Correspondence: Ming Liao, Key Laboratory of Veterinary Vaccine Innovation of the Ministry of Agriculture, College of Veterinary Medicine, South China Agricultural University, 483 Wushan Street, Tianhe District, Guangzhou 510642, China. Tel.: +86 020 85280240; fax: +86 020 85285282; e-mail: email@example.com
Haemophilus parasuis outer membrane protein P2 (OmpP2), the most abundant protein in the outer membrane, has been identified as an antigenic protein and a potential virulence factor. To study the precise function of OmpP2, an ompP2-deficient mutant (ΔompP2) of a H. parasuis serovar 4 clinical strain SC096 was constructed by a modified natural transformation system. Compared with the wild-type SC096 strain, the ΔompP2 mutant showed a pronounced growth defect and exhibited significantly greater sensitivity to the bactericidal action of porcine and rabbit sera, whereas the complemented strain could restore the growth and serum resistance phenotypes. The results indicated that H. parasuisOmpP2 from SC096 strain is an important surface protein involved in serum resistance.
Haemophilus parasuis is the causative agent of Glässer's disease, which is characterized by fibrinous polyserositis, polyarthritis and meningitis. Haemophilus parasuis infection produces significant mortality and morbidity in pig farms, giving rise to important economic losses in the pig industry (Oliveira & Pijoan, 2004). To date, 15 serovars have been described, with apparent differences in virulence (Kielstein & Rapp-Gabrielson, 1992); the virulent serovars 5 and 4 are the most prevalent serovars in China (Cai et al., 2005). Serum-resistance in H. parasuis is frequently associated with systemic disease in swine, suggesting that it is a potential pathogenic mechanism of this bacterium (Cerda-Cuellar & Aragon, 2008). However, the major determinants of serum resistance in this pathogen are largely unknown.
Natural transformation is a process by which bacteria take up extracellular DNA and incorporate it into the host genome by homologous recombination (Wang et al., 2006). Haemophilus parasuis has a cyclic AMP (cAMP)-dependent natural transformation system that enables the uptake of DNA in which the ACCGAACTC sequence signal must be present (Bigas et al., 2005). Using this system, a thy-deficient mutant of H. parasuis has been obtained previously (Bigas et al., 2005). Therefore, natural transformation provides a method for the construction of mutants to study the function of H. parasuis genes.
Haemophilus parasuis outer membrane protein P2 (OmpP2), a member of the porin family, is the most abundant protein in the outer membrane (Zhou et al., 2009). Mullins et al. (2009) reported that H. parasuis OmpP2 proteins exhibit a high level of sequence heterogeneity and that two distinct protein structures exist in this bacterium, suggesting that OmpP2 has experienced high selective pressure which may contribute to virulence. Furthermore, H. parasuis OmpP2 has been identified as a target for protective antibodies and OmpP2 vaccines provide partial protection to mice against this bacterial infection (Zhou et al., 2009). In this study, we constructed a H. parasuis ompP2-deficient mutant (ΔompP2) by a modified natural transformation method to investigate the role of the OmpP2 in serum resistance.
Materials and methods
Bacterial strains, plasmids and culture conditions
Bacterial strains and plasmids used in this study are listed in Table 1. Escherichia coli plasmids were propagated in E. coli DH5α and grown in Luria–Bertani medium. Haemophilus parasuis strains were used and cultivated in trypticase soy agar (TSA) and trypticase soy broth (TSB) (Oxoid, Hampshire, UK) supplemented with 0.002% nicotinamide adenine dinucleotide (NAD) (Sigma, St. Louis, MO) and 5% inactivated bovine serum at 37 °C in a 5% CO2-enriched atmosphere for 36 h. When required, the media were supplemented with kanamycin (30 mg mL−1) or gentamicin (20 mg mL−1).
