Development of a neutralizing mouse–pig chimeric antibody with therapeutic potential against Haemophilus parasuis in Pichia pastoris

Authors

  • Zheng Chai,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Fang Fu,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Fucheng Jiang,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Huabin Tian,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Zhuo Wang,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Nan Zheng,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Xueyun Zhang,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Xiangling Wang,

    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    Search for more papers by this author
  • Xi Li

    Corresponding author
    1. State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Harbin, China
    • Correspondence: Xi Li, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences No. 427 Maduan St., Nangang District, Harbin 150001, China. Tel.: +86 189 4606 6129;

      fax: +86 451 5199 7166;

      e-mail: lx2005@126.com

    Search for more papers by this author

Abstract

Haemophilus parasuis is one of the most important bacterial diseases of pigs worldwide. The lack of a vaccine against a broad spectrum of strains and the limitation of antimicrobial susceptibility hamper the control of disease. In this study, we cloned the constant regions of gamma heavy chains and kappa light chain of pig lymphocytes in frame with the variable regions of heavy and light chains of mouse monoclonal antibody 1D8, which reacts with all 15 serotypes of H. parasuis and has neutralizing activity. The constructed mouse–pig chimeric antibody was expressed in Pichia pastoris. Results demonstrated that the expressed chimeric antibody inhibited the growth of H. parasuis in vitro. Furthermore, the experiments in mice showed that chimeric antibody increased survival rate of the mice compared with that of the control group (< 0.05). Importantly, the chimeric antibody partially protected piglets against H. parasuis infection according to the clinical lesion scores and PCR results of H. parasuis in the tissues from piglets of the chimeric antibody-inoculated group and the PBS group. In summary, our results demonstrated that the mouse–pig chimeric antibody could be a therapeutic candidate to prevent the H. parasuis infection and control the prevalence of disease.

Introduction

Haemophilus parasuis (H. parasuis) is a Gram-negative, nonhemolytic, NAD-dependent bacterium. To date, 15 serovars of H. parasuis have been described, but c. 15% to 41% of the field isolates are untypeable (Rapp-Gabrielson & Gabrielson, 1992; Oliveira & Pijoan, 2004). In China, the 15 serovars and diverse genotypes of H. parasuis were widely distributed (Zhang et al., 2012b). Haemophilus parasuis is the etiological agent of Glässer's disease in swine, pathologically characterized by fibrinous polyserositis, meningitis, polyarthritis, and arthritis syndrome (Nielsen, 1993). In recent years, H. parasuis infection has been considered as a major cause of nursery mortality in swine herd (Oliveira & Pijoan, 2004).

The previous studies have been reported that Glässer's disease can be successfully treated with antimicrobials. However, many H. parasuis isolates were susceptible to antimicrobial agents routinely used for treatment (Aarestrup et al., 2004; de la Fuente et al., 2007). Bacterins have been used to prevent Glässer's disease. However, limited cross-protection among strains has complicated the control of Glässer's disease (Nielsen, 1993). Some virulence factors of H. parasuis have been reported (Sack & Baltes, 2009; Costa-Hurtado et al., 2012; Zhang et al., 2012a; Zhou et al., 2012); however, little is known about protective antigens. Therefore, the development of novel therapeutic methods is necessary.

We have previously reported that the neutralizing monoclonal antibody (MAb) 1D8 can react with all 15 serotypes of H. parasuis and protect mice from both homologous and heterologous challenges with H. parasuis (Tian et al., 2011). It might be a potential therapy method to protect swine against H. parasuis. However, direct application of murine MAbs in other species may be problematic. Antibodies from different animal species may not interact with Fc receptors and/or complement, which lead to a lack of appropriate down-stream functions and rapid clearance from the body (Tabrizi et al., 2006; Goodchild et al., 2011). An attempt to overcome this disadvantage is to generate chimeric MAbs that contain swine Fc domains and retain targeting specificity by incorporating portions of the murine variable regions. The production of chimeric antibodies using recombinant DNA technologies for therapy and prophylaxis in human has been used widely in the last years (Johnson et al., 1997; Zeitlin et al., 1998; Oliphant et al., 2005; Hu et al., 2007). Furthermore, recombinant antibodies have been developed for the control of infectious diseases in animals (Jar et al., 2009; Pyo et al., 2009).

