitutions at specific residues in the PA binding domain showed that K684 , L685 , L687 and Y688 were critical for 14B7 recognition and neutralization [56]. A high affinity (KD = 3.4 nM), highly neutralizing (50% inhibitory concentration, 5.6 nM) Fab isolated from an immunized macaque designated as 35PA83 was found to also bind between P686 and Y694 [57]. Correlation between PA neutralization with antibodies 14B7 and 35PA83 with mutagenesis studies show that the receptor binding loop between K679 to N693 is an important target for anthrax toxin countermeasures. Two additional murine antibodies were found to bind regions of domain IV that inhibit PA binding to receptors and show both in vitro and in vivo neutralization efficacy [54]. Monoclonal antibodies 1-F1 and 2-B12 bind to linear epitopes located between P692 –K703 and T716 –F727 , respectively. High affinities (1-F1, 1.7 nM and 2-B12, 2.0 nM) and in vitro neutralization of LT were reported. Additionally, 550 g/mouse of 1-F1 or 2-B12 protected 2/5 and 3/5 BALB/c mice challenged with a lethal dose of LT, respectively. Given
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
6
the close proximity of these epitopes to the functionally essential loop between K679 and N693 it is not surprising that these antibodies have some ability to inhibit anthrax toxin. It appears though, that these antibodies are not as effective at blocking toxin entry as 14B7 and 35PA83 that recognized the loop between amino acids 679 and 693. Another epitope associated with domain IV was found to exist between L628 and K637 [29]. Antibodies against this peptide sequence were affinity purified from AVA vaccinated human serum and a 30 g/mouse dose was found to protect 6/10 mice from 3 × LD50 LT challenge. The specific mode of inhibition for these antibodies was not determined. Domain IV is a crucial target for anthrax toxin therapies based on PA functional studies, crystal structure analysis, and the potency, by which antibodies targeting this region protect cells and animals against anthrax toxin. 2. Conclusion In summary, PA contains many functional regions that can be targeted for rationally designed therapeutics and vaccines. These regions come from all four domains of PA and include antibody epitopes and functional regions that, if blocked, are sufficient by themselves for anthrax toxin neutralization. On the other hand, PA contains many more partially neutralizing epitopes and functional sites of inhibition. These studies suggest that blocking multiple regions of the PA toxin simultaneously will likely provide more complete protection. A rationally designed vaccine will, more likely than not, need to include a variety of PA antigen targets to induce complete immunity against anthrax and any potential PA mutants that can evade a particular targeted vaccine. Conflict of interest statement The authors report no conflict of interests. References [1] Burton DR. Antibodies, viruses and vaccines. Nat Rev Immunol 2002;2:706–13, http://dx.doi.org/10.1038/nri891. [2] Dormitzer PR, Ulmer JB, Rappuoli R. Structure-based antigen design: a strategy for next generation vaccines. Trends Biotechnol 2008;26:659–67, http://dx.doi.org/10.1016/j.tibtech.2008.08.002. [3] Jardine J, Julien J-P, Menis S, Ota T, Kalyuzhniy O, McGuire A, et al. Rational HIV immunogen design to target specific germline B cell receptors. Science 2013;340:711–6, http://dx.doi.org/10.1126/science.1234150. [4] Kong L, Jackson KN, Wilson IA, Law M. Capitalizing on knowledge of hepatitis C virus neutralizing epitopes for rational vaccine design. Curr Opin Virol 2015;11:148–57, http://dx.doi.org/10.1016/j.coviro.2015.04.001. [5] Kanekiyo M, Wei C-J, Yassine HM, McTamney PM, Boyington JC, Whittle JRR, et al. Self-assembling influenza nanoparticle vaccines elicit broadly neutralizing H1N1 antibodies. Nature 2013;499:102–6, http://dx.doi.org/10.1038/nature12202. [6] Adalja AA, Toner E, Inglesby TV. Clinical management of potential bioterrorism-related conditions. N Engl J Med 2015;372:954–62, http://dx.doi.org/10.1056/NEJMra1409755. [7] Liu S, Moayeri M, Leppla SH. Anthrax lethal and edema toxins in anthrax pathogenesis. Trends Microbiol 2014;22:317–25, http://dx.doi.org/10.1016/j.tim.2014.02.012. [8] Riedel S. Anthrax: a continuing concern in the era of bioterrorism. Proc Bayl Univ Med Cent 2005;18:234–43. [9] Office of Public Health Preparedness and Response. First hours: bioterrorism agents. Cent Dis Control Prev 2013. http://emergency.cdc.gov/ firsthours/bioterrorism.asp (accessed June 9, 2015). [10] Emergent BioSolutions. BioThrax (Anthrax Vaccine Adsorbed) Package Insert; 2012. [11] Committee to Assess the Safety and Efficacy of the Anthrax Vaccine. In: Joellenbeck LM, Zwanziger LL, Durch JS, Strom BL, et al., editors. The anthrax vaccine: Is it safe? Does it work? Washington, DC: National Academy Press; 2002. [12] Bellanti JA, Lin F-YC, Chu C, Shiloach J, Leppla SH, Benavides GA, et al. Phase 1 study of a recombinant mutant protective antigen of Bacillus anthracis. Clin Vaccine Immunol 2012;19:140–5, http://dx.doi.org/10.1128/CVI. 05556-11.
[13] Campbell JD, Clement KH, Wasserman SS, Donegan S, Chrisley L, Kotloff KL. Safety, reactogenicity and immunogenicity of a recombinant protective antigen anthrax vaccine given to healthy adults. Hum Vaccin 2007;3:205–11. [14] Gorse GJ, Keitel W, Keyserling H, Taylor DN, Lock M, Alves K, et al. Immunogenicity and tolerance of ascending doses of a recombinant protective antigen (rPA102) anthrax vaccine: a randomized, double-blinded, controlled, multicenter trial. Vaccine 2006;24:5950–9, http://dx.doi.org/10.1016/j.vaccine.2006.05.044. [15] Brown BK, Cox J, Gillis A, VanCott TC, Marovich M, Milazzo M, et al. Phase i study of safety and immunogenicity of an Escherichia coli-derived recombinant protective antigen (rPA) vaccine to prevent anthrax in adults. PLoS ONE 2010;5:e13849, http://dx.doi.org/10.1371/journal.pone.0013849. [16] Reed MD, Wilder JA, Mega WM, Hutt JA, Kuehl PJ, Valderas MW, et al. Immunization with a recombinant, pseudomonas fluorescensexpressed, mutant form of Bacillus anthracis-derived protective antigen protects rabbits from anthrax infection. PLoS ONE 2015;10:e0130952, http://dx.doi.org/10.1371/journal.pone.0130952. [17] Farchaus JW, Ribot WJ, Jendrek S, Little SF, Fermentation. Purification, and characterization of protective antigen from a recombinant, avirulent strain of Bacillus anthracis. Appl Environ Microbiol 1998;64:982–91. [18] Wagner L, Verma A, Meade BD, Reiter K, Narum DL, Brady RA, et al. Structural and Immunological analysis of anthrax recombinant protective antigen adsorbed to aluminum hydroxide adjuvant. Clin Vaccine Immunol 