KanR, a 990-bp fragment containing the upstream and downstream sequences of the OmpP2 gene in pK18mobsacB
KanR GmR, a 1.790-kp fragment containing the motif of 5′-ACCGAACTC and the ΔompP2::Gm cassette in pK18mobsacB
KanR GmR, a 2.803-kp fragment containing hepII::GmR cassette in pK18mobsacB
KanR GmR, a 1.790-kp fragment containing the motif of 5′-ACCGCTTGT and the Δ ompP2::Gm cassette in pK18mobsacB
KanR GmR, a 4.313-kp fragment containing containing hepII::ompP2+kan cassette in pK18mobsacB
Oligonucleotide primers used for PCR (Table 2) were synthesized at Takara (Dalian, China). A 990-bp PCR fragment containing the 477-bp upstream of the ATG start codon and the 513-bp downstream of the TAA stop codon of ompP2 gene was amplified using overlap PCR with primers (P1 and P4) and subsequently cloned into plasmid pK18mobsacB to create pZB1. Both DNA fragments (upstream and downstream) contained the 9-bp core DNA uptake signal sequence (USS) of 5′-ACCGAACTC (Bigas et al., 2005). Next, an 800-bp gentamicin resistance cassette was amplified from a p34s-Gm plasmid with primers (P5 and P6). Both the pZB1 and the gentamicin resistance cassette were digested with BamHI and SalI and then ligated together to form plasmid pZB2. A pZB3 plasmid contained the entire heptosyltransferase (hep) II gene plus 517-bp upstream of the ATG start codon and 433-bp downstream of the TAA stop codon, and the gentamicin resistance cassette ligated between the hepII gene and the downstream sequence. To obtain plasmid pZB4, a mutation cassette of the OmpP2 gene was amplified from the pZB1 plasmid using primers (P11 and P12) containing a novel putative USS of 5′-ACCGCTTGT. Next, this PCR product was cloned into pK18mobsacB to make pZB4. A 2.32-kb PCR fragment was amplified using overlap PCR with primers (P13 and P16), which contained the complete open reading frame (ORF) of ompP2 gene and the kanamycin resistance cassette. Both the fragment and the pZB3 plasmid were excised with BamHI and SalI and then ligated together to form plasmid pZB5. All plasmids were mobilized into E. coli DH5α by CaCl2-mediated transformation.
Table 2. Sequences of the PCR primers used in this study
A natural transformation assay was performed using the method of Bigas et al. (2005) with some modifications. Recipient bacteria were cultured overnight at 37 °C and resuspended in TSB supplemented with serum and NAD at 5 × 1010 CFU mL−1. A 20-μL aliquot of the suspension was spotted onto a TSA plate supplemented with serum and NAD and spread onto a small area. Next, 1 μg of donor DNA plasmid resuspended in TE buffer was added, mixed and incubated for 5 h at 37 °C. Bacterial cells were scraped up and plated on the selective medium and incubated at 37 °C for 2–3 days. Additionally, TE buffer was added to a bacterial spot, instead of donor DNA, as a negative control.
OMP extraction and SDS-PAGE
To characterize the outer membrane protein (OMP) profiles of the wild-type and mutant strains, OMPs were extracted from H. parasuis according to a previously described method with some modifications (Zhou et al., 2009). Briefly, H. parasuis cultures were harvested by centrifugation for 10 min at 4500 g. The pellet was resuspended in 10 mM HEPES buffer (pH 7.4), and the suspension was subjected to sonication. Cellular debris was removed by centrifugation (10 000 g, 10 min, 4 °C). The supernatant was then removed and centrifuged at 200 000 g for 45 min at 4 °C. The supernatant was discarded, and the pellets were resuspended and washed in 10 mM HEPES buffer (pH 7.4) containing 2% sodium lauryl sarcosinate for 30 min. Pellets containing cell membrane materials were collected by centrifugation at 200 000 g for 45 min at 4 °C and solubilized in 10 mM HEPES buffer (pH 7.4). Finally, OMPs were separated on SDS-PAGE and visualized by Coomassie blue staining.
Sera and serum bactericidal assay
Normal rabbit serum was obtained from the Laboratory Animal Center of South China in Guangzhou, China. Porcine serum consisted of a pool of sera collected from five healthy piglets (3–4 weeks old) from a farm free of Glässer's disease. Both sera were filter-sterilized (0.22 μM) and aliquots were stored at −80 °C. Some aliquots of the sera were treated at 56 °C for 30 min to inactivate the complement.
