Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) 1

ARTICLE Mucosal and systemic immune responses induced by a single time vaccination strategy in mice

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE UNIV on 07/05/15 For personal use only.

Elizabeth González Aznar, Belkis Romeu, Miriam Lastre, Caridad Zayas, Maribel Cuello, Osmir Cabrera, Yolanda Valdez, Mildrey Fariñas, and Oliver Pérez

Abstract: Vaccination is considered by the World Health Organization as the most cost-effective strategy for controlling infectious diseases. In spite of great successes with vaccines, many infectious diseases are still leading killers, because of the inadequate coverage of many vaccines. Several factors have been responsible: number of doses, high vaccine reactogenicity, vaccine costs, vaccination policy, among others. Contradictorily, few vaccines are of single dose and even less of mucosal administration. However, more common infections occur via mucosa, where secretory immunoglobulin A plays an essential role. As an alternative, we proposed a novel protocol of vaccination called Single Time Vaccination Strategy (SinTimVaS) by immunizing 2 priming doses at the same time: one by mucosal route and the other by parenteral route. Here, the mucosal and systemic responses induced by Finlay adjuvants (AF Proteoliposome 1 and AF Cochleate 1) implementing SinTimVaS in BALB/c mice were evaluated. One intranasal dose of AF Cochleate 1 and an intramuscular dose of AF Proteoliposome 1 adsorbed onto aluminum hydroxide, with bovine serum albumin or tetanus toxoid as model antigens, administrated at the same time, induced potent specific mucosal and systemic immune responses. Also, we demonstrated that SinTimVaS using other mucosal routes like oral and sublingual, in combination with the subcutaneous route elicits immune responses. SinTimVaS, as a new immunization strategy, could increase vaccination coverage and reduce time–cost vaccines campaigns, adding the benefits of immune response in mucosa. Key words: adjuvants, mucosal immunization, vaccination strategy, cochleates, outer membrane vesicles. Résumé : L’organisation mondiale de la santé considère que la vaccination est la stratégie la plus économiquement efficiente pour lutter contre les maladies infectieuses. En dépit des succès incontestables de la vaccination, plusieurs maladies infectieuses demeurent des causes majeures de mortalité en raison d’une couverture inadéquate. De multiples facteurs sont mis en cause, dont le nombre de doses, la forte réactogéniocité des vaccins, leurs coûts, et les politiques de vaccination. Contrairement a` la logique, peu de vaccins sont offerts en dose unique et encore moins par voie mucosale. Or, les infections les plus répandues passent par les muqueuses, la` où les immunoglobulines A sécrétoires jouent un rôle essentiel. À titre de solution de rechange, nous proposons un nouveau protocole vaccinal appelé « stratégie de vaccination en un temps » (Single Time Vaccination Strategy, SinTimVaS) fondé sur une immunisation avec deux doses d’initiation simultanées, l’une par la voie mucosale, l’autre par la voie parentérale. Dans la présente étude, on a évalué chez des souris BALB/c les réponses mucosales et systémiques aux adjuvants de Finlay (AF Proteoliposome 1 et AF Cochleate 1) administrés dans le cadre d’une SinTimVaS. Une dose intranasale d’AF Cochleate 1 et une dose intramusculaire d’AF Proteoliposome 1 adsorbés sur de l’hydroxyde d’aluminium, adjoints a` de la BSA ou de la TT comme antigènes modèles, le tout administré simultanément, ont induit de vigoureuses réponses immunitaires mucosales et systémiques spécifiques. En outre, nous avons démontré que la SinTimVaS empruntant d’autres voies, comme les voie orale et sublinguale jumelées a` la voir sous-cutanée, parvenait a` déclencher des réponses immunitaires. La nouvelle stratégie d’immunisation SinTimVas promet de hausser la couverture vaccinale tout en réduisant les coûts et le temps requis pour les campagnes de vaccinations, avantages auxquels s’ajoute la faculté d’activer l’immunité mucosale. [Traduit par la Rédaction] Mots-clés : adjuvants, immunisation mucosale, stratégie de vaccination, cochléaires, vésicules de membrane externe.