2012;19:1465–73, http://dx.doi.org/10.1128/CVI. 00174-12. [19] D’Souza AJM, Mar KD, Huang J, Majumdar S, Ford BM, Dyas B, et al. Rapid deamidation of recombinant protective antigen when adsorbed on aluminum hydroxide gel correlates with reduced potency of vaccine. J Pharm Sci 2013;102:454–61, http://dx.doi.org/10.1002/jps.23422. [20] Reason DC, Ullal A, Liberato J, Sun J, Keitel W, Zhou J. Domain specificity of the human antibody response to Bacillus anthracis protective antigen. Vaccine 2008;26:4041–7, http://dx.doi.org/10.1016/j.vaccine.2008.05.023. [21] Reason D, Liberato J, Sun J, Keitel W, Zhou J. Frequency and domain specificity of toxin-neutralizing paratopes in the human antibody response to anthrax vaccine adsorbed. Infect Immun 2009;77:2030–5, http://dx.doi.org/10.1128/IAI. 01254-08. [22] Chi X, Li J, Liu W, Wang X, Yin K, Liu J, et al. Generation and characterization of human monoclonal antibodies targeting anthrax protective antigen following vaccination with a recombinant protective antigen vaccine. Clin Vaccine Immunol 2015;22:553–60, http://dx.doi.org/10.1128/CVI. 00792-14. [23] Oscherwitz J, Quinn CP, Cease KB. Anthrax vaccine recipients lack antibody against the loop neutralizing determinant: a protective neutralizing epitope from Bacillus anthracis protective antigen. Vaccine 2015;33:2342–6, http://dx.doi.org/10.1016/j.vaccine.2015.03.037. [24] Scobie HM, Rainey GJA, Bradley KA, Young JAT. Human capillary morphogenesis protein 2 functions as an anthrax toxin receptor. Proc Natl Acad Sci USA 2003;100:5170–4, http://dx.doi.org/10.1073/pnas.0431098100. [25] Bradley KA, Mogridge J, Mourez M, Collier RJ, Young JAT. Identification of the cellular receptor for anthrax toxin. Nature 2001;414:225–9, http://dx.doi.org/10.1038/n35101999. [26] Klimpel K, Molloy S, Thomas G, Leppla S. Anthrax toxin protective antigen is activated by a cell-surface protease with the sequence specificity and catalytic properties of furin. Proc Natl Acad Sci USA 1992;89:10277–81, http://dx.doi.org/10.1073/pnas.89.21.10277. [27] Milne J, Furlong D, Hanna P, Wall J, Collier R. Anthrax protective antigen forms oligomers during intoxication of mammalian-cells. J Biol Chem 1994;269:20607–12. [28] Reason D, Liberato J, Sun J, Camacho J, Zhou J. Mechanism of lethal toxin neutralization by a human monoclonal antibody specific for the PA(20) region of Bacillus anthracis protective antigen. Toxins 2011;3:979–90, http://dx.doi.org/10.3390/toxins3080979. [29] Crowe SR, Ash LL, Engler RJM, Ballard JD, Harley JB, Farris AD, et al. Select human anthrax protective antigen epitope-specific antibodies provide protection from lethal toxin challenge. J Infect Dis 2010;202:251–60, http://dx.doi.org/10.1086/653495. [30] Singh Y, Khanna H, Chopra AP, Mehra V. A dominant negative mutant of Bacillus anthracis protective antigen inhibits anthrax toxin action in vivo. J Biol Chem 2001;276:22090–4, http://dx.doi.org/10.1074/jbc.M010222200. [31] Singh Y, Chaudhary VK, Leppla SH. A deleted variant of Bacillus anthracis protective antigen is non-toxic and blocks anthrax toxin action in vivo. J Biol Chem 1989;264:19103–7. [32] Abboud N, De Jesus M, Nakouzi A, Cordero RJB, Pujato M, Fiser A, et al. Identification of linear epitopes in Bacillus anthracis protective antigen bound by neutralizing antibodies. J Biol Chem 2009;284:25077–86, http://dx.doi.org/10.1074/jbc.M109.022061. [33] Rivera J, Nakouzi A, Abboud N, Revskaya E, Goldman D, Collier RJ, et al. A monoclonal antibody to Bacillus anthracis protective antigen defines a neutralizing epitope in domain 1. Infect Immun 2006;74:4149–56, http://dx.doi.org/10.1128/IAI. 00150-06. [34] Cunningham K, Lacy DB, Mogridge J, Collier RJ. Mapping the lethal factor and edema factor binding sites on oligomeric anthrax protective antigen. Proc Natl Acad Sci USA 2002;99:7049–53, http://dx.doi.org/10.1073/pnas.062160399. [35] Mourez M, Yan M, Lacy DB, Dillon L, Bentsen L, Marpoe A, et al. Mapping dominant-negative mutations of anthrax protective antigen
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
by scanning mutagenesis. Proc Natl Acad Sci USA 2003;100:13803–8, http://dx.doi.org/10.1073/pnas.2436299100. Singh Y, Klimpel K, Arora N, Sharma M, Leppla S. The chymotrypsin-sensitive site, Ffd315, in anthrax toxin protective antigen is required for translocation of lethal factor. J Biol Chem 1994;269:29039–46. Benson EL, Huynh PD, Finkelstein A, Collier RJ. Identification of residues lining the anthrax protective antigen channel. Biochemistry (Mosc) 1998;37:3941–8, http://dx.doi.org/10.1021/bi972657b. Miller CJ, Elliott JL, Collier RJ. Anthrax protective antigen: preporeto-pore conversion. Biochemistry (Mosc) 1999;38:10432–41, http://dx.doi.org/10.1021/bi990792d. Santelli E, Bankston LA, Leppla SH, Liddington RC. Crystal structure of a complex between anthrax toxin and its host cell receptor. Nature 2004;430:905–8, http://dx.doi.org/10.1038/nature02763. Zhang J, Xu JJ, Li GL, Dong DY, Song XH, Guo Q, et al. The 2 beta(2)-2 beta(3) loop of anthrax protective antigen contains a dominant neutralizing epitope. Biochem Biophys Res Commun 2006;341:1164–71, http://dx.doi.org/10.1016/j.bbrc.2006.01.080. Gubbins MJ, Berry JD, Corbett CR, Mogridge J, Yuan XY, Schmidt L, et al. Production and characterization of neutralizing monoclonal antibodies that recognize an epitope in domain 2 of Bacillus anthracis protective antigen. Fems Immunol Med Microbiol 2006;47:436–43, http://dx.doi.org/10.1111/j.1574-695X.2006.00114.x. Yin Y, Zhang J, Dong D, Liu S, Guo Q, Song X, et al. Chimeric hepatitis B virus core particles carrying an epitope of anthrax protective antigen induce protective immunity against Bacillus anthracis. Vaccine 2008;26:5814–21, http://dx.doi.org/10.1016/j.vaccine.2008.08.031. Yin Y, Zhang S, Cai C, Zhang J, Dong D, Guo Q, et al. Deletion modification enhances anthrax specific immunity and protective efficacy of a hepatitis B core particle-based anthrax epitope vaccine. Immunobiology 2014;219:97–103, http://dx.doi.org/10.1016/j.imbio.2013.08.008. Oscherwitz J, Yu F, Jacobs JL, Liu T-H, Johnson PR, Cease KB. Synthetic peptide vaccine targeting a cryptic neutralizing epitope in domain 2 of Bacillus anthracis protective antigen. Infect Immun 2009;77:3380–8, http://dx.doi.org/10.1128/IAI. 00358-09. Oscherwitz J, Yu F, Cease KB. A synthetic peptide vaccine directed against the 2 beta 2-2 beta 3 loop of domain 2 of protective antigen protects rabbits from inhalation anthrax. J Immunol 2010;185:3661–8, http://dx.doi.org/10.4049/jimmunol.1001749. Sellman BR, Mourez M, Collier RJ. Dominant-negative mutants of a toxin subunit: an approach to therapy of anthrax. Science 2001;292:695–7, http://dx.doi.org/10.1126/science.109563.
