Vaccine 32 (2014) 290–296

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

Immunogenicity of highly conserved recombinant VacJ outer membrane lipoprotein of Pasteurella multocida Sathish Bhadravati Shivachandra a,∗ , Abhinendra Kumar a , Revanaiah Yogisharadhya a , K.N. Viswas b a b

Clinical Bacteriology Laboratory, Indian Veterinary Research Institute (IVRI), Mukteswar-263138, Nainital (District), Uttarakhand (UK), India Division of Bacteriology and Mycology, Indian Veterinary Research Institute (IVRI), Izatnagar 243122, Uttar Pradesh (UP), India

a r t i c l e

i n f o

Article history: Received 30 June 2013 Received in revised form 12 September 2013 Accepted 23 October 2013 Keywords: Immunogenecity Mouse model Outer membrane protein Pasteurella multocida Protective efficacy Recombinant VacJ lipoprotein

a b s t r a c t Bacterial lipoproteins are emerging targets for inducing protective immunity against many infectious diseases. VacJ is a highly conserved and widely distributed outer membrane lipoprotein of Pasteurella multocida strains, which are known to affect a wide range of domestic as well as wild animals and birds. In the present study, the gene encoding for mature VacJ outer membrane lipoprotein of P. multocida serogroup B:2 strain P52 was cloned and over-expressed in Escherichia coli as a fusion protein. The purified recombinant VacJ protein (∼44 kDa) was used for immunizing mice (6/group) along with adjuvants (FCA and alum) in two experiments. Immunization of mice with rVacJ (30 ␮g and 75 ␮g/mice) elicited humoral immune response with significant (P < 0.01) rise in antigen-specific titers of IgG and its subtypes (IgG1 and IgG2a). No protection was noticed in mice immunized with rVacJ (30 ␮g) along with FCA followed by challenge with 100 LD50 of the homologous strain. On the contrary, higher rVacJ dose (75 ␮g) along with FCA and alum provided 66.7% and 50% protection respectively, at reduced challenge dose (8 LD50 ). The study indicated that a lipidated recombinant VacJ lipoprotein with suitable adjuvants could potentially act as candidate antigen for vaccine development against pasteurellosis in livestock. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Pasteurella multocida, a Gram-negative bacterial pathogen, is the causative agent of several diseases including haemorrhagic septicaemia (HS) in cattle and buffaloes, septicaemic/pneumonic pasteurellosis in sheep and goats, fowl cholera in birds, atrophic rhinitis in pigs and snuffles in rabbits [1–3]. Dog/cat bite-induced P. multocida infection in humans have also been reported [4]. Infections caused by capsular (A, B, D, E, and F) and somatic serogroups (1–16) of P. multocida are collectively called ‘pasteurellosis’, which is characterized by high morbidity/mortality in livestock and the associated economic losses [3,5,6]. Conventional vaccines for pasteurellosis have been used for many decades. However, these vaccines are inefficient in inducing long lasting, cross-protective immunity [3,7]. Recently, several outer membrane proteins have been proposed as candidate antigens following sequence analysis and evaluation of immunogenicity/protective efficacy [3,8–12].

∗ Corresponding author. Tel.: +91 05942 286348; fax: +91 05942 286347; mobile: +91 9756002533. E-mail addresses: [email protected], [email protected] (S.B. Shivachandra). 0264-410X/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.vaccine.2013.10.075

