Genome-wide mining of potential virulence-associated genes in Riemerella anatipestifer using random transposon mutagenesis

Veterinary Microbiology 189 (2016) 52–58

Contents lists available at ScienceDirect

Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic

Genome-wide mining of potential virulence-associated genes in Riemerella anatipestifer using random transposon mutagenesis Xintao Nia,1, Pan Jianga,1, Linlin Xinga , Changcan Oua , Hui Yua , Jingjing Qia , Bingqing Suna , Junsheng Cuia , Guijun Wangb,* , Qinghai Hua,* a b

Shanghai Veterinary Research Institute, the Chinese Academy of Agricultural Sciences, 518 Ziyue Road, Shanghai 200241, China Anhui Agricultural University, College of Animal Science and Technology, 130 West Changjiang Road, Hefei 230036, China

A R T I C L E I N F O

Article history: Received 1 July 2015 Received in revised form 10 April 2016 Accepted 18 April 2016 Keywords: Riemerella anatipestifer Random transposon mutagenesis Virulence Blood bacterial loading Insertion mutation

A B S T R A C T

Riemerella anatipestifer infection is a severe disease confronting the duck industry worldwide. However, little is known about the molecular basis of R. anatipestifer pathogenesis. In this study, we screened 3580 transposon Tn4351 insertion mutagenesis mutants of the highly virulent strain YZb1 in a duckling infection experiment and found 29 of them to be attenuated and 28 potential virulence-associated genes were identified. Molecular characterization of transposon insertion sites showed that of the 28 screened genes, two were predicted to encode TonB-dependent outer membrane receptor (plugs), sixteen encoded enzymes, and seven encoded hypothetical proteins. In addition, of the 28 affected genes, 19 were only found in bacteria belonging to the phylum Bacteroidetes and 10 were only found in the family Flavobacteriaceae. The median lethal dose of the mutants M11 and M29, which was affected in Riean_0060 and Riean_1537 respectively, were about 1700-fold and 210-fold higher than that of the wildtype strain YZb1, and those of the complemented strains M11(pRES-Riean_0060) and M29(pRESRiean_1537) were decreased by 25- and 3-fold respectively compared to those of the mutants M11 and M29. Additional analysis indicated that the blood bacterial loading of ducklings infected with M11 or M29 was decreased significantly, as compared with that in ducklings infected with the wild-type strain YZb1. Thus, our results indicate that Riean_0060 and Riean_1537 were involved in R. anatipestifer pathogenesis. ã 2016 Elsevier B.V. All rights reserved.

1. Introduction Riemerella anatipestifer infection is a contagious disease of domestic ducks, geese, turkeys, and various other domestic and wild birds, which causes major economic losses, especially to the duck industry, worldwide as a result of high mortality, weight loss, condemnations, downgrading, and salvage. However, research into the molecular basis of the pathogenesis of R. anatipestifer infection has been limited and only a few virulence factors have been identified so far, which include OmpA (Hu et al., 2011), SIP (Lu et al., 2013), and TbdR1 (Tu et al., 2014). To date, five full R. anatipestifer genomic sequences have been submitted to the GenBank database (Mavromatis et al., 2011; Wang et al., 2012; Yuan et al., 2013; Yuan et al., 2011). While previous genomic analysis has identified several

* Corresponding authors. E-mail addresses: [email protected] (G. Wang), [email protected] (Q. Hu). 1 These two authors equally contributed to this work. http://dx.doi.org/10.1016/j.vetmic.2016.04.014 0378-1135/ã 2016 Elsevier B.V. All rights reserved.

suspected virulence-associated proteins (e.g., lipopolysaccharide biosynthesis-related proteins), R. anatipestifer was found to lack classical virulence-related sequences and genes, such as those associated with the type III secretion system, and pilus or flagella biogenesis. Therefore, to fully elucidate the molecular basis of R. anatipestifer pathogenesis, unbiased approaches are needed. Random transposon mutagenesis, which allows screening for attenuated mutants in an animal model of infection at the genomewide level, is one of the most powerful technologies for mining of potential virulence-associated genes. The use of this technology has led to the identification of new virulence genes in several pathogenic bacteria. In our previous study, 33 genes involved in biofilm formation by R. anatipestifer CH3 were identified from a library of 2520 random Tn4351 transposon mutants, which suggested that Tn4351 worked well in the construction of a transposon insertion mutant library in R. anatipestifer (Hu et al., 2012a). In the present study, we screened 3580 Tn4351 insertion mutants of the highly virulent R. anatipestifer strain YZb1 of a duckling infection experiment and identified 29 attenuated