Introduction Vaccination is considered by the World Health Organization (WHO) as the most cost-effective strategy for controlling infectious diseases, preventing 2.5 million deaths every year (WHO 2005). Despite the great success of vaccination, many infectious diseases are still leading killers, mainly due to the insufficient coverage of vaccination (Drain et al. 2003; Burton et al. 2009). A number of factors have been responsible for the logistical chal-

lenges to attain high immunization coverage: required doses for immunization schedules, parenteral immunization, high vaccine reactogenicity, few studies targeting antigen combinations in vaccine formulations, vaccine cold chain, vaccination policy, weak immunization programmes for countries, high cost and lack of affordability for poor countries, among others (Jiang et al. 2005). Only live attenuated vaccines like BCG (Bacillus Calmette–Guérin) and polio induce good protection with a single dose. In contrast,

Received 30 January 2015. Revision received 29 April 2015. Accepted 14 May 2015. E. González Aznar,* C. Zayas, M. Cuello, and O. Cabrera. Immunology Department, Finlay Institute, P.O. Box 16017, Havana, Cuba. B. Romeu.* Immunology Department, Finlay Institute, P.O. Box 16017, Havana, Cuba; Permanent Mission of Cuba, Geneva, Switzerland. M. Lastre and O. Pérez. Havana Medical School, Havana, Cuba. Y. Valdez and M. Fariñas. Immunology Department, Finlay Institute, P.O. Box 16017, Havana, Cuba; Animal Models Direction, Finlay Institute, P.O. Box 16017, Havana, Cuba. Corresponding authors: Elizabeth González Aznar (e-mail: elygonzalez@finlay.edu.cu) and Belkis Romeu (e-mail: [email protected]). *These two authors contributed equally. Can. J. Microbiol. 61: 1–8 (2015) dx.doi.org/10.1139/cjm-2015-0063

Published at www.nrcresearchpress.com/cjm on 19 May 2015.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE UNIV on 07/05/15 For personal use only.

2

to obtain good immune response with a single dose by nonliving antigen vaccines, multiple doses are generally needed, to provide sufficient stimulation of the immune system and to achieve longlasting immune responses. The number of doses is especially relevant in developing countries, where the population does not have easy access to medical services and vaccination campaigns are very difficult, much more so when several doses are required, which can cause missed doses over the entire immunization schedule (Aguado and Lambert 1992). So, the development of new immunization strategies that allow a complete immunization scheme with a single visit without further contact could be as important as the development of single-dose vaccines. Traditionally, adjuvants have been used to enhance the magnitude of the adaptive response to a vaccine. However, a second role has become increasingly important: guiding the type of adaptive response to produce the most effective forms of immunity for each specific pathogen (Th1, Th2, Th17, Th9) (Coffman et al. 2010). Adjuvants have been classified as immunopotentiators, delivery systems (O’Hagan and Singh 2003), and immunopolarizors (Pérez et al. 2013), and neither of these activities are mutually excluding, some adjuvants even exhibit at least 2 of these activities. The use of potent adjuvants that exhibit several mechanisms of action and enable comparable immune responses using substantially lower antigen amounts and doses could be very important to develop novel vaccination strategies. AFPL1 (Adjuvant Finlay Proteoliposome 1), a nanoparticle derived from the outer membrane of Neisseria meningitidis serogroup B, has been widely used as the main antigen of the Cuban meningococcal vaccine administered in more than 60 million doses in humans (Campa et al. 1997; Pérez et al. 2001). AFCo1 (Adjuvant Finlay Cochleate 1) is a microparticle with cochlear structure obtained from the interaction of an anionic lipid from AFPL1 with divalent cations like Ca2+ (Pérez et al. 2004). Both Finlay adjuvants contain native lipopolysaccharide (LPS), major outer membrane proteins, phospholipids, traces of bacterial DNA, synergistic MAMPs that interact with TLR-4, TLR-2, and TLR-9, respectively, as immunopotentiator molecules. Also, they induce the activation of different T-cells subsets, with a preferential Th1 pattern (immunopolarized actions) and show delivery system capacities based on the lipids presented in the structures. Their potential as adjuvants has been tested with a variety of antigens: Plasmodium falciparum merozoite surface proteins 4 and 5 (Bracho et al. 2009), Herpes Simplex Virus type 2 (HSV-2) protein (Del Campo et al. 2010), Streptococcus pyogenes proteins (Guilherme et al. 2009), and purified Dermatophagoides siboney allergens (Lastre et al. 2006). Both structures also induce crosspresentation and cytotoxic T-lymphocyte activity (Rodríguez et al. 2005, 2006). AFCo1 can be used mainly as a mucosal adjuvant because it is a stable particle with higher mucosal absorption (Del Campo et al. 2009) and acid-resistant properties. The mucosal surface is exposed to a diverse and very large number of microorganisms and is also the site of infection of diverse pathogens (e.g., influenza virus, Vibrio cholerae, type 1 reovirus, and rhinovirus) (Plotkin 2005). The initial mucosal infection can only be prevented by mucosal immune responses (pathogen-specific secretory immunoglobulin A (SIgA) or local cytotoxic T lymphocytes) (Brandtzaeg 2009). Mucosal immunization elicits systemic specific immune response and stimulates the specific mucosal immune response, leading by the production of SIgA antibody. Also, it is a less invasive route, does not require trained personnel, and significantly decreases the risk of parenteral transmitted diseases (Neutra and Kozlowski 2006; Mann et al. 2009). In our experience, the mucosal route requires larger antigen amounts and more doses than the parenteral route to obtain similar systemic responses (Del Campo et al. 2009). In addition, parenteral immunization has been shown to induce memory IgA in different vaccines, such as influenza, Pseudomonas aeruginosa, and inactivated polio vaccines (attributed to previous natural mucosal infection