Lipoproteins abound in the outer membranes of Gram-negative bacteria and they are considered potential target antigens for vaccine development. Lipoprotein predictor programmes (Lipo and LipoP v 1.0) have been used to predict the presence of 86 and 82 lipoproteins in the genomes of P. multocida avian strain Pm70 and porcine strain 3480 respectively [13]. Among the predicted lipoproteins in the genomes of P. multocida strains, 13–17 and 7–15 were unique/specific to the avian and porcine strains respectively. Till now, only few P. multocida lipoproteins such as plpB [14], plpE [15,16], plp-40 [17], pcP and GlpQ [18], Omp16 [19,20] have been characterized. Many others, including PM0442, PM0659, PM0979, PM1050, PM1064, PM1614, PM1720, PM1578, PM0576, and PM1501 have been identified as putative lipoproteins [9,21]. Therefore, there is a need to understand the immunogenicity of each of these putative lipoproteins in order to select subunit vaccine candidate antigen for pasteurellosis. P. multocida strain Pm70 gene PM1501 has been predicted to encode a homolog of VacJ (virulence-associated chromosome locus J), which is a surface lipoprotein of Shigella flexneri [22] and Hemophilus influenzae [9]. Until now, no attempts were made to characterize and analyze the effect of VacJ on immune system of host. In the present study, for the first time, we describe the immunogenicity and protective efficacy of P. multocida recombinant VacJ using mouse model.

S.B. Shivachandra et al. / Vaccine 32 (2014) 290–296

291

2. Materials and methods

2.4. Mice immunization trials and evaluation of immune response

2.1. Bacteria, vector, primers and mice

A total of thirty healthy Swiss albino mice (6–8 week old) were pre-bled to ascertain their P. multocida antibody-free status. Following bulk purification and quantification of rVacJ protein, adjuvants were incorporated in vaccine formulations. Two independent experiments were carried out as described below:

P. multocida serogroup B:2 strain P52 (an Indian HS vaccine strain) was obtained from the ‘Clinical Bacteriology Laboratory’, Indian Veterinary Research Institute (IVRI), Mukteswar, Uttarakhand, India. For construction and expression clones, pET32a vector (Novagen, USA), Escherichia coli TOP10 and Origami(DE3)pLacI cells were used. The required primers were synthesized and procured (IDT-DNA, USA). Swiss albino mice (6–8 week old), reared in pathogen-free environment and maintained at laboratory animal section, IVRI, Mukteswar, Uttarakhand, were used in immunization trials.

2.2. VacJ sequence analysis and construction of pVacJ clone The vacJ gene from P. multocida B:2 strain P52 was used for prediction of matured VacJ protein characteristics using the PROTEAN program (DNASTAR), PSIPRED, LipoP v1.0 server as well as Protoparam tools from the ExPASy website. For construction of pVacJ clone, a primer set targeting the vacJ gene sequence (Nucleotide region – Nt 61–75 and Nt 720–738) from P. multocida B:2 strain P52 encoding mature full-length VacJ (21 A-D246 aa) without signal peptide sequence and Cysteine residue (1 M-C20 aa) was designed using the reference sequence available at GenBank (Acc # JX184899). The sequences of the oligonucleotides with added restriction sites (underlined) for BamHI and XhoI at 5 end of forward and reverse primer along with tags (small letter) respectively, were as below; VACF: 5 -cgcGGATCCATGGCGACTATTGATTCA-3 and VACR: 5 gtgCTCGAGATCAATTTCATTTAAGATA-3 Chromosomal DNA was purified from P. multocida B:2 strain P52 using standard procedures and was used as template (50 ng) for amplification of vacJ gene. The PCR mixture consisted of 50 pmol of each primer along with other reagents as described earlier [8]. PCR reaction included 30 cycles of denaturation at 94 ◦ C for 45 s, annealing at 55 ◦ C for 45 s, extension at 72 ◦ C for 45 s and a final extension at 72 ◦ C for 3 min. The PCR amplified product (50 ␮g) was digested with BamHI and XhoI; and ligated in to pET32a vector. E. coli TOP10 and Origami(DE3)pLacI cells successfully transformed with pET32a(vacJ) were selected using ampicillin (50 mg/ml) and chloramphenicol (35 mg/ml).