X. Ni et al. / Veterinary Microbiology 189 (2016) 52–58

mutants, while bioinformatics analysis of the mutants revealed that 10 of these genes were only found in the family Flavobacteriaceae. 2. Materials and methods 2.1. Bacterial strains and culture conditions R. anatipestifer serotype 2 strain YZb1 was isolated from the brain of an ill Cherry Valley duck. Escherichia coli strain BW19851 (pEP4351), which carries the plasmid pEP4351, were generously provided by Professor Mark J. McBride of the University of Wisconsin—Milwaukee (Milwaukee, WI, USA). R. anatipestifer cells were cultured at 37  C in tryptic soybean broth (TSB; Difco Laboratories Inc., Detroit, MI, USA), while E. coli strains were routinely grown in Luria broth (LB; Difco Laboratories Inc.) or on LB agar at 37  C. For selective growth of bacterial strains, antibiotics were added at the following concentrations: ampicillin (100 mg/ ml), chloramphenicol (30 mg/ml), erythromycin (1 mg/ml), and kanamycin (50 mg/ml). 2.2. The median lethal dose (LD50) of strain YZb1 to ducklings One-day-old Cherry Valley ducklings were obtained from the Zhuanghang Duck Farm (Shanghai, P. R. China) and housed in cages under a controlled temperature of 28  C–30  C, under a 12 h light/ dark cycle, with free access to food and water during the study period. Animal experiments in this study were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Institutional Animal Care and Use Committee guidelines set by the Shanghai Veterinary Research Institute, the Chinese Academy of Agricultural Sciences (CAAS; Shanghai, China). The study protocol was approved by the Committee on the Ethics of Animal Experiments of the Shanghai Veterinary Research Institute, CAAS (permit no.: 13–19). To determine the LD50 of R. anatipestifer strain YZb1, groups of ten ducklings were infected intramuscularly (i.m.) with increasing inocula: 106, 107, 108, 109, and 1010 colony-forming units (CFU) per 500 ml inoculum. Ten ducklings were infected with phosphatebuffered saline (PBS) as a control group. Ducklings that became moribund were humanely killed and counted as dead. Dead ducklings were subjected to R. anatipestifer identification. The mortality of the ducklings was recorded daily for a period of 10 days after challenge and LD50 values were calculated using the Reed–Muench method (Reed and Muench, 1938).

53

anatipestifer identification. The mutants, which 1–3 infected ducklings were survived in the first screen, were tested in a pure culture to confirm the attenuation further. Three ducklings were infected i.m. with 3  109 CFU of mutant bacteria and compared to a control group of ducklings infected i.m. with the wild-type strain YZb1. Finally, the pathogenicity of the mutants, which 1–3 infected ducklings were survived in the second screen, was measured using six 8-day-old ducklings infected with 1 109 CFU of each bacteria respectively, and the control ducklings (10 ducklings per dosage) were infected with 0 (PBS only), 106, 107, 108 and 109 CFU of the wild type YZb1 respectively. The statistical significance of pathogenicity between each mutant and the wild-type YZb1 was assessed by a two-tailed Fischer’s exact test. Southern blot analysis of the Tn4351 insertions was used for the identification of the number of insertions of the attenuated mutants, according to methods previously described by (Hu et al., 2012a). The transposon-specific probe representing a 410 bp IS4351 fragment was amplified from the plasmid pEP4351 with primers IS4351 P1 plus IS4351 P2 (Table S1). The DIG DNA labeling and detection kit (Roche Diagnostics USA, Indianapolis, IN, USA) was used to prepare probes and perform hybridization. A mutant that was probed with transposon Tn4351 at one site was used to further identify the interrupted gene. 2.5. Identification of transposon insertion sites Genomic DNAs of the attenuated mutants were extracted using the MiniBEST Bacterial Genomic DNA Extraction Kit (TaKaRa, Dalian, China). The nucleotide sequence surrounding the transposon insertion site was determined using inverse PCR (Hu et al., 2012a), or arbitrary PCR (genome walking kit, Takara, Dalian, China; specific primers were listed in Table S1) as described elsewhere (Hu et al., 2012a; Ochman et al., 1988). Briefly, genomic DNA was digested with the restriction enzyme HindIII and then religated, which resulted in the formation of circular molecules. Primer pairs specific for Tn4351 (primers 340 and 341; primers TN1 and IS4351-F, respectively) (Alvarez et al., 2006) were used to amplify the sequences adjacent to the insertion site using the LA PCR kit (TaKaRa, Dalian, China). The sequences of the identified genes were used to search for other known homologous sequences and putative functions using the BLASTX server (http://www.ncbi. nlm.nih.gov/BLASTX/) and the online PSORT v.3.0 program (http:// www.psort.org/) was used to predict the subcellular localization of the proteins. 2.6. Complementation of the mutant strains

