Deletion of the vacJ gene affects the biology and virulence in Haemophilus parasuis serovar 5

Gene 603 (2017) 42–53

Contents lists available at ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

Research paper

Deletion of the vacJ gene affects the biology and virulence in Haemophilus parasuis serovar 5 Liangyou Zhao a,b, Xueli Gao a, Chaonan Liu a, Xiaoping Lv a, Nan Jiang c, Shimin Zheng a,⁎ a b c

College of Veterinary Medicine, Northeast Agricultural University, Harbin 150030, People's Republic of China Drug Safety Evaluation Center of Heilongjiang University of Chinese Medicine, Harbin 150040, People's Republic of China College of Life Science and Technology, Dalian University, Dalian 116622, People's Republic of China

a r t i c l e

i n f o

Article history: Received 12 June 2016 Received in revised form 28 October 2016 Accepted 10 December 2016 Available online 14 December 2016 Keywords: Haemophilus parasuis VacJ lipoprotein Outer membrane integrity Biofilm formation Serum resistance Adherence and invasion Virulence

a b s t r a c t Haemophilus parasuis is an important pathogen causing severe infections in pigs. However, the specific bacterial factors that participate in pathogenic process are poorly understood. VacJ protein is a recently discovered outer membrane lipoprotein that relates to virulence in several pathogens. To characterize the function of the vacJ gene in H. parasuis virulent strain HS49, a vacJ gene-deletion mutant ΔvacJ and its complemented strain were constructed. Our findings supported that VacJ is essential for maintenance of cellular integrity and stress tolerance of H. parasuis, by the demonstrations that the ΔvacJ mutant showed morphological change, increased NPN fluorescence and, and decreased resistance to SDS-EDTA, osmotic and oxidation pressure. The increased susceptibility to several antibiotics in the ΔvacJ mutant further suggested that the stability of the outer membrane was impaired as a result of the mutation in the vacJ gene. Compared to the wild-type strain, the ΔvacJ mutant strain caused a decreased survival ratio from the serum and complement killing, and exhibited a significant decrease ability to adhere to and invade PK-15 cell. In addition, the ΔvacJ mutant showed reduced biofilm formation compared to the wild-type strain. Furthermore, the ΔvacJ was attenuated in a murine (Balb/C) model of infection and its LD50 value was approximately fifteen-fold higher than that of the wild-type or complementation strain. The data obtained in this study indicate that vacJ plays an essential role in maintaining outer membrane integrity, stress tolerance, biofilm formation, serum resistance, and adherence to and invasion of host cells related to H. parasuis and further suggest a putative role of VacJ lipoprotein in virulence regulation. © 2016 Published by Elsevier B.V.

1. Introduction Haemophilus parasuis, a member of the family Pasteurellaceae, is one of the most important bacteria affecting pigs. It is the causative agent of Glässer's disease, which is characterized by fibrinous polyserositis, polyarthritis and meningitis (Oliveira and Pijoan, 2004). This pathogen is considered as an early colonizer and a member of the normal microbiota of the upper respiratory tract of piglets (Møller and Kilian, 1990). However, virulent strains can invade hosts and cause severe systemic disease under certain conditions. H. parasuis infection produces significant morbidity and mortality in contemporary swine production systems, and giving rise to large economic losses in swine-rearing countries (Cerdà-Cuéllar et al., 2010).

Abbreviations: NAD, nicotinamide-adenine dinucleotide; LB, Luria-Bertani; TSA, Tryptic Soy Agar; TSB, Tryptic Soy Broth; OD600, optical density at 600 nm; Kan, kanamycin; MIC, minimal inhibitory concentration; LD50, 50% lethal dose; i.p., intraperitoneally; ORF, open reading frame. ⁎ Corresponding author. E-mail address: [email protected] (S. Zheng).

http://dx.doi.org/10.1016/j.gene.2016.12.009 0378-1119/© 2016 Published by Elsevier B.V.

Multiple different genotypes and serotypes of H. parasuis have been described (Kielstein and Rapp-Gabrielson, 1992). However, there is not a clear association between virulence and H. parasuis phenotypes or genotypes, and the knowledge of the pathogenesis and putative virulence factors mechanisms in H. parasuis are largely unknown (Olvera et al., 2007; Chen et al., 2012). Virulence factor is a material basis of bacterial virulence which plays a major role in the pathogenesis of the pathogen. So the study of the biological characteristics of virulence factor has become the primary task of the pathogenic mechanism. However, very few virulence-associated factors have been identified in H. parasuis to date (Costa-Hurtado and Aragon, 2013; Blackall and Turni, 2013). Outer membrane lipoproteins are widely distributed in Gram-negative bacteria which are involved in diverse mechanisms of physiology/ pathogenesis (Sutcliffe et al., 2012). The VacJ lipoprotein was initially discovered in Shigella flexneri and is attributed for bacterial spreading (Suzuki et al., 1994). A previous study demonstrated that the vacJ gene of Non-typeable Haemophilus influenzae has contributed to serum resistance and IgM binding, thus mediating the escape from complement-dependent killing (Nakamura et al., 2011). VacJ is widely distributed among members of Pasteurellaceae and several other Gram-

L. Zhao et al. / Gene 603 (2017) 42–53

negative bacterial species, however, the functional characterization in the pathogenicity of H. parasuis has not been determined till date. H. parasuis stain SH0165 gene HAPS_2270 has also been predicted to encode VacJ lipoprotein (Yue et al., 2009). In this study, we inactivated the vacJ gene in H. parasuis serovar 5 strain HS49 and investigated the biological characteristics of vacJ mutant strain, including growth characteristics, outer membrane integrity, biofilm formation, serum resistance, adhesion and invasion to preliminary understand the role of vacJ gene in H. parasuis. Then the virulence of the ΔvacJ mutant was tested in mice to determine the effect of vacJ gene in virulence of H. parasuis. 2. Materials and methods 2.1. Bacterial strains, plasmids, and growth conditions The bacterial strains and plasmids used in this study are listed in Table 1. Escherichia coli strains were cultured in Luria-Bertani (LB) medium. H. parasuis was plated on Tryptic Soy Agar (TSA) or grown in Tryptic Soy Broth (TSB) (Difco Laboratories, Detroit, MI, USA), supplemented with 0.01% nicotinamide adenine dinucleotide (NAD) (Sigma, St. Louis, MO, USA) and 10% inactivated bovine serum at 37 °C. If necessary, 50 μg/mL of kanamycin or 20 μg/mL of gentamicin was complemented. To measure bacteria growth curves, cultures of bacteria were grown overnight in TSB supplemented with NAD and serum. The cultures were subinoculated into fresh TSB medium at a ratio of 1:100 and incubated at 37 °C. The optical density at 600 nm (OD600) was measured at 1 h intervals.