2.3. Over-expression, purification and Western blot of rVacJ E. coli Origami(DE3)pLacI cells harboring recombinant plasmid (pVacJ) were grown at 37 ◦ C in 1 l LB broth with appropriate antibiotics and induced with 1 mM IPTG. After 3 h post induction, harvested cell pellet was solubilized in a denaturing buffer and the protein was purified under denaturing condition by affinity chromatography using Ni-NTA superflow cartridges (Qiagen, USA) as per the method described previously [11]. The concentrated rVacJ protein was quantified by NanoDrop 2000 Spectrophotometer (Thermo Scientific, USA) and aliquots were stored at −80 ◦ C. For confirmation, purified rVacJ protein was transferred onto nitrocellulose membrane using semidry immunoblot system (Amersham pharmacia, USA) and detected using anti-P. multocida B:2 polyclonal hyperimmune rabbit serum and goat anti-rabbit IgG HRPO conjugate (Sigma, USA) as primary and secondary antibodies respectively. The blots were incubated in substrate solution (10 mg DAB) for color development.

(1) Experiment #1: Two groups (1a and 1c) each consisting of 6 mice were formed. Each mouse in a Group-1a was inoculated with rVacJ protein (30 ␮g/dose) in a 100 ␮l volume mixed with Freund’s complete adjuvant (FCA) by subcutaneous route. After 21 days, a booster dose with rVacJ protein (30 ␮g/mice) was given to all the immunized mice using Freund’s incomplete adjuvant (FIA). (2) Experiment #2: Three groups (2a, 2b and 2c) with each consisting of 6 mice were formed. Each mouse in a Group-2a and 2b were inoculated with rVacJ protein (75 ␮g/dose) in a 100 ␮l volume mixed with FCA and alum, respectively by subcutaneous route. After 21 days, a booster dose with rVacJ protein (75 ␮g/mice) was given to all the immunized mice using adjuvants FIA and alum respectively. A control group of 6 mice in each experiment (Group-1c and 2c) was injected with PBS alone. All the mice were provided with feed and water ad libitum and monitored regularly. For evaluation of immune response, all the immunized and control mice were bled on 0, 21 and 42 days post-immunization. The sera of each mice from different groups were assayed for the presence of antigen specific IgG and its subtypes (IgG1 and IgG2a ) using an indirect-ELISA as per the standard protocol [23]. Briefly, 96-well flat-bottomed Nunc Maxisorp plates (M/s Nalgene Nunc, Denmark) were coated with rVacJ antigen (100 ng/well) in 0.1 M carbonate/bicarbonate buffer, pH 9.6 (100 ␮l/well) and incubated overnight at 4 ◦ C. Following washing and blocking, dilutions of serum antibodies (1:400) in PBS-SMP were added to wells and incubated for 1 h. Upon washing, plates were added with anti-mouse IgG (1:10,000)/IgG1 (1:8000)/IgG2a (1:8000) horseradish peroxidase (HRP)-conjugated secondary antibodies (M/s Serotec, USA) and substrate OPD (Sigma, USA) before readings at 492 nm on a Sunrise-360063 (M/s Tecan, Austria) plate reader. All the laboratory animal experiments were conducted according to the norms of ‘Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA)’, Government of India, Ministry of Environment and Forests, Animal Welfare Division and approval by the Institute Animal Ethics Committee (IAEC) as well as Institute Biosafety Committee (IBSC), IVRI, Mukteswar, Uttarakhand, India. 2.5. Challenge studies and evaluation of protective efficacy For challenge studies, P. multocida B:2 strain P52 passaged once in mice and the LD50 calculated as per standard procedure [24] was used. On 56th day post immunization, each immunized mice and those in control groups were challenged with 0.2 ml of culture containing 100 LD50 for Expt #1 and 8 LD50 for Expt #2 by subcutaneous route. All the dead mice were subjected to post-mortem examination and heart blood was collected aseptically. Bacterial confirmation was done by conventional method (bacterial isolation, staining, and biochemical tests) and molecular assays (PM and HSB specific PCR assays) as described previously [6,25]. The protective efficacy was determined by mouse survival up to 14 days post challenge. Mortality in challenged mice was expressed as percent survival and mean survival time (MST) as per the previous method [23] and the percentage survival curve was plotted.