In most humanized antibody cases, chimeric antibodies are generated in the mammalian expression system (Xiong et al., 2005; Kim et al., 2011). However, the yield of transient expression in this expression system is usually low. Meanwhile, product cost plays an important role in the development of animal biologics. In this study, we grafted heavy- and light-chain variable domains (VH and VL) of murine MAb 1D8 onto the swine immunoglobulin gamma (IgG1a) heavy- and light-chain constant regions (CH and CL), respectively. The mouse–pig chimeric antibody were generated in Pichia pastoris (P. pastoris) expression system. The chimeric antibodies retained neutralizing activity in vitro and showed the therapeutic efficacy in mice and pigs with H. parasuis infection.

Materials and methods

Ethics statement

This study was carried out in accordance with animal ethics guidelines and approved protocols. All animal studies were approved by the Animal Ethics Committee of Harbin Veterinary Research Institute of the Chinese Academy of Agricultural Sciences (SYXK (H) 2006-032).

Cloning of variable regions of murine kappa and gamma chains and constant regions of porcine kappa and gamma chains

The hybridoma 1D8 cell line was kept in our laboratory, which produced a murine immunoglobulin of IgG2b/kappa isotype (Tian et al., 2011). RNA was separately extracted from murine hybridoma 1D8 or swine lymphocytes and then was used to prime cDNA synthesis (Promega). PCR amplification was performed using the primer pairs (Supporting Information, Table S1) for the variable regions (VH and VL) of the MAb 1D8 and the constant region (CH and CL) of porcine IgG. The fragments H and L chains were separately generated and verified by sequencing.

Construction and expression of recombinant chimeric antibody in P. pastoris

The (Gly)3-Ser linker was designed between the H chain and L chain. The H-L gene fragment was ligated into the pPIC9K vector. The recombinant plasmid was linearized with SalI and then transformed into P. pastoris GS115 cells using a Gene Pulser (Bio-Rad; conditions: 1.5 kV, 25 μF, and 200Ω). His+ transformants were selected and then inoculated in 5-mL buffered complex glycerol medium. The thalli were resuspended in 20-mL buffered complex methanol medium and induction performed for 48 h. Methanol was added to a final concentration of 0.5% (v/v) at 24 h intervals. The supernatant was harvested.

Western blot and Dot blot analysis

The culture supernatant of the recombinant yeast strain was subjected to 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis. For Western blot, proteins in the gel were transferred to nitrocellulose membrane that was then blocked. The membrane was incubated with a horseradish peroxidase-conjugated goat antiporcine IgG antibody (whole molecule) (Sigma) and then incubated with 3, 3-diaminobenzidine tetrahydrochloride as a substrate to visualize the reaction result. The dot blot procedure was performed as Western blot described. Protein denaturation reagents were omitted during sample preparation to preserve the native conformation.

Bactericidal assay in vitro

The bactericidal assay was performed as previously described (Gatto et al., 2002; Tian et al., 2011). The highly virulent H. parasuis HLJ-018 strain used in this study was kept in our laboratory (Tian et al., 2011). The log-phase H. parasuis were diluted to c. 2.4 × 103 CFU mL−1 and mixed with recombinant protein (20 μg), Mab 1D8 (20 μg), and PBS, respectively. The mixtures were incubated for 30 min at 37 °C and then plated onto the TSA agar plates. After growth overnight, colony numbers were measured. This experiment was performed three times.

Treatment of chimeric antibody in mice

For the protective efficacy in mice, eighteen 6-week-old BALB/c mice were randomly assigned to three groups of six each. All of the mice were challenge with a lethal dose of H. parasuis HLJ-018 (c. 3.0 × 1010 CFU) by intraperitoneal (IP) injection. One hour later, the mice were inoculated by IP injection in a 500 μL volume with the supernatant of recombinant yeast cells (25 μg/mouse), MAb 1D8 (25 μg/mouse), or PBS. The survival rates of mice were analyzed.