43

2.3. Construction of H. parasuis ΔvacJ mutant and complemented strains Primers used for amplification (Table 2) in this study were ordered from the Invitrogen (Shanghai, China). A 949 bp PCR fragment containing the 488 bp upstream of the ATG start codon and the 461 bp downstream of the TAA stop codon of the vacJ gene was amplified using overlap extension PCR with primers (P1 and P4), and subsequently cloned into plasmid pK18mobsacB to generate the plasmid pK18ΔvacJ. Both sets of primers (P1 and P4) contained a 9 bp core DNA uptake signal sequence (Zhang et al., 2011). A kanamycin resistance cassette (935 bp) was amplified from pBAD18-Km using primers P5 and P6. Then the pK18-ΔvacJ and the kanamycin resistance cassette were digested with BamHI and SalI and ligated together to create plasmid pK18-ΔvacJ::kan. The recombinant plasmid was introduced into H. parasuis strain HS49 by natural transformation, as described previously (Zhang et al., 2011, 2012). The complementation strain C-vacJ was constructed according to the method of Saeed-Kothe et al. (2004) and Zhang et al. (2012). A PCR fragment was amplified using overlap PCR with primers (P7 and P10), which contained the complete open reading frame (ORF) of vacJ gene and the gentamicin resistance cassette. Both the fragment and the pK18-ΔvacJ plasmid were excised with BamHI and SalI and then ligated together to form a single-copy, chromosome-based complementation plasmid of pK18-C-vacJ, and then this plasmid was transformed into the ΔvacJ mutant. The gentamicin-resistant transformants were checked for specified homologous recombination by PCR with primers P7 and P10. 2.4. Growth characteristics

2.2. Prediction of VacJ characteristics and lipobox motif The subcellular location of VacJ lipoprotein was predicted using the Cell-Ploc package (http://chou.med.harvard.edu/bioinf/Cell-PLoc/) (Chou and Shen, 2008). The VacJ characteristics were predicted with the PROTEAN program as well as proteomics tools from the ExPASy website (Shivachandra et al., 2014). The lipobox sequence in VacJ was predicted using the DOLOP program (Babu et al., 2006). The vacJ sequence was analyzed for identity and similarity to known sequences using BLAST (http://www.ncbi.nlm. nih.gov/BLAST/). Multiple alignment analyses of VacJ protein sequences were performed using ClustalX. Table 1 Bacterial strains and plasmids. Strains and plasmids

Relevant characteristics

Source

E. coli DH5α

FΦ80ΔlacZΔM15Δ(lacZYA-argF) U169 recA1 endA1 hsdR17

Laboratory collection

serotype 5 field isolate

Laboratory collection This work

H. parasuis HS49

ΔvacJ vacJ mutant of H. parasuis HS49, KanR (HS49ΔvacJ::kan) This work C-vacJ The complement of H. parasuis HS49 ΔvacJ::kan containing pK18-C-vacJ, KanR, GmR Plasmids Schafer et al. pK18mobsacB Suicide and narrow-broad-host vector, KanR (1994) pK18-ΔvacJ A 949 bp fragment containing the upstream This work and downstream sequences of the vacJ gene R in pk18mobsacB, Kan pK18-ΔvacJ::kan A 1884 bp fragment containing ΔvacJ::kan This work cassette in pk18mobsacB, KanR This work pK18-C-vacJ A 2496 bp fragment containing vacJ::Gm cassette in pk18mobsacB, KanR GmR R Guzman et pBAD18-Km Km resistance cassette-carrying vector, Kan al. (1995) R p34s-Gm Gm resistance cassette-carrying vector, Gm Yamanaka et al. (1995)

The H. parasuis wild-type strain HS49, mutant strain ΔvacJ, and CvacJ were cultured in 5 mL TSB supplemented with 10% inactivated bovine serum and 0.01% NAD for 16 h, and then diluted to OD600 of 1.0. The fresh cultures were then grown in 50 mL of the same medium at 37 °C. Samples of culture were monitored at 1 h intervals using a Eppendorf BioPhotometer (Eppendorf, Hamburg, Germany). Uninoculated TSB served as the blank control. The colony forming units (CFUs) were determined by performing a dilution series from the culture and counting colonies from the appropriate dilution at 2 h intervals. 2.5. Transmission electron microscopy For morphological detection, the bacteria were grown to logarithmic phase in TSB and harvested by centrifugation. Pellets were resuspended in phosphate buffered saline (PBS) and washed three times. Then the cultures were fixed and dehydrated as described previously (Murphy et al., 2006) to be detected by transmission electron microscopy (TEM). 2.6. SDS-EDTA permeability assay SDS-EDTA permeability assay was conducted as described previously by Carpenter et al. (2014). Briefly, the H. parasuis wild-type strain Table 2 Sequences of PCR oligonucleotide primers. Primers

Sequence (5′–3′)a

P1 (vacJ-up-F) P2 (vacJ-up-R) P3 (vacJ-down-F) P4 (vacJ-down-R) P5 (Kan-F) P6 (Kan-R) P7 (vacJ-F) P8 (vacJ-R) P9(Gm-F) P10(Gm-R)

CGCAAGCTTACCGCTTGTGCAGAAATTGTTGATTACTC GTCGACATGCTCGGATCCATGACTACTTTAAAAAAGGG GGATCCGAGCATGTCGACTGATTTACCTGTAAAAATTT CGGAATTCACAAGCGGTTCACTCGATTCTTTAATGCG CGGGATCCGTAAGGTTGGGAAGCCCTGC CGCGTCGACGGTCGGTCATTTCGAACCCC CGCGGATCCATGAAAAAAATTAAACTTCT CAACCTTACTTACCATAATCAATCAATATTT TTATGGTAAGTAAGGTTGCGAATTGACATAAG ACGTGTCGACGAAGCCGATCTCGGCTTGAAC

a

Restriction sites are underlined, uptake signal sequences (USS) are in italics.