292

S.B. Shivachandra et al. / Vaccine 32 (2014) 290–296

Fig. 1. Predicted characteristics of VacJ lipoprotein of P. multocida. Panel A: (a) Schematic of full length VacJ lipoprotein of P. multocida B:2 strain P52 with signal peptide (red colored tube), (b) hydrophilicity plot as per Kyte-Doolittle, (c) antigenic index as per Jameson-Wolf and (d) surface probability plot as per Emini. Panel B: VacJ lipoprotein sequence with predicted secondary structures. The secondary structures were predicted by PSIPRED of ExPASy. The blue colored aa region denotes signal peptide at Nterminus. Blue colored arrow denotes starting of mature VacJ lipoprotein. Yellow colored tubes denote regions of ␣-helices. ␤-strands are denoted with green colored arrows. Lipobox predicted by LipoP v1.0 is indicated by red colored box. The vertical downward arrow denotes the site of cleavage before lipidation. Abbreviations: N, amino terminus; C, carboxyl terminus; SecStr, secondary structure; aa #, amino acids number; ␣, helix region; ␤, strand region. Numerical denotes sequential number of helix/strand regions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

For statistical analysis, all the data were expressed as mean ± standard error (SE). For determining the significance of the observations, Student’s t-test was carried out using SPSS 16.1 software. P values ≤0.05 were considered statistically significant.

B) was cloned in pET32a vector and recombinant clones were confirmed by colony PCR.

3. Results

Induction of recombinant cells resulted in over-expression of rVacJ protein which accounted for a total molecular weight of approximately 44.2 kDa (399 aa), of which ∼25.4 kDa (226 aa) was VacJ, and ∼18.8 kDa (173 aa) was the coding region of pET32a vector including the two hexa-histidine tags (Fig. 2, Panel C, Lane 2). Solubility analysis following cell lysis indicated the partitioning of over-expressed rVacJ into the insoluble fraction of the lysate. Furthermore, rVacJ was purified under denaturing conditions and renatured on column before elution and dialysis. This protein migrated as a single band at ∼44 kDa during 10% SDS-PAGE (Fig. 2, Panel C, Lanes 3 and 4). The rVacJ protein was also detected by an immunoblot assay using anti P. multocida B:2 strain P52 antibodies (Fig. 2, Panel C, Lane 5).

3.1. VacJ sequence analysis and construction of pVacJ clone The VacJ lipoprotein of P. multocida B:2 strain P52 is encoded by a 741 bp ORF with 39.7% GC. Full length VacJ protein (246 aa, MW ∼27.56 kDa) was predicted to contain a signal peptide (1–19 aa) using PSIPRED (Fig. 1, Panel A). VacJ also had a very high antigenic index, hydrophilicity and surface probability as shown by PROTEAN (Fig. 1, Panel B). Furthermore, full length VacJ was found to contain a total of 29 basic, 28 acidic, 84 hydrophobic and 65 polar residues with an isoelectric point of 8.017. LipoP v1.0 server predicted a lipobox [L-S-G-C] with a cleavage site between 19 G and 20 C aa within VacJ (Fig. 1, Panel B). Predicted secondary structure of VacJ revealed a predominant eight helix region (␣1–␣8) and three minor strands (␤1–␤3) as shown in Fig. 1, Panel B. The expression construct was designed in such that the vacJ region was fused between hexa-histidine tags on both termini (Fig. 2, Panel A). The PCR amplified product, ∼699 bp (Fig. 2, Panel

3.2. Expression, purification and Western blot analysis of rVacJ

3.3. Immunogenecity of rVacJ The titers of serum antibodies against rVacJ, total IgG at 21 and 42 days; and IgG1 and IgG2a subtypes at 42 days post immunization are indicated in Fig. 3, Panel A and B. Serum antibody titers