2.3. Generation of a R. anatipestifer strain YZb1 Tn4351 mutant library Transposon mutagenesis was performed as described previously (Hu et al., 2012a) with E. coli BW19851 (pEP4351) used as the donor strain and R. anatipestifer YZb1 as the recipient. After conjugation of the plasmid from E. coli Bw19851 into R. anatipestifer strain YZb1, transconjugants were selected by growth in the presence of erythromycin and kanamycin. Transposon Tn4351 insertion mutants were confirmed by polymerase chain reaction (PCR) amplification of the ermF and 16S rRNA genes (primers shown in Table S1). 2.4. Screening for attenuated mutants using ducklings For each mutant, three 8-day-old ducklings were infected by i. m. inoculation of bacterial suspension (500 ml, approximately 1 109 CFU). Characteristics of the infected ducklings were recorded daily for a period of 10 days. The ducklings that became moribund were humanely killed and then subjected to R.

To determine whether the virulence-attenuated phenotype was due to an inactivated gene, the attenuated mutant strains, mutants 11 and 29, in which the Riean_0060 and Riean_1537 genes were inactivated by the insertion of Tn4351, respectively, were used for a complementation experiment. An E. coli–R. anatipestifer shuttle plasmid pRES0 or pRES (Hu et al., 2013) was used to construct the complemented strains. The putative Riean_0060 cassette including its putative promoter (from 1 to 546 of the upstream sequences) and Riean_0060 open reading frame (ORF) was amplified and subcloned into the pRES0 vector to generate pRES-Riean_0060, while Riean_1537 ORF was amplified and subcloned into the pRES vector to generate pRES-Riean_1537. The expression of Riean_1537 in the complementing plasmid was under the control of the promoter pompA (Hu et al., 2013). For complementation analysis, the plasmid pRES-Riean_0060 or pRESRiean_1537 was introduced into the M11 (Riean_0060:Tn) or M29 (Riean_1537:Tn) strain by conjugation, to generate the complemented strains M1(pRES- Riean_0060) and M29(pRESRiean_1537), respectively.

54

X. Ni et al. / Veterinary Microbiology 189 (2016) 52–58

Table 1 Description of attenutaed Riemerella anatipestifer YZb1 mutants. Mutants

Pathogenicity to Ducklings

Description of genes

ORF in DSM15868

Putative conserved domain

Function group (COGs)c

CirA, OMP_RagA_SusC

COG1629

CirA, HMA

COG1629 P

K_trans

COG3158 P

YhhN

–d

Riean_1789 Riean_1700 Riean_0395

Outer Membrane Outer Membrane Cytoplasmic Membrane Cytoplasmic Membrane Cytoplasmic Cytoplasmic Cytoplasmic

COG3392, AdoMet_ MTases LysA, PLPDE_III superfamily DEADc, HELICc, PRK11192

COG3392 L COG0019 E COG0513LK J

Riean_0195

Cytoplasmic

Riean_1564

Cytoplasmic

RsuA, S4, PseudoU_ synth_ RsuA_ COG1187 J like RluA, S4, PseudoU_synth RluCD_like COG0564 J

Riean_1780

Cytoplasmic

MetH

COG0646 E

transaldolase

Riean_0060

Cytoplasmic

Transaldolase_FSA,

COG0176 G

0/6

excinuclease ABC, C subunit

Riean_1413

Cytoplasmic

uvrC, UvrC_HhH_N

COG0322 L

0/3

1/6

IMP dehydrogenase / GMP reductase

Riean_1474

Cytoplasmic

PRK05567

COG0516 F

1/3

1/3

3/6

prolyl oligopeptidase family protein

Riean_0403

Cytoplasmic

DAP2, Peptidase_S9,

COG1506 E

1/3

1/3

2/6

phosphoribosylamine-glycine ligase

Riean_0145

Cytoplasmic

GARS, PRK00885

COG0151 F

2/3

2/3

3/6

Riean_0483

Cytoplasmic

DltE

COG0300 R

1/3

1/3

3/6

3-oxoacyl-(acyl-carrier protein) reductase paralog aminoacyl-histidine dipeptidase