44

L. Zhao et al. / Gene 603 (2017) 42–53

HS49, mutant strain ΔvacJ, and C-vacJ were grown to logarithmic phase. Two microliters of serial dilutions were spotted onto TSA plates containing 0.1% SDS and 0.5 mM EDTA. The plates were incubated overnight at 37 °C. 2.7. 1-N-phenylnaphthylamine (NPN) uptake assay NPN uptake assay was performed as described previously (Xie et al., 2016). Briefly, the H. parasuis HS49, ΔvacJ, and C-vacJ were grown to logarithmic phase. The bacteria cells were harvested by centrifugation and the pellets were resuspended in 5 mM HEPES buffer (pH 7.2). NPN (40 mM in HEPES buffer) was added to the bacterial suspension for a final concentration of 10 mM. Fluorescence was monitored using the EnVision Multilabel Reader (PerkinElmer, UK) set as follows: excitation at 350 nm, emission at 420 nm, and slit width of 2 nm. 2.8. Stress resistance assays

2.11. Serum bactericidal assay The serum bactericidal assay was performed as previously described (Zhang et al., 2012). Swine serum was prepared from pooled blood obtained from pigs with no history of H. parasuis infection which were filtered by the filtration membrane of 0.22 μm. A part of the sera were treated at 56 °C for 30 min to inactivate the complement. The H. parasuis stains HS49, ΔvacJ, and C-vacJ used in this assay were obtained from cultures grown until the mid-exponential phase. First, 20 μL bacterial cultures (approximately 1.0 × 107 CFU/mL) were mixed with 80 μL swine serum or heat-treated serum to achieve a final concentration of 80% serum. Then the mixtures were incubated at 37 °C for 1 h with gentle shaking. Subsequently, 10-fold serial dilutions of the bacteria were plated on TSA plates and incubated at 37 °C for 24 h. Bacteria were also incubated in PBS as a control. All experiments were performed in triplicate. 2.12. Adhesion and invasion assays

Stress resistance assays were conducted as described previously (Xie et al., 2013). The H. parasuis wild-type strain HS49, ΔvacJ, and C-vacJ were cultured to OD600 value of approximately 0.8, then the cell cultures were harvested by centrifugation at 4500g for 5 min. In the osmotic stress assay, bacterial cells were resuspended in 1 mL of TSB containing different concentrations of KCl and were incubated at 37 °C for 0.5 h, 1 h, 2 h, and 4 h, respectively. In the oxidative stress assay, bacterial cells were resuspended in 1 mL of TSB containing different concentrations of hydrogen peroxide (H2O2) and were incubated at 37 °C for 0.5 h, 1 h, 2 h, and 4 h, respectively. In the heat-shock assay, bacterial cells were resuspended in 1 mL of TSB and were incubated at 50 °C for 30 min. Bacterial cells of each strain were resuspended in TSB without any treatment as a control. All of the cultures of each stress assay were serially diluted and spread onto TSA plates for enumeration of CFU. Stress resistance values were calculated as [(stressed sample CFU/ mL) / (control sample CFU/mL)] × 100. Each assay was performed independently three times.

The adhesion and invasion assays were performed as previously described with some modifications (Vanier et al., 2006). Briefly, the porcine kidney epithelial cell line (PK-15) was grown as a monolayer in 24-well tissue culture plates. Bacteria were pelleted, washed three times with PBS, and resuspended at 1.0 × 107 CFU/mL in Dulbecco's Modified Eagle Medium (DMEM) (Invitrogen, USA). The monolayer cells were inoculated with 1 mL aliquots of bacterial suspension and incubated for 2 h at 37 °C with 5% CO2. Two hours later, the plates were washed five times with PBS and then 100 μL PBS containing 0.25% trypsin/0.02% EDTA was added to detach and lyse cells. The cell suspensions with adherent bacteria were serially diluted and plated on TSA plates to enumerate the viable bacteria. For the invasion assay, cells were treated with chloramphenicol (25 μg/mL) to kill non-invaded bacteria, prior to lysing the cells. All of the above experiments were performed in triplicate and replicated three times.

2.9. Antimicrobial susceptibility testing

2.13. Determination of ΔvacJ virulence in a mouse model

All determinations of susceptibility test were performed by the microdilution broth method according to the recommendations of the Clinical and Laboratory Standards Institute (Clinical and Laboratory Standards Institute, 2013). Minimal inhibitory concentration (MIC) determinations were performed using commercially sterile 96-well microtiter plates. A total of 10 antimicrobial agents were used as follows: ampicillin (0.06–64 μg/mL); amoxicillin (0.12–4096 μg/mL); cefotaxime (0.002–8 μg/mL); ceftiofur (0.004–16 μg/mL); cefaclor (0.5–512 μg/ mL); ciprofloxacin (0.002–64 μg/mL); levofloxacin (0.004–32 μg/mL); enrofloxacin (0.015–128 μg/mL); florfenicol (0.008–16 μg/mL); tilmicosin (0.12–256 μg/mL) (Zhou et al., 2010; Zhang et al., 2014).

The virulence of the wild-type strain HS49, mutant strain ΔvacJ, and C-vacJ were determined by 50% lethal dose (LD50) value, as previously described (Du et al., 2014). All animal experiments were carried out in strict accordance with the recommendations in the China Regulations for the Administration of Affairs Concerning Experimental Animals 1988. The protocol was approved by China Heilongjiang University of Chinese Medicine (permit number: SYXK(Hei) 2013-012).

2.10. Biofilm formation assays Biofilm formation assay was performed according to a previously described method (Jin et al., 2006). Briefly, bacterial cultures (10 μL) were inoculated into borosilicate glass tubes containing 1 mL TSB supplemented with 10% inactivated bovine serum and 0.01% NAD, which were incubated at 37 °C with circular agitation (150 rpm/min) for 16 h. Then the contents of tubes were removed with a pipette. The tubes were stained with 2 mL of 1% (W/V) Hucker crystal violet solution at room temperature for 5 min. The dyestuff was removed with a pipette and washed under flowing water for 5 min, and then dried at 37 °C for 1 h. Then 1 mL 33% (V/V) glacial acetic acid was added into each tube, and the OD630 was measured. TSB without bacterial cultures was designed as a negative control. All experiments were performed in triplicate.

2.13.1. Determination of the 50% lethal dose (LD50) Seven-week-old female Balb/C mice (purchased from Heilongjiang University of Chinese Medicine) were randomly divided into 16 groups (with six mice in each group). The wild-type strain HS49, ΔvacJ, and CvacJ were cultured at 37 °C until the logarithmic phase. The cultures were then centrifuged, and the cell pellets were resuspended in TSB. Experimental mice were injected intraperitoneally (i.p.) with 0.5 mL of the suspension of different strains at the following concentrations: 2.0 × 107 to 4.0 × 108 CFU (wild-type strain HS49 or C-vacJ), 2.0 × 108 to 4.0 × 109 CFU (mutant strain ΔvacJ). The control mice were inoculated only with TSB. The number of surviving mice was recorded 14 days after infection. LD50 value was calculated according to Karber's method. 2.13.2. Mouse survival and mortality studies A total of 45 seven-week-old female Balb/C mice were infected by i.p. injection with 0.5 mL of HS49, ΔvacJ or C-vacJ (approximately 4.0 × 108 CFU) in TSB, and mice infected with sterile TSB were used as controls. The mortality and clinical signs of infection were recorded daily post-infection over a 14-day period.