S.B. Shivachandra et al. / Vaccine 32 (2014) 290–296

293

Fig. 2. Schematic of VacJ lipoprotein construct, expression, purification and detection. Panel A: Mature VacJ lipoprotein with both N- and C-terminus hexa-histidine tag. Panel B: PCR amplification of vacJ gene from P. multocida B:2 strain P52. Lane M: DNA standard ladder, Lane 1: Amplified vacJ gene product (∼699 bp), Lane 2: Negative control. Panel C. Over-expression, purification and Western blot of rVacJ lipoprotein. Lanes-M: Protein standard marker, Lane 1: Uninduced E. coli cell lysate without 1mM IPTG induction, Lane 2: Induced E. coli cell lysate indicating expressed rVacJ (∼44 kDa) protein, Lane 3 and 4: Purified rVacJ lipoprotein fractions, Lane M: Protein standard marker, Lane 5: Immunoblot of lane 4 indicating the brown color development following detection using polyclonal anti-P. multocida B:2 rabbit antibodies, goat anti-rabbit IgG HRP conjugate and DAB substrate in chromogenic reaction observed on nitrocellulose membrane.

Fig. 3. Immunogenecity of rVacJ lipoprotein in mouse model. Panel A: The bar diagram indicating the rVacJ specific serum antibody titers (total IgG) of sera collected from all the mice belonging to Expt #1 and #2 at 0, 21 and 42 days post immunization. Panel B: The bar diagram indicating the serum antibody (subtype IgG1 and IgG2a ) titers of sera collected from all the mice belonging to Expt #1 and #2 at 42 days post immunization. The rVacJ specific antibody titres mice belonging to both experiments were measured by indirect ELISA and recorded at OD492nm. The absorbance values in both panels, A and B are means ± standard error of six mice per group.

(IgG, IgG1 and IgG2a ) of mice were significantly different (P < 0.01) among the immunized and control groups at 21 and 42 days post immunization. There was a significant rise (P = 0.002) in total IgG antibody response in group 2b compared to group 2a at 42 days post immunization. In Expt #1 and #2, there was a ∼2 fold increase in rVacJ specific IgG titers (P = 0.000) following the booster dose (Fig. 3, Panel A). A rise in amount/dose of rVacJ protein in immunized mice (group 2a and 2b) in Expt #2 contributed to substantial rise (P = 0.000) in serum antibody titers (total IgG) in comparison to Expt #1. Variable titers of subtypes IgG1 /IgG2a between Expt #1 and #2 were noticed. In Expt #1, the IgG1 /IgG2a ratio was nearly one (0.97). However, a more pronounced rise in IgG1 titer (P = 0.000) was observed 42 days post immunization in Expt#2 with IgG1 /IgG2a ratio in the range of 2.05–2.29 (Fig. 3, Panel B and Table 1). In Expt #1, there was no significance (P = 0.102) between IgG1 and IgG2a , whereas, significant (P = 0.000) rise in subtypes (group 2a and 2b) were noticed in Expt #2.

3.4. Protective efficacy of rVacJ The challenge strain (P. multocida B:2 strain P52) had LD50 of 15 CFU. Results of the challenge studies are indicated in Table 1 and Fig. 4. In Expt #1, all the immunized and control mice succumbed to infection (100% mortality) within 2-5 days post challenge. However, rVacJ (30 ␮g) + alum immunized mice survived by 2 days increase in MST (4.67 ± 0.76) with a significance (P = 0.025) over control group (1c). In Expt #2, immunization with rVacJ (75 ␮g) + FCA resulted in 66.7% protection (P = 0.025) and 6 days increase in MST (10.67 ± 2.11) with a significance (P = 0.008) over lower dose rVacJ and 8 days increase in MST over the PBS control (2c). Additionally, rVacJ (75 ␮g) + alum immunized mice resulted in 50% protection (P = 0.07) and 6 days increase in MST (8.83 ± 2.33) with a significance (P = 0.027) over control group (2c). However, no significant difference (P = 0.363) was found between the MST (P = 0.261) of rVacJ (75 ␮g) along with FCA (2a) and alum