Can. J. Microbiol. Vol. 61, 2015

by these organisms) (Kaul and Ogra 1998). Indeed, a combined mucosal and parenteral poliovirus immunization strategy has been shown to effectively induce neutralizing antibodies in serum and mucosal immune responses in humans. Strategies that combine parenteral and mucosal immunizations (Treanor et al. 1992; Morrison et al. 1998), such as mucosal prime + parenteral booster (Faden et al. 1990; Lee et al. 1999) or parenteral prime + mucosal booster (Keren et al. 1988), have generated stronger immune responses than either of these routes by themselves. In the light of the ambitious goal of producing vaccines that provide maximum protection at both the parenteral and mucosal levels with a minimum number of doses, we set out to develop a novel vaccination strategy called the Single Time Vaccination Strategy (SinTimVaS) (Pérez et al. 2008; González et al. 2009). SinTimVaS combines 2 priming doses (1 mucosal and 1 parenteral) at the same time without a subsequent boost. In this work, we assessed the ability of SinTimVaS to induce an efficient systemic and mucosal immune response in mice against bovine serum albumin (BSA) protein and tetanus toxoid (TT) by different mucosal and parenteral immunizations routes. We showed that mice immunized with 1 AFCo1 intranasal (i.n.) dose and 1 AFPL1 intramuscular (i.m.) dose at the same time combined with BSA or TT antigens triggered a specific serum IgG and mucosal IgA immune response against these specific antigens. Furthermore, we demonstrated that the new strategy also works for oral (o.r.), sublingual (s.l.), and subcutaneous (s.c.) routes combinations.