7
[47] Yan M, Collier RJ. Characterization of dominant-negative forms of anthrax protective antigen. Mol Med 2003;9:46–51. [48] Brossier F, Levy M, Landier A, Lafaye P, Mock M. Functional analysis of Bacillus anthracis protective antigen by using neumonoclonal antibodies. Infect Immun 2004;72:6313–7, tralizing http://dx.doi.org/10.1128/IAI.72.11. 6313-6317.2004. [49] Mogridge J, Mourez M, Collier RJ. Involvement of domain 3 in oligomerization by the protective antigen moiety of anthrax toxin. J Bacteriol 2001;183:2111–6, http://dx.doi.org/10.1128/JB.183.6. 2111-2116.2001. [50] Lacy DB, Wigelsworth DJ, Melnyk RA, Harrison SC, Collier RJ. Structure of heptameric protective antigen bound to an anthrax toxin receptor: a role for receptor in pH-dependent pore formation. Proc Natl Acad Sci USA 2004;101:13147–51, http://dx.doi.org/10.1073/pnas.0405405101. [51] Brady RA, Verma A, Meade BD, Burns DL. Analysis of antibody responses to protective antigen-based anthrax vaccines through use of competitive assays. Clin Vaccine Immunol 2010;17:1390–7, http://dx.doi.org/10.1128/CVI. 00145-10. [52] Little SF, Novak JM, Lowe JR, Leppla SH, Singh Y, Klimpel KR, et al. Characterization of lethal factor binding and cell receptor binding domains of protective antigen of Bacillus anthracis using monoclonal antibodies. Microbiology-UK 1996;142:707–15. [53] Little S, Leppla S, Cora E. Production and characterization of monoclonalantibodies to the protective antigen component of Bacillus–Anthracis toxin. Infect Immun 1988;56:1807–13. [54] Kelly-Cirino CD, Mantis NJ. Neutralizing monoclonal antibodies directed against defined linear epitopes on domain 4 of anthrax protective antigen. Infect Immun 2009;77:4859–67, http://dx.doi.org/10.1128/IAI. 00117-09. [55] Varughese M, Teixeira AV, Liu SH, Leppla SH. Identification of a receptorbinding region within domain 4 of the protective antigen component of anthrax toxin. Infect Immun 1999;67:1860–5. [56] Rosovitz MJ, Schuck P, Varughese M, Chopra AP, Mehra V, Singh Y, et al. Alanine-scanning mutations in domain 4 of anthrax toxin protective antigen reveal residues important for binding to the cellular receptor and to a neutralizing monoclonal antibody. J Biol Chem 2003;278:30936–44, http://dx.doi.org/10.1074/jbc.M301154200. [57] Laffly E, Danjou L, Condemine F, Vidal D, Drouet E, Lefranc MP, et al. Selection of a macaque Fab with framework regions like those in humans, high affinity, and ability to neutralize the protective antigen (PA) of Bacillus anthracis by binding to the segment of PA between residues 686 and 694. Antimicrob Agents Chemother 2005;49:3414–20, http://dx.doi.org/10.1128/AAC.49.8. 3414-3420.2005.
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
ARTICLE IN PRESS
JVAC-17154; No. of Pages 7
Vaccine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Review
Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine Ryan C. McComb, Mikhail Martchenko ∗ Keck Graduate Institute, School of Applied Life Science, 535 Watson Dr., Claremont, CA, United States
a r t i c l e
i n f o
Article history: Received 7 August 2015 Received in revised form 5 October 2015 Accepted 9 November 2015 Available online xxx Keywords: Anthrax Bacillus anthracis Protective antigen Neutralizing antibody mapping Functional mapping Rational vaccine design
a b s t r a c t Anthrax is defined by the Centers for Disease Control and Prevention as a Category A pathogen for its potential use as a bioweapon. Current prevention treatments include Anthrax Vaccine Adsorbed (AVA). AVA is an undefined formulation of Bacillus anthracis culture supernatant adsorbed to aluminum hydroxide. It has an onerous vaccination schedule, is slow and cumbersome to produce and is slightly reactogenic. Next-generation vaccines are focused on producing recombinant forms of anthrax toxin in a well-defined formulation but these vaccines have been shown to lose potency as they are stored. In addition, studies have shown that a proportion of the antibody response against these vaccines is focused on non-functional, non-neutralizing regions of the anthrax toxin while some essential functional regions are shielded from eliciting an antibody response. Rational vaccinology is a developing field that focuses on designing vaccine antigens based on structural information provided by neutralizing antibody epitope mapping, crystal structure analysis, and functional mapping through amino acid mutations. This information provides an opportunity to design antigens that target only functionally important and conserved regions of a pathogen in order to make a more optimal vaccine product. This review provides an overview of the literature related to functional and neutralizing antibody epitope mapping of the Protective Antigen (PA) component of anthrax toxin. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction to rationally designed vaccines Developments in neutralizing antibody epitope mapping and structural determination of functional regions of pathogens and virulence factors are enhancing the capabilities for designing vaccine antigens that elicit targeted neutralizing antibody responses. This approach has been termed “rational” vaccination since the starting points for antigen design begin with thoroughly understanding the mechanism, by which virulence is blocked by the immune system [1,2] and is currently being explored for a number of intractable diseases such as HIV [3], Hepatitis C [4] and Influenza [5]. The strength in this approach lies in the fact that native antigens are not adapted for stable storage, optimal production, or elicitation of neutralizing antibodies to conserved functional regions leading to subsequent immunity [2]. Guided antigen design, by utilizing structural information and neutralizing antibody epitopes, provides an opportunity for developing potentially more effective
∗ Corresponding author. Tel.: +1 909 6070038. E-mail address: Mikhail [email protected] (M. Martchenko).
immunotherapeutics and vaccines that are better suited for production and storage in addition to being optimized for inducing a targeted immune response that may lead to long-lasting immunity. 1.1. Anthrax pathogenesis and biowarfare Anthrax is caused by the Gram-positive bacterium Bacillus anthracis, which has afflicted humans and their livestock for centuries. B. anthracis forms endospores and infection occurs when an animal inhales, ingests, or has contact with the spores through broken skin [6]. Spores germinate into vegetative bacteria and secrete the proteins Protective Antigen (PA), Edema Factor (EF) and Lethal Factor (LF). Anthrax PA, EF, and LF are the major pathogenesis factors of B. anthracis, as these toxins have been shown to inhibit phagocytosis process of myeloid cells. LF–PA causes host lethality through targeting the cardiomyocytes and vascular smooth-muscle cells, whereas EF–PA induces lethality mostly by targeting hepatocytes [7]. High mortality rates [6,8] require special actions for public health preparedness against an attack with B. anthracis [9]. Anthrax Vaccine Adsorbed (AVA) is the only vaccine approved to prevent anthrax infection. AVA consists of B. anthracis culture
http://dx.doi.org/10.1016/j.vaccine.2015.11.025 0264-410X/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
2
Fig. 1. Anthrax toxin cellular pathogenesis; PA83 or PA63 binds to cell surface receptors CMG2 and TEM8. PA83 undergoes cleavage by cellular furin releasing the 20 kD fragment. PA63 monomers bound to cellular receptors form an oligomer that is capable of binding three LF or EF molecules. Endocytosis of the PA63 oligomer bound to LF or EF, and acidification of the endosome, result in translocation of LF or EF to the cytosol where they act by cleaving Mitogen Activated Protein Kinase Kinases and Nlrp1 and causing cell death or converting ATP to cAMP and inhibiting phagocytosis.