L. Zhao et al. / Gene 603 (2017) 42–53

2.13.3. Determination of viable bacteria in organs A total of 45 female Balb/C mice were assigned to three groups and used to assess the presence of viable bacteria in infected organs. Mice were inoculated by i.p. injection with 0.5 mL of HS49, ΔvacJ, or C-vacJ (4.0 × 10 7 CFU). Five mice infected with sterile TSB were used as controls. The tissue samples of kidney, liver, spleen, and lung were fragmented into small pieces and then homogenized with PBS as follow: 0.05 g/organ were placed in 500 μL of PBS and homogenized with a vortex. After serial 10-fold dilutions with TSB, 100 μL tissue mixtures of different dilutions were plated onto TSA plates and incubated at 37 °C overnight. The number of colonies were counted and expressed as CFU/0.05 g for organ samples.

2.14. Statistical analysis Experimental data were expressed as the mean ± SD. The difference between two groups was analyzed using the two-tailed Student's t-test and survival analysis was performed using the log rank test. P b 0.01 was considered highly statistical significant. P b 0.05 was considered statistical significant.

3. Results 3.1. Bioinformatic analysis of VacJ The VacJ of H. parasuis is encoded by a 747 bp ORF, which is a lipoprotein located in the outer membrane. The predicted characteristics of VacJ protein indicated hydrophilicityas, high antigenic index, and surface probability as shown by PROTEAN (Fig. 1A). VacJ possesses a short putative signal sequence and a characteristic lipobox sequence, Leu-Thr-Ala-Cys (LTAC) (Fig. 2B). Homology analysis of vacJ based on BLAST indicated that the vacJ sequence is highly conserved in currently reported H. parasuis isolates (99%). Multiple sequence alignments showed high homology to VacJ from other Gram-negative bacteria such as Actinobacillus pleuropneumoniae (99%), H. influenzae (55%), and S. flexneri (46%) (data not shown).

45

3.2. Construction of HS49ΔvacJ mutant strain and complemented strain To determine whether the function of the vacJ gene was crucial for H. parasuis, the vacJ deletion mutant of H. parasuis HS49 was constructed using the natural transformation system. The complemented strain construction method was transformation of H. parasuis with a complementation construct directs integration of a gene of interest into the chromosome. Two pairs of primers were used for identification of the vacJ deletion and complemented. The fragment containing upstream and downstream homologous arms with intermediate region was amplified using primers P1/P4. The wild-type strain produced a slightly smaller band because the sequence of vacJ is slightly shorter than the kan resistance cassette, while C-vacJ strain produced a larger band because it contains both two sequences of vacJ and gentamicin cassette. The primers P7/P8 amplified a 747 bp fragment in the wild-type HS49 and C-vacJ strains, while did not obtain a band from ΔvacJ strain (data not shown). These results indicated the successful generation of a vacJ gene deletion mutant and its complemented strain. 3.3. VacJ is required for growth and maintaining bacterial morphology of H. parasuis For examining the effect of vacJ deletion on growth, the OD600 of cultures of HS49, ΔvacJ, and C-vacJ strains in TSB at 37 °C were determined. As shown in Fig. 2A, the growth curve of the ΔvacJ mutant was clearly slower than that of the wild-type HS49 strain, whereas the growth phenotype was restored by introducing the vacJ gene on a shuttle plasmid. The viable cell counts of both the wild-type strain and the complemented strain C-vacJ were more than that of the ΔvacJ strain (Fig. 2B). Bacterial morphology change was observed through TEM after vacJ deletion. As shown in Fig. 2C, bacterial cells of the ΔvacJ deletion mutant showed lysed morphology and slightly swelled compared to the wild-type HS49. The change was restored by the complemented strain. The results indicated that loss of vacJ results in bacterial morphology change of H. parasuis. 3.4. Loss of vacJ affects the outer membrane integrity of H. parasuis In SDS-EDTA permeability assay, the wild-type HS49, ΔvacJ, and C-vacJ strains were diluted and spotted onto TSA plates containing

Fig. 1. Predicted characteristics of VacJ lipoprotein of H. parasuis. (A) Hydrophilicity plot as per Kyte-Doolittle, antigenic index as per Jameson-Wolf, and surface probability plot as per Emini. (B) Sequence for fragment of VacJ that include the signal sequence and lipobox.

46

L. Zhao et al. / Gene 603 (2017) 42–53

Fig. 2. (A) Growth of the wild-type HS49, ΔvacJ, and C-vacJ strains. Cultures were grown in TSB broth containing serum and NAD overnight at 37 °C and were subcultured 1:100 into fresh medium. The OD600 was measured at 1 h intervals. (B) The viable cell counts of the wild-type HS49, ΔvacJ, and C-vacJ strains in log CFU/mL. (C) Bacterial morphology analysis in the logarithmic phase by TEM.

SDS and EDTA to determine the relative sensitivity (Fig. 3A). Bacterial cultures that did not contain SDS or EDTA were spotted onto TSA plates as control (Fig. 3B). The results showed that the ΔvacJ mutant was more sensitive to SDS-EDTA than the wildtype and complemented strains. Furthermore, the NPN uptake assay was carried out to examine the outer membrane integrity of each strain. Compare to the HS49 and C-vacJ strains, the final fluorescence level of ΔvacJ was significantly higher (P b 0.05) (Fig. 3C). Taken together, these observations indicated that the vacJ gene is essential for outer membrane integrity of H. parasuis.

3.5. VacJ is required for the stress tolerance of H. parasuis The wild-type HS49, ΔvacJ, and C-vacJ strains were cultured under a variety of stress conditions. When the bacteria cells were treated with different concentrations of salt for 1 h, the ΔvacJ mutant survival rate was much lower than that of the HS49 or the C-vacJ (P b 0.05). Bacteria survival rates were obviously reduced with KCl concentration increasing (Fig. 4A) and the action time prolonging (data not shown). The results indicated that the deletion of the vacJ gene increased the sensitivity of H. parasuis to hyperosmotic shock which effects were dose and time dependent. Similar results were obtained in the hydrogen

L. Zhao et al. / Gene 603 (2017) 42–53

47

Fig. 3. (A and B) SDS-EDTA permeability assay. Cultures were grown to mid-log phase, and 2 μL of each dilution was spotted onto TSA plates. (A) TSA supplemented with 0.1% SDS and 0.5 mM EDTA. (B) TSA medium. (C) Uptake of NPN occurs in bacterial cells suspensions. The hydrophobic probe NPN fluorescence for the HS49, ΔvacJ, and C-vacJ strains, * denotes a P-value (t-test) b 0.05 following comparison with the HS49.

peroxide assay (Fig. 4B). However, when the cells incubated at 50 °C for 30 min, the survival rate of the ΔvacJ mutant was approximately equal to that of HS49 and C-vacJ (Fig. 4C). These results indicated that the deletion of the vacJ gene impairs the ability of H. parasuis to successfully respond to osmotic stress and oxidative but not to thermal stress.