Materials and methods Animals The experiments were assessed using female BALB/c mice (age: 8–10 weeks old; mass: 16–18 g) from the National Center for the Production of Laboratory Animals, Havana, Cuba. Mice were randomly distributed into 5 mice per experimental group or control group. All mice were kept in ventilated cages, watered ad libitum, under specific conditions at the Finlay Institute Animal Research Facilities. All animal studies were carried out following the guidelines for the use of laboratory animals (EEC Council Directive, 1987) and were approved by the Ethics Committee for Laboratory Animals in Havana, Cuba. Adjuvants and antigens TT and AFPL1, of the epidemic meningococcal strain Cu385 (B: 4:P1.19,15), were produced at industrial scale under Good Manufacturing Practices conditions by manufacturing plants at the Finlay Institute, Havana, Cuba. AFPL1 was adsorbed onto aluminum hydroxide. AFCo1 was obtained in our laboratory, starting from proteoliposome (PL), through a process of exchange of sodium deoxycholate and the divalent cation calcium, as described by Pérez et al. 2004. The formation of the cochleate structure was visualized by the appearance of a white suspension in the preparation and by light microscopy analysis. BSA protein was supplied by Sigma-Aldrich. Immunization schedules Groups of 5 female mice were immunized simultaneously by i.m. injection of AFPL1 (12.5 ␮g) or by mucosal (i.n., s.l., or o.r.) administration of AFCo1 (50 ␮g). Also, groups of mice were immunized 2 times (days 0 and 14) i.m. with AFPL1 (12.5 ␮g) and 3 times (days 0, 7, and 14) i.n. (12.5 ␮L/nostril) with AFCo1 (50 ␮g). With heterologous antigens, mice were immunized using the same scheme described above. Each dose contained BSA (10 ␮g per mouse per dose) plus AFPL1 (12.5 ␮g) or TT (2 Lf (limit to flocculation) per dose per mouse) plus AFPL1 (12.5 ␮g) for i.m. or s.c. immunizations. For i.n., s.l., and o.r. immunizations, each dose contained BSA (20 ␮g per dose per mouse) or TT (5 Lf per dose per mouse) plus AFCo1 (50 ␮g). The volume delivered in each dose was 12.5 ␮L/nostril for i.n., 10 ␮L under the tongue for s.l., and 100 ␮L for o.r. routes. Groups of animals receiving saline were used as Published by NRC Research Press

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) González Aznar et al.

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE UNIV on 07/05/15 For personal use only.

experimental control for immunization. Table 1 shows the different SinTimVaS combinations used. Samples collection Samples were collected as follows: saliva and faeces were collected on the 7th day after the last dose, serum and vaginal washes on the 21st day after the last dose. Saliva samples were taken following salivation stimulation with 50 ␮L of intraperitoneal injection of pilocarpine (0.5% (v/v), Quimefa, Cuba). Saliva was inactivated for 15 min at 56 °C, centrifuged at 10 000g for 10 min, and stored until it was used. Three to 6 pieces of freshly voided faeces were collected into 1.5 mL preweighed microcentrifuge tubes and phosphate-buffered saline (PBS, pH 7.2) was added with protease inhibitors (1 mmol/L phenylmethyl sulfonylfluoride in ethanol; 5 ␮g/mL aprotinin; and 1 ␮L/mL leupeptin, antipain, and pepstatin (all from Sigma)) in a ratio of 20 ␮L/mg of faeces. Solid matter was resuspended by extensive vortex and centrifuged at 10 000g for 10 min, and the supernatants were stored. Vaginal cavities were rinsed 3 times with 50 ␮L of PBS, and the wash was centrifuged and immediately stored. Blood was collected in microhematocrit capillary tubes following puncturing of the retroorbital plexus. To remove the serum, the blood was removed at 5000g for 10 min. All samples were stored at –20 °C until enzymelinked immunosorbent assays (ELISAs) were conducted. Analysis of immune response by ELISA Anti-PL-, anti-BSA-, or anti-TT-specific IgA, IgG, IgG1, and IgG2a antibodies were measured. Briefly, 96-well ELISA plates (Nunc MaxiSorp F96; Roskilde, Denmark) were coated with (per well) 100 ␮L of PL (20 ␮g/mL), BSA (10 ␮g/mL), or TT (5 Lf/mL) in carbonate– bicarbonate buffer (pH 9.6) and incubated overnight at 4 °C. After blocking with PBS – 1% gelatin (blocking solution) and incubating at room temperature for 1 h, plates were washed with PBS containing 0.05% Tween 20. Then, samples were diluted in blocking solution as follows: 1:2 for saliva and faeces, 1:10 for vaginal washes, and 1:100 for sera, adding 100 ␮L/well and incubating for 2 h at 37 °C. The plates were incubated at 37 °C for 2 h with the different conjugates: anti-mouse IgA biotinylated antibody (4 ␮g/mL, Sigma, USA), anti-mouse IgG peroxidase-conjugated antibodies (1:5000, Sigma, USA), and anti-mouse IgG1 and IgG2a biotinylated antibody (1:2500, Sigma, USA) for IgG isotypes. Streptavidin peroxidase conjugate (1:2000 dilution, Sigma, USA) was added and incubated for 2 h at 37 °C. Each plate was developed with o-phenylenediamine (Sigma, USA) in citrate–phosphate buffer (pH 5.0), and the reaction was stopped with 2 mol/L H2SO4. Optical densities (OD) were read at 492 nm, and the results were expressed as OD. Anti-PL IgG and IgA titers were expressed as arbitrary units per millilitre (AU/mL), referring to an in-house standard. The cutoff value was established as the mean OD492 plus 2 times the standard deviation (SD) of the control group. In all plates, positive and negative sera were included to validate the assay. Statistical analysis Data were analyzed using the GraphPAD prism software (San Diego, California, USA). The statistical significance of the difference between group means was analyzed by a Student’s 2-tailed t test for 2 groups, or by 1-factor analysis of variance (ANOVA) followed by Tukey’s test for 3 or more groups. Differences were considered significant with p < 0.05.