supernatant adsorbed to aluminum hydroxide and requires a long initial immunization time frame of intramuscular injections at 0, 1 and 6 months. Vaccine recipients are considered protected after completing this initial series. Following the initial series, booster doses are administered at 12 and 18 months and annually thereafter [10]. This schedule is not ideal for use in emergency scenarios or for military personnel with inflexible deployment schedules. AVA consists of undefined components and causes local (e.g., erythema, swelling, pain or tenderness, itching, and nodules) and systemic (e.g., fever, chills, myalgia, arthralgia, and malaise) reactions in some individuals, which have fostered a perception that the vaccine is unsafe. In response to these shortcomings, a Department of Defense commissioned study in 2001 outlined goals for future anthrax vaccine development [11]. According to this report, an ideal anthrax vaccine would consist of defined components that elicit sufficient immunity within 30 days after 2–3 doses and protect against aerosolized anthrax for at least 1 year after initial immunization. Furthermore, this vaccine should remain stable for a long period of time and have minimal local and systemic adverse reactions. Finally, this vaccine should be easily scaled up to ensure product consistency [11]. 1.2. Next generation anthrax vaccines Most of the focus on next-generation anthrax vaccines has been on recombinantly produced PA (rPA) from various expression systems. Recombinant PA vaccines that are furthest along in development are those that are produced in prokaryotic systems such as avirulent, nontoxigenic and sporulation deficient strains of B. anthracis, Escherichia coli, or Pseudomonas fluorescens [12–16]. Several of these rPA vaccines have been evaluated in animal studies and human phase 1 clinical trials and show that rPA with Alhydrogel® adjuvant produces comparable Total Neutralizing Antibody titers and anti-PA IgG titers to AVA. Production of rPA in prokaryotic cells has several advantages including expression and secretion of authentically folded protein into the culture medium for easier purification. Also, the development of avirulent, non-toxigenic and
sporulation deficient strains of B. anthracis improves safety during manufacturing [17]. Although rPA vaccines are a step forward, limitations still exist. For example, rPA adsorbed to aluminum hydroxide adjuvant has been shown to lose its ability to induce toxin neutralizing antibodies as it is stored [18,19]. Furthermore, recent studies have also found that the human antibody response induced by AVA is biased toward non-neutralizing epitopes [20–22]. Also, important functional regions of PA have been shown to be shielded from the immune response induced by AVA [23]. These limitations provide justification for a rationally designed vaccine that targets functionally important and neutralizing epitopes on PA toxin, which can be optimized for stability, safety, production, and a reduced vaccine schedule. This article provides an overview of the critical regions of PA that might be targeted for production of such a vaccine. 1.3. Neutralizing antibody and functional epitope mapping of PA The PA molecule is an 83 kD protein comprised of four domains. After binding to cellular receptors [24,25], PA83 undergoes cleavage by a cellular furin to release a 20 kD fragment from domain I [26]. Furin cleavage exposes the ligand (LF/EF) binding sites and allows for receptor bound PA63 to oligomerize with other PA63 molecules (Fig. 1) [27]. Domain I is thus broken into PA20, which has no known function at this point in time, and Domain I , which remains attached to the PA63 molecule. Amino acids and sequences necessary for the function of PA have been discovered through amino acid mutagenesis and crystal structure determination (Fig. 2). In addition, monoclonal neutralizing antibodies have been used to elucidate vulnerable epitopes of PA (Fig. 3). These PA epitopes are located on all four domains of PA, and some of them, when blocked, are sufficient by themselves for toxin neutralization. 1.3.1. PA20 Although the PA20 region is not known to have a functionally important role in anthrax toxin pathogenesis, neutralizing antibodies targeting this domain have been identified. The monoclonal
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
3
Fig. 2. Amino acids associated with functionally important processes in cellular pathogenesis; Location and process of functionally important amino acids (A); location of residues described in table on PA63 bound to cellular receptor CMG2 (B) (Picture, PD:1T6B [39] visualized with Cn3D software); Residues not shown: 1–174 (PA20), 276–287, 304–319, 457–476, and 596–607.
antibody (mAb) 47F12, which was isolated from a human donor vaccinated with AVA, showed the ability to neutralize LF–PA (LT) in vitro [28]. Experiments with radiolabeled PA83 showed that 47F12 neutralizes through inhibition of furin cleavage. Analysis of 47F12 binding against a yeast library displaying randomly mutated PA20 on its surface found that binding was dependent on the fidelity of the amino acids corresponding to E95 , N98 , A100 , N104 , and I106 of PA20. In a separate study, a screen of AVA vaccinated donor serum tested against overlapping peptides of PA identified N102 –Q115 as an epitope, which bound 5/6 donor samples associated with high neutralization of LT as determined in an in vitro cellular assay with Raw264.7 macrophages [29]. This shows that
antibodies directed against this region contribute to neutralization, but it is unknown whether they are sufficient on their own for providing protection against anthrax toxin. The sequence R164 KKR167 that spans the interface between PA20 and PA63 is the furin cleavage site [30]. Variants of PA with residues 163–168 deleted are insensitive to furin cleavage and are non-toxic when administered with LF in cellular assays and in rats [31]. The mAb 7.5G was isolated from mice immunized with PA83 and recognizes the sequence between L156 and 170 S, which overlaps this site [32,33]. However, it is unclear from the data whether this antibody can inhibit furin cleavage [33]. When purified and administered to mice 24 h prior to LT challenge, it was found that 1 mg of 7.5G
Fig. 3. mAbs with defined epitopes that bind PA63; mAb epitope location, inhibited function, and neutralization capacity (A); Location of epitopes described in table on PA83 bound to cellular receptor CMG2 (B) (Picture, PD: 1T6B [39] visualized with Cn3D software); Residues not shown: 1–15, 159–174, 276–287, 302–319, 457–476, and 596–607.
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7 4
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
was needed to prolong survival of mice challenged with 100 g of LT by 3 days. When the peptide epitope of 7.5G was synthesized as a Multiple Antigenic Peptide (MAP) and administered to mice in Complete Freund’s Adjuvant (CFA), it was claimed that sera samples from vaccinated animals “moderately” protected J774.A macrophages from LT challenge [32]. Since the mice were not challenged with toxin, however, it is impossible to know if the antibody species and titers obtained post-vaccination were sufficient for in vivo neutralization. In another study, human antibodies directed against the sequence between N162 and N180 of PA were affinity purified from AVA vaccinated donors and tested for their ability to neutralize LT in vitro and in vivo [29]. This study showed that furin cleavage site directed antibodies protected 70% of Raw264.7 macrophages after treatment with LT but only 1 of 10 A/J mice survived challenged with 3 × LD50 LT after treatment with 30 g of affinity purified antibodies [29]. It appears that, although furin cleavage is an attractive target for focused therapies from a functional standpoint, there are limited levels of protection that can be achieved with antibodies directed against this epitope. 1.3.2. Domain I Domain I is the remainder of domain I after PA20 is cleaved by furin. A number of antibody epitopes have been mapped to this domain and correspond with functionally important amino acids sequences. For example, point mutations introduced into domain I showed that replacement of any of seven residues (R178 , K197 , R200 , P205 , I207 , I210 and K214 ) with alanine almost completely eliminated LF/EF binding to PA [34]. Analysis of the location of these seven residues in their 3-dimensional conformation on the crystal structure of the PA oligomer reveals that the LF/EF binding site spans across two PA molecules. Two murine antibodies, 19D9 and 20G7, bind the peptide sequence from V196 to I210 that include many of these vulnerable amino acids associated in LF/EF ligand binding [32]. Interestingly, 19D9 (IgG) is neutralizing at 1 g/ml against LT in cellular assays while 20G7 (IgM) is not neutralizing at concentrations up to 50 g/ml. The peptide sequence was synthesized as a MAP and used to immunize mice in the same manner as discussed above with the 7.5G peptide epitope. Neutralizing titers of 1:800 were observed from serum of vaccinated animals in a cellular LT cytotoxicity assay but no in vivo survival data was reported. Finally, Crowe and colleagues report that sera from AVA vaccinated human donors reacted with the peptide Y192 to P205 in a solid-phase epitope mapping experiment but no neutralization assays or toxin survival experiments in animals were performed [29]. These studies show that antibodies targeting the LF/EF binding site of PA are likely critical for toxin neutralization and support including this as a potential vaccine target. However, the fact that the LF/EF binding site spans two PA molecules in the oligomer may require an antigen mimic that closely resembles its native conformational structure. Another region with functional significance in domain I was found that included amino acids I210 , K225 , T240 , and K245 . Cysteine substitutions at any of these 4 amino acids resulted in the inhibition of PA activity by at least 100-fold [35]. Antibodies directed against the epitope P232 to V247 were found in 4 of 6 serum samples from AVA vaccinated human donors with high neutralizing serum titers [29]. Affinity purified antibodies directed against P232 to V247 were tested against LT and shown to provide 50% protection of Raw264.7 cells. Furthermore, 30 g of these affinity purified antibodies were shown to protect 3 of 10 mice challenged with 3 × LD50 LT [29]. Although this sequence is near the LF/EF binding site it is unclear what effect inhibition of this epitope has on the normal function of PA. The proximity of this sequence to the LF/EF binding site and its ability to partially protect cells and mice from LT warrants further investigation into its potential usefulness as a vaccine antigen target.