Riean_0533

Cytoplasmic

Peptidase_M20, ArgE, PRK08651

COG2195 E

1/3

2/3

3/6

cell division protein FtsA

Riean_1038

Cytoplasmic

COG0849 D

0/3

1/3

2/6

chorismate synthase (aroC)

Riean_0535

Cytoplasmic

NBD_sugar-kinase_HSP70; Cell_division_protein FtsA Chorismate_synthase

COG0082 E

1/3

2/3

3/6

(p)ppGpp synthetase I, SpoT/RelA

Riean_0227

Cytoplasmic

SPOT, HDc, ACT_RelA-Spot

COG0317 TK

1/3

1/3

3/6

Oligopeptidase B

Riean_1210

Unknown

PtrB, Peptidase_S9

COG1770 E

2/3

2/3

3/6

hypothetical protein

Riean_1020

Unknown

–

–

2/3

2/3

3/6

hypothetical protein

Riean_1201

Unknown

MecA_N

–

1/3

2/3

3/6

hypothetical protein

Riean_1098

Unknown

–

–

2/3

2/3

3/6

hypothetical protein

Riean_0211

Unknown

OMP_b-brl_2

–

2/3

2/3

3/6

hypothetical protein

M949_0169e Unknown

PTZ00121, SMC

–

1/3

2/3

3/6

hypothetical protein

Riean_1454

Unknown

–

COG1196D

1/3

2/3

4/6

hypothetical protein

Riean_1537

Cytoplasmic

–

–

1st Screen

2nd Screen

3rd Chall

mutant 1

0/3a1

2/3a2

1/6a3,* TonB-dependent receptor plug

Riean_1748

mutant 2

1/3

1/3

2/6

TonB-dependent receptor

Riean_1284

mutant 3

0/3

2/3

2/6

potassium transporter

Riean_1616

mutant 4

0/3

1/3

1/6

YhhN family protein

Riean_0217

mutant 5 mutant 6 mutant 7

1/3 0/3 1/3

2/3 1/3 1/3

3/6 1/6 2/6

mutant 8

0/3

0/3

0/6

mutant 9

1/3

1/3

2/6

mutant 10 mutant 11* mutant 12 mutant 13 mutant 14 mutant 15 mutant 16 mutant 17 mutant 18 mutant 19 mutant 20 mutant 21 mutant 22 mutant 23 mutant 24 mutant 25 mutant 26, 27 mutant 28 mutant 29*

0/3

0/3

0/6

Site-specific DNA-methyltransferase Orn/DAP/Arg decarboxylase 2 DEAD/DEAH box helicase domain protein ribosomal large subunit pseudouridine synthase B (RluB) ribosomal large subunit pseudouridine synthase D (RluD) homocysteine S-methyltransferase

0/3

1/3

2/6

0/3

0/3

0/3

a1

Gene products Subcellular locationb

P

For 1st screen, 3 ducklings were injected intramuscularly with 109 CFU of each mutant. For 2nd screen, 3 ducklings were injected intramuscularly with 3  109 CFU of each mutant which was screened out from 1st screening. a3 For 3rd challenge, 6 ducklings were injected intramuscularly with 109 CFU of each mutant, which was screened out by 2nd screening, to confirm consistent virulence defects. Deaths were recorded during 7 days postinoculation. Ten ducklings were all dead when they were challenged intramuscularly with 108 or 109 CFU of the wild type strain YZb1. a1–3 The ratio represents the number of dead ducklings out of total challenged ducklings. b Subcellular locations were predicted by the PSORTb v.3.0 server (http://www.psort.org/). c COG functional categories: (1) Information storage and processing: (J: Translation, ribosomal structure and biogenesis; K: Transcription; L: DNA replication, recombination and repair); (2) Cellular processes: (D: Cell division and chromosome partitioning; M: Cell wall/membrane/envelope biogenesis; P: Inorganic ion transport and metabolism; T: Signal transduction mechanisms); (3) Metabolism: (E: Amino acid transport and metabolism; F: Nucleotide transport and metabolism; G: Carbohydrate transport and metabolism; H. Coenzyme transport and metabolism). (4) Poorly characterized: (R: General function prediction only). d —: No related COG. e ORF not found on the genome R. anatipestifer DSM15868 (accession No: CP002346), but on that of strain CH3 (accession No: CP006649). * The LD50 value of mutant 1, mutant 11 and mutant 29 was 7.53  109. 2.20  109 and 1.86  108 respectively. a2