294

S.B. Shivachandra et al. / Vaccine 32 (2014) 290–296

Table 1 Details of survival/mortality pattern of rVacJ immunized mice following challenge with P. multocida serogroup B:2 strain P52. Experiment #

Groups

Antigen + adjuvant

Challenge dose

Number of mice survived/number of mice challenged [% protection]

Expt #1

1a 1c

rVacJ (30 ␮g) + FCA Control-1 [PBS]

100 LD50 100 LD50

0/6 [0.0%] 0/6 [0.0%]

Expt #2

2a 2b

rVacJ (75 ␮g) + FCA rVacJ (75 ␮g) + Alum Control-2 [PBS]

8 LD50 8 LD50

4/6 [66.7%]* 3/6 [50.0%]

8 LD50

0/6 [0.0%]

2c

Mean survival time [mean days ± SE]

Ratio of IgG1 /IgG2a

4.67 ± 0.76* 2.67 ± 0.21

0.97 0.75

10.67 ± 2.11* 8.83 ± 2.33*

2.29 2.05

3.00 ± 0.45

0.77

Abbreviations: Expt, experiment; FCA, Freund’s complete adjuvant; PBS, phosphate buffered saline; LD, Lethal dose 50%; SE, standard error. * Statistically significant as compared to control group, P < 0.05.

Fig. 4. Protective efficacy of rVacJ lipoprotein in mouse model. The percentage survival curve showing survival/mortality pattern of immunized as well as control mice (6 mice/group) in both Expt #1 and Expt #2 following challenge with 100 LD50 and 8 LD50 of P. multocida B:2 strain P52 respectively. The mean data of survival and mortality are indicated in Table 1.

(2b) groups. Furthermore, P. multocida B:2 was re-isolated from the heart blood of each challenged mouse after death. These isolates were confirmed by standard bacteriological tests and PCR assays, which produced amplicons of ∼460 bp and ∼620 bp, specific for P. multocida and serogroup B:2. 4. Discussion Genes encoding outer membrane lipoproteins are ubiquitous in bacteria and constitute approximately 1–3% of their total gene set [26]. Recent analysis of P. multocida genomes identified genes encoding several putative vaccine candidates including outer membrane lipoproteins [13,21] as they are associated with diverse known/unknown functions [27,28]. In the present study, comparative analysis of vacJ sequences from several P. multocida strains revealed that they are highly conserved irrespective of serogroup, disease, host species or geographical origin (unpublished data). This indicates that VacJ is widely distributed among different P. multocida strains and could be universally prevalent in all members of the species/genus. Previously, Wu et al. [16] also noted higher percentage (90.8–100%) of plpE gene sequence homology among different P. multocida strains. However, the percentage of identity between VacJ and PlpE was less than 8%. Interestingly, widespread distribution of VacJ-like proteins with varying gene/protein lengths among several Gram-negative bacteria has also been reported. However, the role of VacJ in generating protective immunity has not yet been elucidated.

Structurally, majority of triacylated bacterial lipoproteins are considered to be similar in each bacterium and known to play important diverse roles in bacterial physiology and virulence [28]. A lipobox motif, as predicted by LipoP v1.0 which is known to have a sensitivity of 0.964, is typically L-3-[A/S/T]-2-[G/A]–1-C+1 with the +1 cysteine absolutely conserved in all bacterial lipoproteins [29]. The presence of a characteristic lipoprotein motif ‘Leu-X-GlyCys’ at the C-terminal end of the signal sequence (17 L-S-G-C20 ) strongly suggested that VacJ was a lipoprotein exposed on the bacterial surface. In bacteria, lipoproteins are synthesized with a N-terminal hydrophobic signal peptide that is cleaved from the mature polypeptide by lipoprotein signal peptidase (LSP) prior to covalent linkage of fatty acids [30]. The predominant hydrophilic regions were found dispersed across the entire length of protein and this observation was also supported by surface probability and high antigenicity. VacJ was found to possess predominately ␣-helices with fewer (