1.3.3. Domain II Neutralizing antibodies with epitopes that overlap known functional regions also exist in domain II of PA. The 22–23 loop (E302 to S325 ), which contains a chymotrypsin site (F313 to F314 ), has been shown to be involved in translocating LF/EF to the cytosol [36]. Crystal structure determination of PA and experimentation with cysteine substitutions along amino acids 302–325 provide strong evidence that 22–23 loops from individual PA molecules in the heptamer complex combine to form a 14-stranded transmembrane -barrel induced by the low-pH of the endosome [37–39]. Furthermore, when this loop was substituted with a homologous membrane-inserting loop from iota-b toxin of Clostridium perfringens and mixed with wild-type PA in equimolar ratios, a dominant-negative phenotype was observed, in which LT killing was completely inhibited in cellular studies and in rats [30]. These studies provide strong evidence that this region is necessary for LF/EF toxicity mediated through PA. Various studies report antibodies that bind PA at this location but differ in effectiveness for neutralizing LT. Murine monoclonal antibodies (5E12, 2A8 and 5E1) mapped to the chymotrypsin sequence (312–315) and were shown to inhibit chymotrypsin cleavage of PA into 47 kD and 37 kD fragments [40]. Although the neutralization capacity of antibodies 5E12, 2A8 and 5E1 is claimed, no percent survival values are reported from cellular assays. In another report, antibodies 2H9 and 16A12 mapped to the region between amino acids 312–326 and were not shown to be neutralizing [32]. However, Gubbins and colleagues report the discovery of three antibodies (F20G75, F20G76 and F20G77) that recognize the consensus sequence A311 SFFD315 within the 22–23 loop [41]. Cellular assays, in which dilutions of each antibody was incubated with LT toxin for 1 h at 37 ◦ C prior to addition onto J774.A macrophages, show that 90–100% protection was achieved for F20G75, F20G76 and F20G77 antibodies at concentrations as low as 12.5, 11.8 and 16.0 ng/ml, respectively. Interestingly, when the assay format was changed such that PA was allowed to incubate with cells prior to the addition of LF and antibodies, 90% protection was achieved only with higher concentrations (1–10 g/ml). This observation provides evidence that this epitope is inaccessible once PA has bound to the cellular membrane. Discordance between antibody neutralization capabilities targeting the 22–23 loop are possibly indicative of specific affinity/avidity, antibody class, or epitope fine tuning requirements that are necessary to achieve effective neutralization. Functional analysis and mAb neutralization studies give a rational basis behind targeting the 22–23 loop in a vaccine formulation. Based on this data, there have been reports of vaccines that target the epitope comprising amino acids G305 to S319 . The first attempt at targeting this epitope was performed by inserting this amino acid sequence onto the self-assembling Hepatitis B core protein scaffold [42,43]. Chimeric Hepatitis B core proteins carrying the 22–23 loop assembled to form virus like particles. Additionally, 50 g of virus particles administered without adjuvant was sufficient for protecting 4 of 7 guinea pigs challenged with B. anthracis spores (40 × LD50 ) compared to 2 of 3 animal survivors that received rPA vaccination with adjuvant [42]. The second attempt was made by creating a synthetic peptide constructed as a MAP on a lysine backbone. This vaccine was shown to induce protective antibodies in vitro and in vivo [44,45]. Seven New Zealand White Rabbits were immunized initially with 250 g of MAP comprising the 305–319 sequence from PA and a helper T-cell epitope from Plasmodium falciparum in CFA followed by four 125 g vaccine boosts in Incomplete Freund’s Adjuvant (IFA) at 2 week intervals. Vaccinated rabbits were challenged through the inhalational route with 200 × LD50 B. anthracis spores. Seven out of seven MAP-305319 vaccinated rabbits survived spore challenge compared to 7/7, 0/6 and 0/6 PA vaccinated, MAP control vaccinated or naïve control rabbits, respectively [45]. Antibodies targeting this region are
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
induced at low frequencies in humans, rabbits, and non-human primates by AVA or whole rPA vaccines [23]. The 22–23 loop has a necessary role in anthrax toxin pathogenesis and is an important target for next generation vaccines especially since current vaccines induce low antibody titers against this site. Mutational analysis and mAb neutralization has implicated a second region in domain II necessary for anthrax toxin pathogenesis. PA mutants with single amino acid substitutions in the region between amino acids 337 S to 458 N were found to reduce LT toxicity by at least 100-fold in a cellular toxicity assay [35]. More specifically, cysteine substitutions at I364 , T380 , S382 , T393 , N399 , Y411 , N422 and changes at D425 K, K397 D and F427 A exhibited a dominantnegative phenotype, in which PA containing any of these mutations mixed with wild-type PA inhibited death caused by LT in treated cells [35,46,47]. Further analysis showed that these PA mutants all retained the ability to bind cellular receptors, form heptamers, and bind LF but were deficient in endosomal pore formation and LF translocation to the cytosol [35]. Only one neutralizing mAb has mapped to the region between I364 and F427 and is the IgG1 mAb 48.3, which recognizes an epitope consisting of S412 to I419 [48]. Unexpectedly, the mechanism, by which 48.3 provides protection is through inhibition of the furin cleavage step of PA83. There is no obvious explanation for the ability of this antibody to inhibit furin cleavage based on its epitope location since analysis of the crystal structure shows that the furin cleavage site and amino acids 412–419 are not co-localized and are even located on opposite faces of the PA molecule. It could be that this site is required for furin recognition and stabilization for subsequent cleavage at R164 KKR167 site. Protective efficacy of mAb 48.3 was shown in mice challenged with a subcutaneous infection of B. anthracis spores. In this study, 6/8 and 2/8 mice treated with 100 g or 10 g, respectively, of antibody survived challenge of a 10 × LD50 dose of spores [48]. Further evidence for the importance of this region is found in the fact that human serum antibodies from 4/6 AVA vaccinated donors cross-reacted with peptides corresponding to amino acids 406–419 [29]. These studies suggest that amino acids located between I364 and F427 play a primary functional role in anthrax toxin pathogenesis and that antibody epitopes within this region are an important component of a protective immune response. This region, therefore, represents another potential vaccine target for epitope focused therapies. 1.3.4. Domain III From a functional standpoint, domain III appears to be involved with PA oligomer formation. Insertion of missense point mutations within domain III at amino acids D512 , L514 , and D520 blocked detection of PA63 oligomer assembly suggesting that these residues are important in this step anthrax toxin pathogenesis [49]. In a separate study, a point mutation at E515 inhibited the activity of PA by 100-fold in cellular toxicity assays [35]. From the crystal structure it is apparent that this region of PA forms a loop that is in close contact with amino acids P173 to A258 of domain I in a neighboring PA monomer stabilizing the oligomer complex [50]. Although functionally important, no neutralizing mAbs have been reported that bind this region of PA. It has been determined that antibodies directed against domain III contribute to neutralization of LT in vitro. However, none of these antibodies appear to bind regions, which are correlated with functional importance as described previously. For example, recombinantly expressed and purified domain III of PA reduces the ability of serum from PA vaccinated rabbits, non-human primates and humans to neutralize native PA [51]. This study did not identify epitopes on domain III, which were associated with this competitive inhibitory effect. In another study, antibodies from mouse hybridomas (2D3, 2D5 and 10D2) were discovered that bind between I581 and N601 , which spans the end of domain III and