Results SinTimVaS applied by a combination of i.n. and i.m. routes induces anti-PL mucosal and systemic immune responses To determine if SinTimVaS (i.n.+i.m.) might induce systemic and mucosal immune responses, we compared this novel strategy (SinTimVaS) with the traditional immunization schedule of 3 i.n. and 2 i.m. doses. Groups of BALB/c mice were primed mucosally (i.n.) and systemically (i.m.) at the same time without a subsequent

3

Table 1. Single time vaccination (SinTimVaS) combinations used in the studies. SinTimVaS combination AFCo1 + BSA or TT

AFPL1 + BSA or TT

i.n. s.l. o.r.

i.m. i.m. i.m.

Note: Two primings at the same time, without subsequent boost. Adjuvants and antigens concentrations used in SinTimVaS were twice those used in traditional parenteral and mucosal routes. AFCo1, AF cochleate 1; AFPL1, AF proteoliposome 1; BSA, bovine serum albumin; TT, tetanus toxoid; i.n., intranasal; i.m., intramuscular; s.l., sublingual; o.r., oral.

booster. As AFCo1 and AFPL1 have conserved in their structure several protective outer membrane proteins, we can determine the immune responses against these proteins. Twenty-one days after the last dose, the levels of systemic and mucosal immunity were evaluated by determining serum anti-PL IgG and IgA responses. The serum anti-PL IgG induced by SinTimVaS was similar to that induced by 3 i.n. doses of AFCo1 or 2 i.m. doses of AFPL1 (Fig. 1A), maintaining the same behavior in the subclasses IgG1 and IgG2a (Fig. 1B). Thus, the observed levels of IgG2a in mice immunized by different SinTimVaS combinations suggest a Th1 polarization (Fig. 2B), as is the case of the immune response of mice immunized with 3 i.n. or 2 i.m. doses. In saliva, anti-PL IgA was significant higher in SinTimVaS (p < 0.001) than in 2 i.n. doses of AFCo1 or 2 i.m. doses of AFPL1. In addition, neither 1 i.n. dose nor 2 i.m. doses induced specific IgA (Fig. 1C). Overall, SinTimVaS overcomes the need for several required doses for mucosal immune response induction when it is supplemented with a simultaneous parenteral dose. SinTimVaS works with other mucosal routes To evaluate if SinTimVaS can be extended to other mucosa routes, we administered s.l.+i.m. and o.r.+i.m. combinations and compared the results with that of traditional immunization schedules. SinTimVaS combinations of s.l.+i.m. or o.r.+i.m. priming induced a significantly higher (p < 0.01) systemic immune response than 3 s.l. and 3 o.r. doses of AFCo1 (Fig. 2A). We also tested specific IgG subclasses in the sera of immunized mice and found that IgG1 and IgG2a were induced in all SinTimVaS combinations (Fig. 2B). In addition, the presence of high levels of IgG2a in sera constitutes direct evidence of Th1. The anti-PL IgA response was detected in all mucosal samples (saliva, faeces, and vagina). Both SinTimVaS combinations s.l.+i.m. and o.r.+i.m. induced similar mucosal specific IgA in all mucosal samples when we compared with 3 s.l. and 3 o.r. doses of AFCo1. Nevertheless, the antiPL-specific IgA response induced by the i.n. route was higher than that of other mucosal routes (Fig. 2C). Lastly, mucosal specific IgA was induced not only at the regional site of immunization but also at distal places like the vagina and intestine when i.n. and s.l. routes were used in SinTimVaS combinations. These results confirmed that SinTimVaS works even when other mucosal routes of immunization were used. Mucosal and systemic immune response induced by SinTimVaS with s.c. route as parenteral priming To determine whether the s.c. route as parenteral priming in SinTimVaS induces systemic and mucosal immune responses, we compared different combinations of mucosal (i.n., s.l., o.r.) routes plus s.c. route with 3 mucosal doses of AFCo1 and 2 s.c. doses of AFPL1 (Fig. 3). At the systemic or mucosal levels no differences were observed when s.c. priming was used. Also, all SinTimVaS Published by NRC Research Press