Treatment of chimeric antibody in piglets

The piglets were obtained from a farm with a standard health status and no reported Glässer's disease outbreaks. All of the piglets were negative for the detection of Actinobacillus pleuropneumoniae, porcine reproductive and respiratory syndrome virus, porcine pseudorabies virus, and porcine circovirus. Eighteen large white piglets (33 days old) were randomly assigned to three groups of six each. All animals were intranasally inoculates with 2 × 1010 CFU/animal of H. parasuis HLJ 018. After 1 h, the recombinant protein (220 μg/piglet), MAb 1D8 (220 μg/piglet), or PBS was intravenously inoculated. Rectal temperatures were recorded once a day during the whole experiment. All piglets were euthanized after 7 days postchallenge. A lesion score was calculated as the sum of individual lesions/signs (lack of lesion = 0; presence of lesion = 1): pleuritis, peritonitis, arthritis, meningitis, and pericarditis. The heart, liver, spleen, lung, kidney, inguinal lymph node tissues, and pericardial fluid were removed from each euthanized piglet for re-isolating bacteria. Randomly selected bacterial isolates recovered from different tissues of all the animals were confirmed as H. parasuis by PCR (Oliveira et al., 2001).

Statistical analysis

The graphpad prism version 5.01 was used for statistical analysis. Survival curve was analyzed with the log-rank test. < 0.05 was defined as a significant difference.

Results

Cloning of mouse–pig chimeric antibody

Each of the genes encoding VH and VL of murine MAb 1D8 was amplified from hybridoma MAb 1D8. The genes of swine IgG constant regions (CH and CL) were separately amplified from the cDNA of lymphocytes that were isolated from swine peripheral blood mononuclear cells. As shown in Fig. 1a, the fragments of VH, VL, CH, and CL were obtained, respectively. The sequences of the VH, VL, and CL were identical to the gene sequence reported at GenBank (KF561241, KF561242 and KF561240, respectively). Porcine constant gamma chain showed 100% identity with previously reported gamma (GenBank accession number U03781). The variable regions from murine MAb 1D8 and constant regions from swine IgG1a were spliced into the whole H and L chains of chimeric antibody using overlapping PCR (Fig. 1b).

Figure 1.

Product of nucleotide sequence of mouse–pig chimeric antibody. (a) RT-PCR products of VH, VL, CL, and CH genes were generated from hybridoma 1D8 cells and swine PMBC lymphocytes and were analyzed by agarose gel electrophoresis. (b) PCR products of H and L chain.

Expression of chimeric antibodies in baculovirus and P. pastoris system

The constructed vector was integrated into the His4 locus of P. pastoris GS115 by electroporation. The recombinant yeast strain was selected from the His+Mut+ Pichia transformants and induced with 0.5% methanol in shake flask culture. The chimeric antibody proteins were obtained from the supernatants of culture media after 48 h of methanol induction. The recombinant protein using P. pastoris expression system accumulated up to 55 μg mL−1.

The expression of chimeric antibodies was confirmed by Western bolt (Fig. 2a) and dot blot (Fig. 2b). The expressed H and L chains with molecular weight of about 50 kDa and 25 kDa were observed in the cultural supernatant of the yeast cells (Fig. 2a). In contrast, no similar band (Fig. 2a) or spot (Fig. 2b) appeared in negative control strain GS115.

Figure 2.

Expression of the recombinant chimeric antibody protein in Pichia pastoris. (a) Western blot analysis. (b) Dot blot analysis. A 10 μL culture supernatant of the recombinant P. pastoris strain induced for 48 h was separated by SDS-PAGE and blotted onto a nitrocellulose membrane. The second antibody was detected using antiporcine IgG (H + L). The supernatant of GS115 strain after 48 h of induction was negative control.

Assessment of chimeric antibody in bacteria activity assay

To ensure that the chimeric antibodies have the ability to inhibit the growth of H. parasuis HLJ-018 strain, the bactericidal activity of chimeric antibody proteins was assessed by the differences of colony numbers between experimental and control groups (Fig. 3). The colony numbers of H. parasuis that were treated with chimeric antibody or MAb 1D8 were reduced significantly compared with those of PBS-inoculated group (< 0.01). Moreover, there was no significant difference between chimeric antibody and MAb 1D8. The results suggested that chimeric antibodies as well as MAb 1D8 can effectively inhibit the growth of H. parasuis in vitro.

Figure 3.