5
the beginning of domain IV [52]. Interestingly, the data suggests that the mechanism, by which 2D3, 2D5 and 10D2 confer protection seems to be by inhibition of LF binding to PA suggesting that this region interacts closely with domain I . Although these mAbs showed neutralization of LT in J774.A cellular assay, they were not capable of protecting Fisher 344 Rats from death. Rats challenged with 13 × LD50 LT incubated with up to 1.7 mg IgG from ascites for 1 h only delayed death by 18 h (2D3 and 2D5) or 0 h (10D2 and IgG negative control) [52,53]. These studies show that mAbs targeting discrete epitopes on domain III do contribute to a neutralizing response against LT but are not sufficient on their own to neutralize LT in vivo. This fact is supported by a number of other studies that report mAbs that bind domain III but on their own are incapable of neutralizing LT. For example, antibody 2-A7, that was found to bind amino acids G532 to Q543 , did not protect mice against LT challenge [54]. Also, Fisher 344 Rats challenged with LT and administered domain III binding antibody 8A7 at a 1:1 (antibody: PA) molar ratio also all died [22]. However, when 9 g of 8A7 was administered in combination with 9 g of 2A6 (an antibody whose epitope is undetermined but also was unable to protect Fisher 344 Rats against LT challenge on its own) 5/5 rats survived a lethal dose of anthrax toxin. This data supports the conclusion that domain III antibodies may need to work in concert with antibodies against other regions on PA in order to provide sufficient protection against anthrax toxin. 1.3.5. Domain IV Domain IV contains important functional sequences for epitope targeted therapeutics. Numerous studies have shown that antibodies directed solely against domain IV are sufficient for neutralizing anthrax toxin. Crystal structure analysis of PA bound to cellular receptor shows that this interaction is primarily attributed to amino acids found in three loops comprising amino acid regions E654 –M662 , Y681 –Y688 , and E712 –G714 of domain IV and one loop from domain II [39,50]. Consistent with the prediction that these amino acids have important functional activity due to their role in binding cellular receptors is the observation that cysteine substitutions at residues I656 , N657 , I665 , N682 , D683 , and L687 inhibit PA activity by at least 100-fold [35]. An additional study shows that mutations in the loop containing amino acids K679 to N693 were most detrimental to PA toxicity, whereas mutations between E704 to K722 allowed PA to retain its toxicity [55]. Interestingly, it was found that an alanine substitution at D683 had the most significant effect (1000-fold reduction) on PA toxicity and binding [56]. Some of the most potent mAbs for neutralizing PA have the ability to block binding to cellular receptors. Neutralizing murine antibodies 3B6 (926 g) and 14B7 (23 g) protected 4/4 Fisher 344 Rats challenged with LT [53]. These antibodies were subsequently shown to bind between amino acids D671 and I721 [52]. Further studies with alanine substitutions at specific residues in the PA binding domain showed that K684 , L685 , L687 and Y688 were critical for 14B7 recognition and neutralization [56]. A high affinity (KD = 3.4 nM), highly neutralizing (50% inhibitory concentration, 5.6 nM) Fab isolated from an immunized macaque designated as 35PA83 was found to also bind between P686 and Y694 [57]. Correlation between PA neutralization with antibodies 14B7 and 35PA83 with mutagenesis studies show that the receptor binding loop between K679 to N693 is an important target for anthrax toxin countermeasures. Two additional murine antibodies were found to bind regions of domain IV that inhibit PA binding to receptors and show both in vitro and in vivo neutralization efficacy [54]. Monoclonal antibodies 1-F1 and 2-B12 bind to linear epitopes located between P692 –K703 and T716 –F727 , respectively. High affinities (1-F1, 1.7 nM and 2-B12, 2.0 nM) and in vitro neutralization of LT were reported. Additionally, 550 g/mouse of 1-F1 or 2-B12 protected 2/5 and 3/5 BALB/c mice challenged with a lethal dose of LT, respectively. Given
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
6
the close proximity of these epitopes to the functionally essential loop between K679 and N693 it is not surprising that these antibodies have some ability to inhibit anthrax toxin. It appears though, that these antibodies are not as effective at blocking toxin entry as 14B7 and 35PA83 that recognized the loop between amino acids 679 and 693. Another epitope associated with domain IV was found to exist between L628 and K637 [29]. Antibodies against this peptide sequence were affinity purified from AVA vaccinated human serum and a 30 g/mouse dose was found to protect 6/10 mice from 3 × LD50 LT challenge. The specific mode of inhibition for these antibodies was not determined. Domain IV is a crucial target for anthrax toxin therapies based on PA functional studies, crystal structure analysis, and the potency, by which antibodies targeting this region protect cells and animals against anthrax toxin. 2. Conclusion In summary, PA contains many functional regions that can be targeted for rationally designed therapeutics and vaccines. These regions come from all four domains of PA and include antibody epitopes and functional regions that, if blocked, are sufficient by themselves for anthrax toxin neutralization. On the other hand, PA contains many more partially neutralizing epitopes and functional sites of inhibition. These studies suggest that blocking multiple regions of the PA toxin simultaneously will likely provide more complete protection. A rationally designed vaccine will, more likely than not, need to include a variety of PA antigen targets to induce complete immunity against anthrax and any potential PA mutants that can evade a particular targeted vaccine. Conflict of interest statement The authors report no conflict of interests. References [1] Burton DR. Antibodies, viruses and vaccines. Nat Rev Immunol 2002;2:706–13, http://dx.doi.org/10.1038/nri891. [2] Dormitzer PR, Ulmer JB, Rappuoli R. Structure-based antigen design: a strategy for next generation vaccines. Trends Biotechnol 2008;26:659–67, http://dx.doi.org/10.1016/j.tibtech.2008.08.002. [3] Jardine J, Julien J-P, Menis S, Ota T, Kalyuzhniy O, McGuire A, et al. Rational HIV immunogen design to target specific germline B cell receptors. Science 2013;340:711–6, http://dx.doi.org/10.1126/science.1234150. [4] Kong L, Jackson KN, Wilson IA, Law M. Capitalizing on knowledge of hepatitis C virus neutralizing epitopes for rational vaccine design. Curr Opin Virol 2015;11:148–57, http://dx.doi.org/10.1016/j.coviro.2015.04.001. [5] Kanekiyo M, Wei C-J, Yassine HM, McTamney PM, Boyington JC, Whittle JRR, et al. Self-assembling influenza nanoparticle vaccines elicit broadly neutralizing H1N1 antibodies. Nature 2013;499:102–6, http://dx.doi.org/10.1038/nature12202. [6] Adalja AA, Toner E, Inglesby TV. Clinical management of potential bioterrorism-related conditions. N Engl J Med 2015;372:954–62, http://dx.doi.org/10.1056/NEJMra1409755. [7] Liu S, Moayeri M, Leppla SH. Anthrax lethal and edema toxins in anthrax pathogenesis. Trends Microbiol 2014;22:317–25, http://dx.doi.org/10.1016/j.tim.2014.02.012. [8] Riedel S. Anthrax: a continuing concern in the era of bioterrorism. Proc Bayl Univ Med Cent 2005;18:234–43. [9] Office of Public Health Preparedness and Response. First hours: bioterrorism agents. Cent Dis Control Prev 2013. http://emergency.cdc.gov/ firsthours/bioterrorism.asp (accessed June 9, 2015). [10] Emergent BioSolutions. BioThrax (Anthrax Vaccine Adsorbed) Package Insert; 2012. [11] Committee to Assess the Safety and Efficacy of the Anthrax Vaccine. In: Joellenbeck LM, Zwanziger LL, Durch JS, Strom BL, et al., editors. The anthrax vaccine: Is it safe? Does it work? Washington, DC: National Academy Press; 2002. [12] Bellanti JA, Lin F-YC, Chu C, Shiloach J, Leppla SH, Benavides GA, et al. Phase 1 study of a recombinant mutant protective antigen of Bacillus anthracis. Clin Vaccine Immunol 2012;19:140–5, http://dx.doi.org/10.1128/CVI. 05556-11.