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

combinations were capable of inducing high levels of IgG1 and IgG2a (Fig. 3B). These results demonstrated that this novel strategy of immunization also works with the s.c. route as parenteral priming. Mucosal and systemic immune responses induced by SinTimVaS when AFPL1 and AFCo1 were used as adjuvants against BSA and TT To address if Finlay adjuvants (AFPL1 to parenteral routes, and AFCo1 to mucosal routes) are able to elicit an immune response against 2 model antigens (BSA and TT) using the proposed strategy, mucosal and systemic anti-BSA and anti-TT immune responses were examined. Figures 4A and 4B show anti-BSA-specific IgG in sera and anti-BSA-specific IgA in saliva, respectively. The SinTimVaS i.n.+i.m. combination induced an anti-BSA IgG response superior (p < 0.05) to that induced by the s.l.+i.m. or o.r.+i.m. combinations (Fig. 4A). Also, the responses at the systemic level for all SinTimVaS combinations were significantly superior (p < 0.05) to that of 3 doses of each mucosal route. Figure 4B shows that we detected a slight increase in anti-BSA IgA response in mice immunized with the o.r.+i.m. combination when compared with each only-mucosal scheme. Figure 4C shows anti-TT systemic response induced by SinTimVaS combinations. The antiTT-specific IgG in sera of o.r.+i.m. and s.l.+i.m. combinations was stimulated in comparison with antibodies levels induced by 3 o.r. or 3 s.l. doses. Moreover, the anti-TT IgG response induced by the i.n.+i.m. combination was similar to those induced by 3 i.n. doses of AFCo1-TT or 2 i.m. doses of AFPL1-TT. However, in saliva anti-TT response of the SinTimVaS combinations was not superior to the response induced by 3 doses of AFCo1-TT by different routes (Fig. 4D), but it was significantly higher (p < 0.05) than the IgA detected in mice immunized with 3 doses of TT alone (data not show). These results indicate that Finlay adjuvants in SinTimVaS are highly effective in eliciting the production of both mucosal and systemic immune responses against heterologous antigens.

p < 0.001

Anti PL IgG (U/mL)

5000

A

p < 0.001

4000 3000 2000 1000 0

Doses

Anti PL IgG subclasses (OD 492 nm)

Fig. 1. Proteoliposome (PL)-specific antibody responses induced by Single Time Vaccination Strategy (STVS). Groups of female BALB/c mice (n = 5) were STVS (intranasal+intramuscular (i.n./i.m.)) immunized with Adjuvant Finlay Proteoliposome 1 (AFPL1) (i.m. route) and Adjuvant Finlay Cochleate 1 (AFCo1) (i.n. route). Included were control groups immunized with 1, 2, or 3 doses of AFPL1 (i.m.) or AFCo1 (i.n.). The PL-specific immunoglobulin A (IgA), IgG, and IgG subtype levels were determined using a PL-specific ELISA. Panel A shows the anti-PL IgG levels in sera, panel B shows the anti-PL IgG subclasses in sera, and panel C shows the anti-PL IgA in saliva. Data are expressed as the mean arbitrary units (AU) + standard deviation from 3 independent experiments. Data from subclasses evaluation are expressed as the mean optical density (OD) of samples measured at 492 nm + standard error of the mean. Significant differences between the means of different groups were determined by a Tukey’s multiple comparison test using the Graph Pad Prism 4 software (California). A p value of

Mucosal and systemic immune responses induced by a single time vaccination strategy in mice.

Vaccination is considered by the World Health Organization as the most cost-effective strategy for controlling infectious diseases. In spite of great ...
813KB Sizes 0 Downloads 8 Views