Bactericidal activity of mouse–pig chimeric antibody in vitro. Haemophilus parasuis was mixed with recombinant protein, Mab 1D8, and PBS, respectively. The mixtures were plated onto the TSA agar plates. After growth overnight, colony numbers were measured. Asterisk indicated a statistically significant difference compared with the PBS control. ns indicated no statistically significant difference.

Treatment of chimeric antibodies against H. parasuis in mice

To determine the therapeutic efficacy of chimeric antibodies in vivo, all the mice were challenged with a lethal dose of H. parasuis HLJ-018 strain (c. 3.0 × 1010 CFU/mouse). After 1 h, chimeric antibodies, MAb 1D8, or PBS was intraperitoneally inoculated into BALB/c mice. The result is shown in Fig. 4. After 72 h, the survival rate of the recombinant protein group was 16.67%. There is no significant difference in the survival rate between the groups of recombinant protein and MAb 1D8 (= 0.0719). However, there is a significant difference between the groups of recombinant protein and PBS (< 0.05, = 0.0422).

Figure 4.

Survival of mice treated with mouse–pig chimeric antibody against Haemophilus parasuis HLJ-018 challenge. Kaplan–Meier survival curves of Balb/c mice are shown. All of the mice were challenged with a lethal dose of H. parasuis HLJ-018 (c. 3.0 × 1010 CFU) via intraperitoneal injection. One hour later, the mice received recombinant chimeric antibody, MAb 1D8, or PBS intraperitoneally.

Treatment efficacy of chimeric antibodies against H. parasuis in piglets

To characterize the therapeutic potential of mouse–pig chimeric antibody in pigs, all of the piglets were challenged with high dose of H. parasuis HLJ-018 strain. The animals treated with MAb 1D8 or PBS showed high temperatures at 1 day postchallenged (dpc.). The highest temperatures were obtained from 2 to 4 dpc. (Fig. 5). A similar tendency was observed for the group chimeric antibody (Fig. 5). However, there were significant differences in comparison with the piglets that were treated with MAb 1D8 or PBS from 2 to 4 dpc. (< 0.05). The results demonstrated that all the piglets that were inoculated with chimeric antibody showed the improved clinical performance.

Figure 5.

Rectal temperatures in piglets at different intervals after challenge with Haemophilus parasuis.

All the PBS control piglets and some of those inoculating chimeric antibodies or MAb 1D8 showed the characteristic inflammatory caused by H. parasuis (Table 1). Mild or moderate fibrinous polyserositis in the pericardial, pleural, and peritoneal cavities was observed in all of the piglets (Table 1). However, some piglets from MAb 1D8 or PBS group further presented fibrinous polyarthritis and mild meningitis (Table 1). Furthermore, pure cultures were recovered from heart, liver, spleen, lung, kidney, lymph node tissues, and pericardial fluid of most of PBS control piglets, while the decreased isolation rates were found for chimeric antibody group. In addition, all of the positive cultures were confirmed by PCR (Table 1).

Table 1. Numbers of piglets showing the lesions and recovery of Haemophilus parasuis after challenge
Group (n)No. of different lesions attributable to H. parasuisNo. of piglets positive for H. parasuis
PleuritisPeritonitisArthritisMeningitisPericarditisHeartLiverSpleenLungKidneyLymph node tissuesPericardial fluid
  1. a

    Significant difference compared with the PBS group (< 0.05).

  2. b

    Significant difference compared with the MAb 1D8 (< 0.05).

Recombinant protein (6)4/63/60/6a0/6a,b3/6a4/61/6a0/6a,b3/61/62/62/6a
MAb 1D8 (6)4/64/61/63/65/64/62/62/63/63/63/64/6
PBS (6)6/65/63/65/66/66/65/66/65/63/65/66/6

Discussion

Haemophilus parasuis is one of the most serious porcine respiratory diseases and causes large economic losses in the pig industry (Oliveira & Pijoan, 2004). The limited cross-protection of traditional vaccines and antimicrobial susceptibility hampers control of the disease. In this study, mouse–pig chimeric antibody were expressed in P. pastoris, maintained the inhibiting bacterium activity in vitro and showed its therapeutic efficacy against H. parasuis challenge in mice and piglets.