[13] Campbell JD, Clement KH, Wasserman SS, Donegan S, Chrisley L, Kotloff KL. Safety, reactogenicity and immunogenicity of a recombinant protective antigen anthrax vaccine given to healthy adults. Hum Vaccin 2007;3:205–11. [14] Gorse GJ, Keitel W, Keyserling H, Taylor DN, Lock M, Alves K, et al. Immunogenicity and tolerance of ascending doses of a recombinant protective antigen (rPA102) anthrax vaccine: a randomized, double-blinded, controlled, multicenter trial. Vaccine 2006;24:5950–9, http://dx.doi.org/10.1016/j.vaccine.2006.05.044. [15] Brown BK, Cox J, Gillis A, VanCott TC, Marovich M, Milazzo M, et al. Phase i study of safety and immunogenicity of an Escherichia coli-derived recombinant protective antigen (rPA) vaccine to prevent anthrax in adults. PLoS ONE 2010;5:e13849, http://dx.doi.org/10.1371/journal.pone.0013849. [16] Reed MD, Wilder JA, Mega WM, Hutt JA, Kuehl PJ, Valderas MW, et al. Immunization with a recombinant, pseudomonas fluorescensexpressed, mutant form of Bacillus anthracis-derived protective antigen protects rabbits from anthrax infection. PLoS ONE 2015;10:e0130952, http://dx.doi.org/10.1371/journal.pone.0130952. [17] Farchaus JW, Ribot WJ, Jendrek S, Little SF, Fermentation. Purification, and characterization of protective antigen from a recombinant, avirulent strain of Bacillus anthracis. Appl Environ Microbiol 1998;64:982–91. [18] Wagner L, Verma A, Meade BD, Reiter K, Narum DL, Brady RA, et al. Structural and Immunological analysis of anthrax recombinant protective antigen adsorbed to aluminum hydroxide adjuvant. Clin Vaccine Immunol 2012;19:1465–73, http://dx.doi.org/10.1128/CVI. 00174-12. [19] D’Souza AJM, Mar KD, Huang J, Majumdar S, Ford BM, Dyas B, et al. Rapid deamidation of recombinant protective antigen when adsorbed on aluminum hydroxide gel correlates with reduced potency of vaccine. J Pharm Sci 2013;102:454–61, http://dx.doi.org/10.1002/jps.23422. [20] Reason DC, Ullal A, Liberato J, Sun J, Keitel W, Zhou J. Domain specificity of the human antibody response to Bacillus anthracis protective antigen. Vaccine 2008;26:4041–7, http://dx.doi.org/10.1016/j.vaccine.2008.05.023. [21] Reason D, Liberato J, Sun J, Keitel W, Zhou J. Frequency and domain specificity of toxin-neutralizing paratopes in the human antibody response to anthrax vaccine adsorbed. Infect Immun 2009;77:2030–5, http://dx.doi.org/10.1128/IAI. 01254-08. [22] Chi X, Li J, Liu W, Wang X, Yin K, Liu J, et al. Generation and characterization of human monoclonal antibodies targeting anthrax protective antigen following vaccination with a recombinant protective antigen vaccine. Clin Vaccine Immunol 2015;22:553–60, http://dx.doi.org/10.1128/CVI. 00792-14. [23] Oscherwitz J, Quinn CP, Cease KB. Anthrax vaccine recipients lack antibody against the loop neutralizing determinant: a protective neutralizing epitope from Bacillus anthracis protective antigen. Vaccine 2015;33:2342–6, http://dx.doi.org/10.1016/j.vaccine.2015.03.037. [24] Scobie HM, Rainey GJA, Bradley KA, Young JAT. Human capillary morphogenesis protein 2 functions as an anthrax toxin receptor. Proc Natl Acad Sci USA 2003;100:5170–4, http://dx.doi.org/10.1073/pnas.0431098100. [25] Bradley KA, Mogridge J, Mourez M, Collier RJ, Young JAT. Identification of the cellular receptor for anthrax toxin. Nature 2001;414:225–9, http://dx.doi.org/10.1038/n35101999. [26] Klimpel K, Molloy S, Thomas G, Leppla S. Anthrax toxin protective antigen is activated by a cell-surface protease with the sequence specificity and catalytic properties of furin. Proc Natl Acad Sci USA 1992;89:10277–81, http://dx.doi.org/10.1073/pnas.89.21.10277. [27] Milne J, Furlong D, Hanna P, Wall J, Collier R. Anthrax protective antigen forms oligomers during intoxication of mammalian-cells. J Biol Chem 1994;269:20607–12. [28] Reason D, Liberato J, Sun J, Camacho J, Zhou J. Mechanism of lethal toxin neutralization by a human monoclonal antibody specific for the PA(20) region of Bacillus anthracis protective antigen. Toxins 2011;3:979–90, http://dx.doi.org/10.3390/toxins3080979. [29] Crowe SR, Ash LL, Engler RJM, Ballard JD, Harley JB, Farris AD, et al. Select human anthrax protective antigen epitope-specific antibodies provide protection from lethal toxin challenge. J Infect Dis 2010;202:251–60, http://dx.doi.org/10.1086/653495. [30] Singh Y, Khanna H, Chopra AP, Mehra V. A dominant negative mutant of Bacillus anthracis protective antigen inhibits anthrax toxin action in vivo. J Biol Chem 2001;276:22090–4, http://dx.doi.org/10.1074/jbc.M010222200. [31] Singh Y, Chaudhary VK, Leppla SH. A deleted variant of Bacillus anthracis protective antigen is non-toxic and blocks anthrax toxin action in vivo. J Biol Chem 1989;264:19103–7. [32] Abboud N, De Jesus M, Nakouzi A, Cordero RJB, Pujato M, Fiser A, et al. Identification of linear epitopes in Bacillus anthracis protective antigen bound by neutralizing antibodies. J Biol Chem 2009;284:25077–86, http://dx.doi.org/10.1074/jbc.M109.022061. [33] Rivera J, Nakouzi A, Abboud N, Revskaya E, Goldman D, Collier RJ, et al. A monoclonal antibody to Bacillus anthracis protective antigen defines a neutralizing epitope in domain 1. Infect Immun 2006;74:4149–56, http://dx.doi.org/10.1128/IAI. 00150-06. [34] Cunningham K, Lacy DB, Mogridge J, Collier RJ. Mapping the lethal factor and edema factor binding sites on oligomeric anthrax protective antigen. Proc Natl Acad Sci USA 2002;99:7049–53, http://dx.doi.org/10.1073/pnas.062160399. [35] Mourez M, Yan M, Lacy DB, Dillon L, Bentsen L, Marpoe A, et al. Mapping dominant-negative mutations of anthrax protective antigen
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025
G Model JVAC-17154; No. of Pages 7