The serum bactericidal assay was performed with porcine and rabbit sera as previously described (Cerda-Cuellar & Aragon, 2008) with some modifications. Briefly, 100 μL of each aliquot of fresh serum or heat-treated serum was mixed with 100 μL of bacterial suspension (approximately 1 × 108 CFU mL−1) to achieve a final concentration of 50% serum. Then, 180 μL of each aliquot of fresh serum or heat-treated serum was mixed with 20 μL of bacterial suspension (approximately 1 × 107 CFU mL−1) to achieve a final concentration of 90% serum. The mixtures were incubated at 37 °C for 1 h with gentle shaking. After incubation, 10-fold serial dilutions of the samples were made and placed on TSA plates containing inactive bovine serum and NAD. The plates were incubated at 37 °C with 5% CO2 for 36 h, at which point the colonies were counted. The percent survival was calculated by the ratio of colonies in fresh serum to those in heat-treated serum. Each H. parasuis strain was tested in three independent experiments.
Comparison of several test series was evaluated by analysis of variance (anova). The significance of differences was determined using Student's t-test. A P value of < 0.05 was considered statistically significant.
Results and discussion
Development of a modified natural transformation method of H. parasuis
Using the method of Bigas et al. (2005), no transformants were obtained when the seven different clinical isolates and four reference strains listed in Table 1 were transformed with the pZB2 plasmid carrying the ompP2::GmR cassettes. This result suggested that the strains might not share the reported USS (5′-ACCGAACTC) or might be non-transformable strains. Therefore, we searched the H. parasuis SH0165 strain genome (GenBank accession no. NC_011852) to determine the prevalence of the alternative motif, 5′-ACCGCTTGT. In total, 523 occurrences of this motif were found, a much higher number than of the reported USS (13 occurrences, including its complement). Recently, Xu et al. (2011) also reported the 5′-ACCGCTTGT motif as a DNA USS in the SH0165 strain genome. To confirm that the 5′-ACCGCTTGT motif was required for H. parasuis transformation, the hepII gene, containing this motif 842 bp from its translational start point was selected for test transformations. Of the seven isolates and four reference strains, only the SC096 strain was transformable with plasmid pZB3 under the conditions tested. Hence, this strain will be used as a model strain to investigate the gene function by genetic manipulation.
Previously, Bigas et al. (2005) demonstrated failure of transformation in the absence of cAMP. In this study, we tested the effect at any concentration of cAMP in the transformation assay (Fig. 1). The results showed no significant difference at any concentration of cAMP in the transformation assay.
Construction of the H. parasuis ΔompP2 mutant and complemented strains
To obtain the ΔompP2 mutant, the pZB4 plasmid was used as donor DNA, introduced by natural transformation into strain SC096. Colony PCR was used to check the gentamicin-resistant transformants (Fig. 2). As expected, the primers P1 and P4 amplified a 2.045-kb fragment from the wild-type strain. In the ΔompP2 mutant, this fragment was decreased to 1.753 kb by replacement of the ompP2 sequence with the GmR cassette. Sequencing of PCR products further confirmed replacement of the ompP2 sequence by the GmR cassette in the ΔompP2 mutant.
According to the method of Saeed-Kothe et al. (2004), transformation of Haemophilus influenzae with a complementation construct directs integration of a gene of interest into the chromosome. In this study, a single-copy, chromosome-based complementation plasmid of pZB5 was constructed and transformed into the ΔompP2 mutant. Many kanamycin-resistant transformants were obtained and checked for specified homologous recombination by PCR with primers P13 and P16 (Fig. 2). As predicted, the primers amplified a 2.32-kb fragment containing the ompP2 gene and the KanR cassette in the complemented strain, whereas no fragment was observed in the ΔompP2 mutant. Sequencing further confirmed that the complete OmpP2 ORF was integrated into the non-coding region of the hepII gene, 76 bp downstream of the TAA stop codon.
To further describe the ΔompP2 mutant, the OMP profiles showed that the expression of a protein of approximately 37 kDa was absent in the ΔompP2 mutant compared with the wild-type strain (Fig. 2). The ORF for OmpP2 in the wild-type strain is 1.092 kb (GenBank accession no. HQ709244) and the cleavage of the signal sequence (the first 20 N-terminal residues) results in a mature OmpP2 protein with a predicted molecular mass of 37.2 kDa, which closely approximates the size of the protein absent in the ΔompP2 mutant determined by SDS-PAGE of the OMP preparation. The expression of OmpP2 was restored in the complemented strain. Thus, the result further confirmed that the ompP2 gene was deleted from the genome of strain SC096. In addition, there appeared to be two bands of approximately 25 kDa present in the ΔompP2 mutant strain, suggesting further alterations in the protein composition of the outer membrane as a possible result of changes in protein expression or instability of other outer membrane proteins.