3.6. The vacJ gene deletion affects antibiotics sensitivity of H. parasuis To determine whether the vacJ gene deletion mutant exhibited altered outer membrane properties, the sensitivity of the HS49, ΔvacJ, and C-vacJ strains to antibiotics was analyzed. As shown in Table 3, in comparison to the wild-type strain, the ΔvacJ mutant showed 4- to 8-fold decreases in MIC values for cefotaxime, cefaclor, levofloxacin, and tilmicosin, and there were just a slight decrease in MIC values for enrofloxacin and florfenicol. The complemented strain could restore the phenotypes of resistance to antibiotics. However, no change in sensitivity to ampicillin,

amoxicillin, ceftiofur or ciprofloxacin was observed in the ΔvacJ mutant compared to HS49. 3.7. Biofilm formation assay The ability of H. parasuis to form biofilms at the air-liquid interface in glass tubes and stained with Hucker crystal violet solution was determined for HS49, ΔvacJ, and C-vacJ. As shown in Fig. 4D, the biofilm formation phenotypes of the HS49 and C-vacJ strains were stronger, while the ΔvacJ mutant exhibited poor biofilm formation. Furthermore, biofilm formation was quantitatively measured using a microplate reader. The value of HS49 and C-vacJ strains were significantly (P b 0.01) higher than that of ΔvacJ strain (Fig. 4E). The results revealed that the vacJ mutant significantly weakens the ability to form biofilm. 3.8. Serum bactericidal assay In order to determine whether vacJ was involved in serum resistance, the serum bactericidal assay was performed. Compared

48

L. Zhao et al. / Gene 603 (2017) 42–53

to the wild-type HS49 strain, loss of vacJ gene resulted in significantly increased sensitivity to serum killing (P b 0.01) in 80% serum (Fig. 4F). Complementation of the ΔvacJ mutant restored the serum-resistant phenotype. The results indicated that vacJ is important for H. parasuis survival in swine serum.

3.9. Adhesion and invasion to PK-15 cells To evaluate whether the vacJ was involved in the ability of H. parasuis to interact with host cells, PK-15 cell was utilized to compare the adherence and invasion abilities in the wild-type HS49, ΔvacJ, and

L. Zhao et al. / Gene 603 (2017) 42–53

49

Fig. 4. (A) Analysis of the stress tolerance of the wild-type HS49, ΔvacJ, and C-vacJ strains. Overnight cultures were washed in sterile PBS and diluted to an OD600 value of 0.8. Bacterial cells were then treated with different concentrations of KCl or H2O2 for 1 h, or incubation at 50 °C for 30 min. (B) Biofilm formation of the wild-type HS49, ΔvacJ, and C-vacJ strains at the airliquid interface of cultures grown in glass tubes. The tubes were stained with 1% Hucker crystal violet. (C) Quantitative analysis of the biofilm formation in the HS49, ΔvacJ, and C-vacJ strains. (D) Survival of H. parasuis strains in 80% porcine sera. The ΔvacJ mutant strain exhibited significantly increased sensitivity to serum killing compared with the wild-type HS49 strain (P b 0.01). The complemented strain restored the serum-resistant phenotype. Error bars represent the standard deviation from three independent experiments. * denotes a Pvalue (t-test) b 0.05 following comparison with the ΔvacJ. ** denotes a P-value (t-test) of P b 0.01 following comparison with the ΔvacJ.

C-vacJ strains. As shown in Fig. 5A, the ΔvacJ mutant exhibited significantly decreased levels of cell binding after 2 h of infection, compared to those of the wild-type and C-vacJ strains. Similarly, it also showed a significantly poor capacity to invade host cells in chloramphenicol protection assays (Fig. 5B). These results confirmed that the vacJ of the HS49 strain was involved in H. parasuis-specific adhesion to and invasion of PK-15 cells. 3.10. Role of ΔvacJ in pathogenesis of H. parasuis The effect of the vacJ gene on virulence was evaluated using the Balb/ C mouse model by i.p. with various doses of HS49, ΔvacJ, and C-vacJ. The mortality of mice was observed within 14 days after the challenge. As shown in Table 4, the LD50 of the mutant strain (1.90 × 109 CFU) was increased compared with the wild-type strain HS49 (1.30 × 108 CFU), indicating that vacJ deletion significantly reduced the virulence of H. parasuis serovar 5. The virulence of the complemented mutant strain C-vacJ (1.67 × 108 CFU) was restored. For pathogenicity experiment, the survival rates were significantly lower in mice infected with the

HS49 or C-vacJ strain than in those infected with the ΔvacJ strain (Fig. 6A). During 24 h after inoculation, nearly half of the mice in the HS49 or C-vacJ-infected group were dead. The surviving mice of these two groups developed clinical signs including depression, rough coat, tremble, and prostration, and these mice died within 7 days after infection. During the entire period of monitoring, the mice infected with the ΔvacJ mutant showed no mortality. The virulence attenuation of the ΔvacJ was further investigated in vivo colonization experiments. Based on the results of the LD50 assessment, mice were i.p. inoculated with approximately 4.0 × 107 CFU of HS49, the ΔvacJ mutant or C-vacJ. Bacterial counts from each specific tissue of ΔvacJ-infected mice were significantly decreased compared with those of HS49 or C-vacJ-infected mice (Fig. 6B). These results suggested that the vacJ deletion attenuates the pathogenicity of H. parasuis. 4. Discussion H. parasuis is an important swine pathogen responsible for a diverse group of diseases. The mechanism of H. parasuis infection pathogenesis

50

L. Zhao et al. / Gene 603 (2017) 42–53

Table 3 Minimum inhibitory concentrations (MIC) for H. parasuis strains HS49, ΔvacJ, and C-vacJ. Antibiotics

Ampicillin Amoxicillin Cefotaxime Ceftiofur Cefaclor Ciprofloxacin Levofloxacin Enrofloxacin Florfenicol Tilmicosin

MIC (μg/mL) HS49

ΔvacJ

C-vacJ

16 4 8 2 4 2 16 32 16 16

16 4 2 2 0.5 2 4 16 8 4

16 4 8 2 4 2 16 32 16 16

was not well understood, because of limited information regarding its virulence factors. Outer membrane lipoprotein genes are universally distributed in bacteria and constitute approximately 1–3% of their total genes (Sutcliffe et al., 2012). The VacJ lipoprotein (also known as MlaA) is part of a complex system of phospholipids retrograde recycling from the bacterial outer membrane to the cytoplasm, involved in

Fig. 5. Adherence (A) and invasion (B) of wild-type HS49, ΔvacJ, and C-vacJ stains in PK-15 cells. The data represent the number of bacteria that adhered to or invaded the cells in each well of a 24-well plate. Error bars represent the standard deviation from three independent experiments performed in triplicate. **P b 0.01.