ARTICLE IN PRESS R.C. McComb, M. Martchenko / Vaccine xxx (2015) xxx–xxx
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
by scanning mutagenesis. Proc Natl Acad Sci USA 2003;100:13803–8, http://dx.doi.org/10.1073/pnas.2436299100. Singh Y, Klimpel K, Arora N, Sharma M, Leppla S. The chymotrypsin-sensitive site, Ffd315, in anthrax toxin protective antigen is required for translocation of lethal factor. J Biol Chem 1994;269:29039–46. Benson EL, Huynh PD, Finkelstein A, Collier RJ. Identification of residues lining the anthrax protective antigen channel. Biochemistry (Mosc) 1998;37:3941–8, http://dx.doi.org/10.1021/bi972657b. Miller CJ, Elliott JL, Collier RJ. Anthrax protective antigen: preporeto-pore conversion. Biochemistry (Mosc) 1999;38:10432–41, http://dx.doi.org/10.1021/bi990792d. Santelli E, Bankston LA, Leppla SH, Liddington RC. Crystal structure of a complex between anthrax toxin and its host cell receptor. Nature 2004;430:905–8, http://dx.doi.org/10.1038/nature02763. Zhang J, Xu JJ, Li GL, Dong DY, Song XH, Guo Q, et al. The 2 beta(2)-2 beta(3) loop of anthrax protective antigen contains a dominant neutralizing epitope. Biochem Biophys Res Commun 2006;341:1164–71, http://dx.doi.org/10.1016/j.bbrc.2006.01.080. Gubbins MJ, Berry JD, Corbett CR, Mogridge J, Yuan XY, Schmidt L, et al. Production and characterization of neutralizing monoclonal antibodies that recognize an epitope in domain 2 of Bacillus anthracis protective antigen. Fems Immunol Med Microbiol 2006;47:436–43, http://dx.doi.org/10.1111/j.1574-695X.2006.00114.x. Yin Y, Zhang J, Dong D, Liu S, Guo Q, Song X, et al. Chimeric hepatitis B virus core particles carrying an epitope of anthrax protective antigen induce protective immunity against Bacillus anthracis. Vaccine 2008;26:5814–21, http://dx.doi.org/10.1016/j.vaccine.2008.08.031. Yin Y, Zhang S, Cai C, Zhang J, Dong D, Guo Q, et al. Deletion modification enhances anthrax specific immunity and protective efficacy of a hepatitis B core particle-based anthrax epitope vaccine. Immunobiology 2014;219:97–103, http://dx.doi.org/10.1016/j.imbio.2013.08.008. Oscherwitz J, Yu F, Jacobs JL, Liu T-H, Johnson PR, Cease KB. Synthetic peptide vaccine targeting a cryptic neutralizing epitope in domain 2 of Bacillus anthracis protective antigen. Infect Immun 2009;77:3380–8, http://dx.doi.org/10.1128/IAI. 00358-09. Oscherwitz J, Yu F, Cease KB. A synthetic peptide vaccine directed against the 2 beta 2-2 beta 3 loop of domain 2 of protective antigen protects rabbits from inhalation anthrax. J Immunol 2010;185:3661–8, http://dx.doi.org/10.4049/jimmunol.1001749. Sellman BR, Mourez M, Collier RJ. Dominant-negative mutants of a toxin subunit: an approach to therapy of anthrax. Science 2001;292:695–7, http://dx.doi.org/10.1126/science.109563.
7
[47] Yan M, Collier RJ. Characterization of dominant-negative forms of anthrax protective antigen. Mol Med 2003;9:46–51. [48] Brossier F, Levy M, Landier A, Lafaye P, Mock M. Functional analysis of Bacillus anthracis protective antigen by using neumonoclonal antibodies. Infect Immun 2004;72:6313–7, tralizing http://dx.doi.org/10.1128/IAI.72.11. 6313-6317.2004. [49] Mogridge J, Mourez M, Collier RJ. Involvement of domain 3 in oligomerization by the protective antigen moiety of anthrax toxin. J Bacteriol 2001;183:2111–6, http://dx.doi.org/10.1128/JB.183.6. 2111-2116.2001. [50] Lacy DB, Wigelsworth DJ, Melnyk RA, Harrison SC, Collier RJ. Structure of heptameric protective antigen bound to an anthrax toxin receptor: a role for receptor in pH-dependent pore formation. Proc Natl Acad Sci USA 2004;101:13147–51, http://dx.doi.org/10.1073/pnas.0405405101. [51] Brady RA, Verma A, Meade BD, Burns DL. Analysis of antibody responses to protective antigen-based anthrax vaccines through use of competitive assays. Clin Vaccine Immunol 2010;17:1390–7, http://dx.doi.org/10.1128/CVI. 00145-10. [52] Little SF, Novak JM, Lowe JR, Leppla SH, Singh Y, Klimpel KR, et al. Characterization of lethal factor binding and cell receptor binding domains of protective antigen of Bacillus anthracis using monoclonal antibodies. Microbiology-UK 1996;142:707–15. [53] Little S, Leppla S, Cora E. Production and characterization of monoclonalantibodies to the protective antigen component of Bacillus–Anthracis toxin. Infect Immun 1988;56:1807–13. [54] Kelly-Cirino CD, Mantis NJ. Neutralizing monoclonal antibodies directed against defined linear epitopes on domain 4 of anthrax protective antigen. Infect Immun 2009;77:4859–67, http://dx.doi.org/10.1128/IAI. 00117-09. [55] Varughese M, Teixeira AV, Liu SH, Leppla SH. Identification of a receptorbinding region within domain 4 of the protective antigen component of anthrax toxin. Infect Immun 1999;67:1860–5. [56] Rosovitz MJ, Schuck P, Varughese M, Chopra AP, Mehra V, Singh Y, et al. Alanine-scanning mutations in domain 4 of anthrax toxin protective antigen reveal residues important for binding to the cellular receptor and to a neutralizing monoclonal antibody. J Biol Chem 2003;278:30936–44, http://dx.doi.org/10.1074/jbc.M301154200. [57] Laffly E, Danjou L, Condemine F, Vidal D, Drouet E, Lefranc MP, et al. Selection of a macaque Fab with framework regions like those in humans, high affinity, and ability to neutralize the protective antigen (PA) of Bacillus anthracis by binding to the segment of PA between residues 686 and 694. Antimicrob Agents Chemother 2005;49:3414–20, http://dx.doi.org/10.1128/AAC.49.8. 3414-3420.2005.
Please cite this article in press as: McComb RC, Martchenko M. Neutralizing antibody and functional mapping of Bacillus anthracis protective antigen—The first step toward a rationally designed anthrax vaccine. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.11.025