We previously found that a neutralized MAb 1D8 could reduce the colony numbers and prevalence of H. parasuis in various organs in mice that were challenged by H. parasuis (Tian et al., 2011). To limit or avoid possible side effects linked to the murine origin of the immunoglobulin, a cloning strategy replacing the constant regions of mice by the swine counterparts produces a mouse–pig chimeric antibody.

Heterologous expression of mouse–pig chimeric antibody has been reported in Escherichia coli as well as in baculovirus (Jar et al., 2009; Pyo et al., 2009). Escherichia coli expression system is simple and low costs; however, the expressed proteins may lack important post-translational modifications. Moreover, as recombinant protein in baculovirus system was generated in eukaryotic cell, its cost was very expensive. The lytic baculovirus cycle of Sf9 cells releases intracellular proteases that not only may alter the stability of secreted functional antibodies, but also could interfere with functional antibody qualification (Prigent et al., 2010). Pichia pastoris as a eukaryote has the advantages of higher eukaryotic expression systems and is easier and less expensive to use than other eukaryotic expression systems. In P. pastoris system, the α-factor secretion signal attached upstream of chimeric antibody. The expressed chimeric antibody was secreted into the medium. The advantage of secretory expression is that P. pastoris secretes very few native proteins and its culture medium is almost protein-free, and therefore, the secreted chimeric antibody should comprise the majority of the total protein in the medium (Cregg et al., 1987; Romanos, 1995). Therefore, we used P. pastoris expression systems to produce mouse–pig chimeric antibody for production and application of chimeric antibody in animals.

Western blot showed that the mouse–pig chimeric antibody was expressed in P. pastoris (Fig. 2a). Dot blot indicated that the constant region of chimeric antibody from swine was biological activite as chimeric antibody protein without denaturation reacted with anti-pig secondary antibody (Fig. 2b). Furthermore, chimeric antibodies significantly inhibited the bacterial growth in vitro (Fig. 3), which suggested that expression of variable regions for murine MAb 1D8 grafted with pig IgG antibodies does not impair the neutralization of the paratope in vitro.

To our knowledge, there was no data to show the therapeutic effect of mouse–pig chimeric antibody in animals, although the mouse–pig chimeric antibody have been reported (Jar et al., 2009; Pyo et al., 2009). In this study, our findings showed that mouse–pig chimeric antibody partially protected mice against lethal challenge with H. parasuis HLJ-018, suggesting that variable region of chimeric antibody has biologic activity in vivo (Fig. 4). Furthermore, the piglets were used to determine the therapeutic efficacy of mouse–pig chimeric antibody. Unfortunately, all of the piglets were survival, including the PBS control group. Because it is difficult to adjust a dose to reproduce the disease, limitations of the disease model used should be noticed (Olvera et al., 2011). However, results of clinical performance and lesion analysis showed that the chimeric antibody provided partial protection against H. parasuis infection (Fig. 5 and Table 1). On the one hand, the rectal temperatures of the piglets that were inoculated with MAb 1D8 or PBS were significantly higher than those of piglets with chimeric antibody treatment after H. parasuis infection (< 0.01), indicating that the chimeric antibody can alleviate the clinical manifestations of infected piglets On the other hand, the lesion records and bacterial isolations from tissues showed that the piglets with chimeric antibody treatment were partially protected against H. parasuis infection compared with the results of PBS control piglets. In further study, we would determine the therapeutic efficacy of the chimeric antibody against multiserotype strains of H. parasuis in vivo. Meanwhile, we would attempt to enhance the yields of chimeric antibody protein in P. pastoris by means of increasing chimeric antibody copy numbers or producing chimeric antibody in P. pastoris bioreactor.

This study described a novel H. parasuis prophylactic by expressing mouse–pig chimeric antibody in P. pastoris. The results demonstrated that the chimeric antibody was capable to inhibit bacterial growth in vitro and showed significantly therapeutic efficacy in the piglets with H. parasuis infection, which represents a potentially preventive and therapeutic tool in animals. Furthermore, chimeric antibody could be used alone or combined with other vaccination strategy to provide instant immunity in the face of an outbreak and before the establishment of effective adaptive immune response of the swine.

Acknowledgements

We thank Dr Chunlai Wang for providing experimental suggestions. This study was supported by Special Fund for Agro-scientific Research in the Public Interest (201203056).

Ancillary