Table 4 Virulence of the wild-type, the mutant ΔvacJ, and C-vacJ strains in mice. Strains

Challenge dose (CFU)

Wild-type (HS49)

2.0 4.2 8.9 1.9 4.0 2.0 4.2 8.9 1.9 4.0 2.0 4.2 8.9 1.9 4.0

Mutant (ΔvacJ)

C-vacJ

a

× × × × × × × × × × × × × × ×

107 107 107 108 108 108 108 108 109 109 107 107 107 108 108

Number dead 0 1 2 3 6 0 1 1 2 6 0 0 1 3 6

Value of LD50 (CFU)

Fold attenuationa

1.30 × 108

1

1.90 × 109

15

1.67 × 108

1

Fold attenuation normalised to the wild-type strain.

maintaining the lipid asymmetry (Malinverni and Silhavy, 2009). A originally study by Suzuki et al. (1994) has proved that VacJ is a surface-exposed lipoprotein of S. flexneri. The latest research by Shivachandra et al. (2014) further confirmed that various pathogenic bacterial strains belonging to the family-Pasteurellaceae have several surface exposed virulence factors including VacJ and VacJ-like lipoproteins. We also used bioinformatic analysis to predict characteristics of VacJ lipoprotein. We discovered that the vacJ gene is high conserved in several H. parasuis serovars. Recent research has revealed new insights into the role of the VacJ lipoprotein in several virulence-associated properties of A. pleuropneumoniae (Xie et al., 2016). In our present study, the vacJ gene-deletion mutant HS49ΔvacJ and its complemented strain were constructed to investigate the role of vacJ in the pathogenicity of H. parasuis. Bacterial outer membrane integrity is associated with the morphology of the bacterial colonies and the osmotic stress response. Several previous studies have already demonstrated that VacJ lipoprotein has a common role in maintaining outer membrane integrity (Nakamura et al., 2011; Carpenter et al., 2014; Xie et al., 2016). In this study, we assayed the bacterial colony morphology. The deletion of vacJ gene resulted in significant cell morphological variation. Further, we analyzed the role of vacJ in H. parasuis stress tolerance by examining three strains survival under a variety of stress conditions. The results showed that the vacJ gene deletion caused more vulnerable to SDS plus EDTA, KCl, and H2O2, which demonstrated that VacJ is essential for stress tolerance. Our observation of increased NPN fluorescence in the ΔvacJ mutant further substantiate above findings. The outer membrane of Gram-negative bacteria is also one of the most important structures for resistance to antibiotics (Chung et al., 2007). Our results suggested that loss of VacJ in the H. parasuis HS49 strain showed increased sensitive to the some of cephalosporins and fluoroquinolones. The targets for cephalosporins are penicillin binding proteins on the cytoplasmic membrane (Murphy et al., 2006). Fluoroquinolones act by inhibiting the action of target enzymes, DNA gyrase and topoisomerase IV, with both enzymes playing a principal role in DNA replication. We speculate that the loss of vacJ increases the accessibility of the targets for these agents by alteration of the outer membrane permeability and integrity, accounting for the increased antibiotic susceptibility. Many bacteria have the ability to form biofilms, which play a key role in causing persistent infections and the ability of bacterial adaption to environmental changes (Hall-Stoodley et al., 2004; Joo and Otto, 2012; Bove et al., 2012). Most serovars of H. parasuis strains can form biofilm in vitro, which is essential for long-term colonization of the upper respiratory tract by H. parasuis (Jin et al., 2006). The expression of homologues of genes with putative functions in biofilm formation (siaB, htrA, fumC, gcpE and acrR) has been detected during pulmonary infection (Jin et al., 2008). We found that the VacJ lipoprotein was implicated

L. Zhao et al. / Gene 603 (2017) 42–53

51

Fig. 6. Determination of the wild-type HS49, ΔvacJ, and C-vacJ virulence in a mouse model. (A) Survival curves for Balb/C mice infected with the wild-type HS49, ΔvacJ mutant, and C-vacJ strains. (B) Bacterial distribution in kidney, liver, spleen, and lung from mice infected i.p. the wild-type HS49, ΔvacJ mutant, and C-vacJ strains.

in the biofilm formation of H. parasuis. The similar result was found in another member of Pasteurellaceae (Xie et al., 2016). Several lipoproteins have also been shown to be involved in bacterial biofilm formation, such as the SciN and SslE lipoprotein of enteroaggregative E. coli (Aschtgen et al., 2008; Baldi et al., 2012). Serum resistance represents an important virulence strategy of bacterial pathogens. Serum-resistance may play an important role in the virulence of H. parasuis. As the previous study revealed that once

virulent H. parasuis enter the bloodstream, the bacterium is able to avoid complement-mediated killing in an antibody-independent manner. Serum resistant H. parasuis strains are better able to resist the bactericidal effect of serum and thus are able to reach systemic sites, causing polyserositis and polyarthritis (Cerdà-Cuéllar and Aragon, 2008). In the current study, we also compared the survival of the ΔvacJ mutant and wild-type strains in swine serum. The results indicated that the ΔvacJ mutant was much more sensitive to swine serum than