A pronounced growth defect in the H. parasuis ΔompP2 mutant
In Gram-negative bacteria, porins form transport channels that are involved in the uptake of essential nutrients required for bacterial growth (Achouak et al., 2001). In this study, deletion of the ompP2 gene in H. parasuis SC096 strain led to a pronounced growth defect (Fig. 3). The generation time of the ΔompP2 mutant (~68 min) was significantly longer than that of the wild-type strain (~50 min) (P < 0.001), whereas the complemented strain restored the growth phenotype (~52 min). The results suggested that OmpP2 played an important role in the growth of the H. parasuis SC096 strain. Furthermore, our findings were similar to those in a previous study that described a severe growth defect in an H. influenzae type b ΔompP2 mutant (Cope et al., 1990). Our results thus indicated that OmpP2 had a similar function in growth in both H. parasuis SC096 and H. influenzae DL42 strains.
The H. parasuis ΔompP2 mutant exhibits increased sensitivity to complement
The ability of bacteria to produce systemic infection often corresponds to resistance to the bactericidal activity of the host complement, allowing bacteria effectively to evade immune responses and to survive in the blood stream (Cerda-Cuellar & Aragon, 2008). Thus, serum resistance represents an important virulence strategy of bacterial pathogens. The porins of Por1A and Por1B in Neisseria gonorrhoeae were both involved in serum resistance (Ram et al., 1998, 2001). In H. influenzae type b, loss of OmpP2 expression only led to a slower growth rate in normal infant rat serum but did not increase the serum bactericidal activity (Cope et al., 1990).
In this study, we first investigated the effect on serum resistance of the wild-type SC096 strain in 50% and 90% serum compared with the reference strains SW114, Nagasaki, C5, 84-17975 and the clinical isolate SC003 (Fig. 4). The level of survival in 90% serum of the Nagasaki strain was similar to that previously described (Cerda-Cuellar & Aragon, 2008). In 50% and 90% serum, the SC096, Nagasaki and 84-17975 strains showed significantly increased resistance to serum killing compared with the SW114, C5 and SC003 strains. Therefore, the results indicated that the SC096 strain is highly resistant to the bactericidal activity. In addition, the ORFs for OmpP2 in the Nagasaki (1.08-kb), 84-17975 (1.08-kb) and SC096 (1.092-kb) strains are shorter in length than those of the SW114 (1.191-kb), C5 (1.203-kb) and SC003 (1.182-kb, GenBank accession no. JN571296) strains. The results indicated that H. parasuis strains possessing shorter length OmpP2 proteins exhibited significantly increased resistance to complement killing. There are two distinct OmpP2 structures in H. parasuis, and two discontinuous sequence insertions of the longer ompP2 gene result in an additional extracellular loop in the predicted protein structure (Mullins et al., 2009). Accordingly, it was suggested that the additional extracellular loop of OmpP2 proteins might contribute to serum susceptibility in H. parasuis.
Compared to the wild-type SC096 strain, loss of OmpP2 expression resulted in significantly increased sensitivity to serum killing, with the mutant exhibiting extremely low levels of survival in porcine and rabbit sera (Fig. 4). Complementation of the ΔompP2 mutant restored the serum-resistant phenotype. The results suggested that an important role of H. parasuis OmpP2, at least in the SC096 strain, appeared to be its ability to protect against the bactericidal activity of complement. Future in vivo studies are required to investigate this further.
In conclusion, in this study, a modified natural transformation method in H. parasuis was developed that could provide an avenue to identify the function of different genes. Using this genetic manipulation system, the ΔompP2 mutant of the H. parasuis SC096 strain was determined to be significantly more sensitive to serum killing than its wild-type strain. The results indicated that OmpP2 is required for serum resistance in H. parasuis SC096, belonging to serovar 4.
This work was supported by the Program for New Century Excellent Talents in University (Grant No. NCET-06-0752), the Program for Changjiang Scholars and Innovative Research Teams in Chinese Universities (Grant No. IRT0723) and the Innovative Research Teams Program of Guangdong Natural Science Foundation (Grant No. 5200638).