52

L. Zhao et al. / Gene 603 (2017) 42–53

the wild-type strain. Similar result has been reported in A. pleuropneumoniae for lipoprotein VacJ mutant (Xie et al., 2016). Further investigations are necessary to determine which complement molecules were inhibited and which complement pathway was suppressed by VacJ. Adherence of bacteria to host cells surfaces is an essential determinant for bacterial colonization and cellular invasion, which contribute to the breaching of the cell barrier, persistence in the host, ultimately leading to systemic disease (Vahle et al., 1997). Adhesion and invasion of host cells in H. parasuis are also considered as important factors to pathogenesis. The virulent H. parasuis strains have been confirmed be able to breach the nasal mucosal barrier and get into the bloodstream (Vanier et al., 2006). It has also been confirmed that the systemic isolates of H. parasuis can adhere to and invade host cells and may be important in the first step of infection (Frandoloso et al., 2012; Zhang et al., 2013). In this study, compared to the wild-type HS49 strain, the ΔvacJ mutant strain demonstrated significantly decreased adherence and invasion abilities of PK-15 cells. Based on the result of the bacteria growth, the curve of ΔvacJ mutant was clearly slower than that of the wild-type HS49 strain, which seems to affect the result of cell adhesion and invasion to host cells. We speculate that the difference of cell adhesion might due to the fact that the cell growth is slower. To our knowledge, this study is the first to show the role of the VacJ lipoprotein in bacterial adhesion to and invasion of host cells. The lipoprotein VacJ was previously identified as a virulence determinant contributing to intracellular survival by S. flexneri (Suzuki et al., 1994). All the changes in the ΔvacJ mutant described above ultimately caused the attenuation of virulence. To further evaluate the role of vacJ in the pathogenesis of H. parasuis in vivo, a murine infection model was used. We demonstrated that the virulence of the vacJ-deficient strain was attenuated. The LD50 value of the mutant strain was fifteen-fold higher compared with the wild-type HS49, indicating that the absence of vacJ has a significant effect on bacterial virulence. The bacterial counts in different tissues of ΔvacJ-infected mice were significantly lower than those of HS49-infected mice. This suggested that vacJ is required during colonization and bacterial survival in vivo. In summary, we constructed the vacJ deletion mutant and its complemented strain from H. parasuis serovar 5, and preliminary investigated the effects of vacJ gene on several virulence-associated properties. The present results suggest that VacJ lipoprotein involves in maintaining the outer membrane integrity the formation of biofilms, adhesion and invasion, and resistance to complement killing. Compared with the wild-type strain, the ΔvacJ mutant has lower virulence in mice and shows reduced abilities to colonize the tissues, which indicating that vacJ may play a critical role in the virulence of H. parasuis. Our study provides new insights into the role of the VacJ lipoprotein in the pathogenesis of H. parasuis infection. Conflicts of interest The authors have no conflict of interest to declare. Acknowledgments This study was supported by the National Natural Science Foundation of China (No. 31472169). References Aschtgen, M.S., Bernard, C.S., De Bentzmann, S., Lloubes, R., Cascales, E., 2008. SciN is an outer membrane lipoprotein required for type VI secretion in enteroaggregative Escherichia coli. J. Bacteriol. 190, 7523–7531. Babu, M.M., Priya, M.L., Selvan, A.T., Madera, M., Gough, J., Aravind, L., Sankaran, K., 2006. A database of bacterial lipoproteins (DOLOP) with functional assignments to predicted lipoproteins. J. Bacteriol. 188, 2761–2773. Baldi, D.L., Higginson, E.E., Hocking, D.M., Praszkier, J., Cavaliere, R., James, C.E., BennettWood, V., Azzopardi, K.I., Turnbull, L., Lithgow, T., Robins-Browne, R.M.,

Whitchurch, C.B., Tauschek, M., 2012. The type II secretion system and its ubiquitous lipoprotein substrate, SslE, are required for biofilm formation and virulence of enteropathogenic Escherichia coli. Infect. Immun. 80, 2042–2052. Blackall, P.J., Turni, C., 2013. Understanding the virulence of Haemophilus parasuis. Vet. J. 198, 549–550. Bove, P., Capozzi, V., Garofalo, C., Rieu, A., Spano, G., Fiocco, D., 2012. Inactivation of the ftsH gene of Lactobacillus plantarum WCFS1: effects on growth, stress tolerance, cell surface properties and biofilm formation. Microbiol. Res. 167, 187–193. Carpenter, C.D., Cooley, B.J., Needham, B.D., Fisher, C.R., Trent, M.S., Gordon, V., Payne, S.M., 2014. The Vps/VacJ ABC transporter is required for intercellular spread of Shigella flexneri. Infect. Immun. 82, 660–669. Cerdà-Cuéllar, M., Aragon, V., 2008. Serum-resistance in Haemophilus parasuis is associated with systemic disease in swine. Vet. J. 175, 384–389. Cerdà-Cuéllar, M., Naranjo, J.F., Verge, A., Nofrarias, M., Cortey, M., Olvera, A., Aragon, V., 2010. Sow vaccination modulates the colonization of piglets by Haemophilus parasuis. Vet. Microbiol. 14, 315–320. Chen, L., Wu, D., Cai, X., Guo, F., Blackall, P.J., Xu, X., Chen, H., 2012. Electrotransformation of Haemophilus parasuis with in vitro modified DNA based on a novel shuttle vector. Vet. Microbiol. 155, 310–316. Chou, K.C., Shen, H.B., 2008. Cell-ploc: a package of web servers for predicting subcellular localization of proteins in various organisms. Nat. Protoc. 3, 153–162. Chung, J.W., Ng-Thow-Hing, C., Budman, L.I., Gibbs, B.F., Nash, J.H., Jacques, M., Coulton, J.W., 2007. Outer membrane proteome of Actinobacillus pleuropneumoniae: LC-MS/ MS analyses validate in silicopredictions. Proteomics 7, 1854–1865. Clinical and Laboratory Standards Institute (CLSI), 2013. Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals. Clinical and Laboratory Standards Institute, Wayne, PA, USA, CLSI Documents VET01-A4 and VET01-S2. Costa-Hurtado, M., Aragon, V., 2013. Advances in the quest for virulence factors of Haemophilus parasuis. Vet. J. 198, 571–576. Du, B., Ji, W., An, H., Shi, Y., Huang, Q., Cheng, Y., Fu, Q., Wang, H., Yan, Y., Sun, J., 2014. Functional analysis of c-di-AMP phosphodiesterase, GdpP, in Streptococcus suis serotype 2. Microbiol. Res. 169, 749–758. Frandoloso, R., Martinez-Martinez, S., Gutierrez-Martin, C.B., Rodriguez-Ferri, E.F., 2012. Haemophilus parasuis serovar 5 Nagasaki strain adheres and invades PK-15 cells. Vet. Microbiol. 154, 347–352. Guzman, L.M., Belin, D., Carson, M.J., Beckwith, J., 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177, 4121–4130. Hall-Stoodley, L., Costerton, J.W., Stoodley, P., 2004. Bacterial biofilms: from the natural environment to infectious diseases. Nat. Rev. Microbiol. 2, 95–108. Jin, H., Zhou, R., Kang, M., Luo, R., Cai, X., Chen, H., 2006. Biofilm formation by field isolates and reference strains of Haemophilus parasuis. Vet. Microbiol. 118, 117–123. Jin, H., Wan, Y., Zhou, R., Li, L., Luo, R., Zhang, S., Hu, J., Langford, P.R., Chen, H., 2008. Identification of genes transcribed by Haemophilus parasuis in necrotic porcine lung through the selective capture of transcribed sequences (SCOTS). Environ. Microbiol. 10, 3326–3336. Joo, H.S., Otto, M., 2012. Molecular basis of in vivo biofilm formation by bacterial pathogens. Chem. Biol. 19, 1503–1513. Kielstein, P., Rapp-Gabrielson, V.J., 1992. Designation of 15 serovars of Haemophilus parasuis on the basis of immunodiffusion using heat-stable antigen extracts. J. Clin. Microbiol. 30, 862–865. Malinverni, J.C., Silhavy, T.J., 2009. An ABC transport system that maintains lipid asymmetry in the gram-negative outer membrane. Proc. Natl. Acad. Sci. U. S. A. 106, 8009–8014. Møller, K., Kilian, M., 1990. V factor-dependent members of the family Pasteurellaceae in the porcine upper respiratory tract. J. Clin. Microbiol. 28, 2711–2716. Murphy, T.F., Kirkham, C., Lesse, A,J., 2006. Construction of a mutant and characterization of the role of the vaccine antigen P6 in outer membrane integrity of nontypeable Haemophilus influenzae. Infect. Immun. 74, 5169–5176. Nakamura, S., Shchepetov, M., Dalia, A.B., Clark, S.E., Murphy, T.F., Sethi, S., Gilsdorf, J.R., Smith, A.L., Weiser, J.N., 2011. Molecular basis of increased serum resistance among pulmonary isolates of non-typeable Haemophilus influenzae. PLoS Pathog. 7, e1001247. Oliveira, S., Pijoan, C., 2004. Haemophilus parasuis: new trends on diagnosis, epidemiology and control. Vet. Microbiol. 99, 1–12. Olvera, A., Segales, J., Aragon, V., 2007. Update on the diagnosis of Haemophilus parasuis infection in pigs and novel genotyping methods. Vet. J. 174, 522–529. Saeed-Kothe, A., Yang, W., Mills, S.D., 2004. Use of the riboflavin synthase gene (ribC) as a model for development of an essential gene disruption and complementation system for Haemophilus influenzae. Appl. Environ. Microbiol. 70, 4136–4143. Schafer, A., Tauch, A., Jager, W., Kalinowski, J., Thierbach, G., Puhler, A., 1994. Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 69–73. Shivachandra, S.B., Kumar, A., Mohanty, N.N., Yogisharadhya, R., Chacko, N., Viswas, K.N., Ramakrishnan, M.A., 2014. Homogeneity of VacJ outer membrane lipoproteins among Pasteurella multocida strains and heterogeneity among members of Pasteurellaceae. Res. Vet. Sci. 96, 415–421. Sutcliffe, I.C., Harrington, D.J., Hutchings, M.I., 2012. A phylum level analysis reveals lipoprotein biosynthesis to be a fundamental property of bacteria. Protein. Cell 3, 163–170. Suzuki, T., Murai, T., Fukuda, I., Tobe, T., Yoshikawa, M., Sasakawa, C., 1994. Identification and characterization of a chromosomal virulence gene, vacJ, required for intercellular spreading of Shigella flexneri. Mol. Microbiol. 11, 31–41.

L. Zhao et al. / Gene 603 (2017) 42–53 Vahle, J.L., Haynes, J.S., Andrews, J.J., 1997. Interaction of Haemophilus parasuis with nasal and tracheal mucosa following intranasal inoculation of cesarean derived colostrum deprived (CDCD) swine. Can. J. Vet. Res. 61, 200–206. Vanier, G., Szczotka, A., Friedl, P., Lacouture, S., Jacques, M., Gottschalk, M., 2006. Haemophilus parasuis invades porcine brain microvascular endothelial cells. Microbiology 152, 135–142. Xie, F., Zhang, Y., Li, G., Zhou, L., Liu, S., Wang, C., 2013. The ClpP protease is required for the stress tolerance and biofilm formation in Actinobacillus pleuropneumoniae. PLoS One 8, e53600. Xie, F., Li, G., Zhang, W., Zhang, Y., Zhou, L., Liu, S., Liu, S., Wang, C., 2016. Outer membrane lipoprotein VacJ is required for the membrane integrity, serum resistance and biofilm formation of Actinobacillus pleuropneumoniae. Vet. Microbiol. 183, 1–8. Yamanaka, H.I., Kirii, Y., Ohmoto, H., 1995. An improved phage display antibody cloning system using newly designed PCR primers optimized for Pfu DNA polymerase. J. Biochem. 117, 1218–1227. Yue, M., Yang, F., Yang, J., Bei, W., Cai, X., Chen, L., Dong, J., Zhou, R., Jin, M., Jin, Q., Chen, H.C., 2009. Complete genome sequence of Haemophilus parasuis SH0165. J. Bacteriol. 191, 1359–1360.

53

Zhang, B., He, Y., Xu, C., Xu, L., Feng, S., Liao, M., Ren, T., 2011. Cytolethal distending toxin (CDT) of the Haemophilus parasuis SC096 strain contributes to serum resistance and adherence to and invasion of PK-15 and PUVEC cells. Vet. Microbiol. 157, 237–242. Zhang, B., Feng, S., Xu, C., Zhou, S., He, Y., Zhang, L., Zhang, J., Guo, L., Liao, M., 2012. Serum resistance in Haemophilus parasuis SC096 strain requires outer membrane protein P2 expression. FEMS Microbiol. Lett. 326, 109–115. Zhang, B., Xu, C., Zhang, L., Zhou, S., Feng, S., He, Y., Liao, M., 2013. Enhanced adherence to and invasion of PUVEC and PK-15 cells due to the overexpression of RfaD, ThyA and Mipin the ΔompP2 mutant of Haemophilus parasuis SC096 strain. Vet. Microbiol. 162, 713–723. Zhang, J., Xu, C., Shen, H., Li, J., Guo, L., Cao, G., Feng, S., Liao, M., 2014. Biofilm formation in Haemophilus parasuis: relationship with antibiotic resistance, serotype and genetic typing. Res. Vet. Sci. 97, 171–175. Zhou, X., Xu, X., Zhao, Y., Chen, P., Zhang, X., Chen, H., Cai, X., 2010. Distribution of antimicrobial resistance among different serovars of Haemophilus parasuis isolates. Vet. Microbiol. 141, 168–173.