X. Ni et al. / Veterinary Microbiology 189 (2016) 52–58

2.7. Real-time PCR To determine whether Reian_0060 or Riean_1537 gene inactivation had a polar effect on transcription of adjacent genes, total RNA was isolated from the wild-type YZb1, mutants M11 and M29, and the complemented strains. Real-time PCR was performed to measure mRNA levels of Reian_0059/Riean_0060/Riean_0061, and Riean_1536/Riean_1537/Riean_1538. The primers were designed using Primer Express 3.0 software (Applied Biosystems, Inc.) and were listed in Table S1. Relative quantification of gene expression was calculated using the DCT method based on the signal intensity of the PCR products according to the following formula: 2 DCT = 2 (sampl Ct normalizerCt) (Ct = threshold cycle of real-time PCR). DnaB gene was used as endogenous control for sample normalization. Results were presented as fold changes relative to the mRNA expressions in the wild type YZb1. 2.8. Characteristics of mutants M11 and M29 The wild-type strain YZb1, the mutants M11 and M29, and their complemented strains, were grown in TSB at 37  C with shaking, and the growth curves of each bacteria were determined as described previously (Hu et al., 2002). The LD50 of those strains, to 8-day-old ducklings was determined as described for the wildtype strain YZb1 (Section 2.2). In addition, the blood bacterial loads of ducklings infected with the mutant strain M11 or M29, the corresponding complemented strain M11(pRES-Riean_0060) or M29(pRES_1537), and the wild-type strain YZb1, were compared as described previously (Hu et al., 2011). Briefly, 8-day-old Cherry Valley ducklings were injected intramuscularly with 108 (for M11, M11(pRES-Riean_0060) and YZb1), or 109 (for M29, M29 (pRES_1537) and YZb1) CFU of bacteria. The blood were collected at 12 h and 24 h post inoculation (six ducklings per bacterium at each time-point), and plated on TSB agar plates for bacteria counting. 3. Results 3.1. R. anatipestifer strain YZb1 is highly pathogenic to ducklings The LD50 for strain YZb1 was determined to evaluate its virulence. The results showed that the LD50 for the YZb1 was about 100–1000-fold greater than that of pathogenic strains CH3 and Th4 (5.71 105 vs. 6.48  107 and 4.41 108 CFU, respectively) (Hu et al., 2011; Hu et al., 2010). Therefore, in this study, we chose strain YZb1 to construct the Tn4351 insertion mutant library to screen for potential virulence-associated genes. 3.2. Screening of the attenuated mutants of strain YZb1 Biparental mating of E. coli 19851 carrying pEP4351 with R. anatipestifer strain YZb1 was performed to construct a library of 3580 mutants. For each mutant, three ducklings were infected with 1 109 CFU of bacteria in the first screen, and 61 mutants were screened out because 1–3 infected ducklings were survived. Of these, 39 mutants were screened out in the second screen using a higher inoculation dosage, but only 28 mutants had significant virulence defects (p < 0.05) when 6 ducklings were challenged with 109 CFU of per mutant(Table 1), compared with that of wild type YZb1, while the pathogenicity of the other one, mutant 29, was decreased about 25 fold than that of wild type YZb1 (Section 3.5). To determine whether these 29 mutants harbored unique Tn4351 insertions, genomic DNA was digested with a restriction enzyme and probed using the ermF cassette amplified from pEP4351. The results showed that the digestion of 29 mutants

55

yielded single restriction fragments, which further confirmed successful insertion of a single transposon element in each mutant (data not shown). The attenuating defects in the systemically impaired mutants may arise directly from the insertions themselves, or from the potential effects of these mutations on downstream gene expression. Therefore, complementation of the mutants or regeneration of the mutants by gene deletion, and other further experimentations are required to confirm whether or not the genes are related to the virulence of R. anatipestifer directly. 3.3. Molecular analysis of the attenuated mutants The insertion site of Tn4351 in the genomic DNA of the attenuated mutants was amplified either by inverse PCR or arbitrary PCR. Homology searches were conducted using the flanking sequences against R. anatipestifer genomic sequences. The function of each disrupted gene was predicted using genome annotation and the BLAST software for each coding sequence against a non-redundant protein sequence database. The amino acid sequences were analyzed using the CD-search algorithm (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) to detect conserved domains. The results of the bioinformatics searches are shown in Table 1. Putative conserved domains were detected in 24 of the 28 disrupted genes. A total of 7 mutants had transposon insertions within genes coding putative products of unknown function among which three had putative conserved domains. Nineteen of 28 affected genes are only found in bacteria belonging to the phylum Bacteroidetes. Of these, 10 genes were only found in the family Flavobacteriaceae, and none of the mutated genes were only found in R. anatipestifer. In addition, gene M949_0169, which was affected in the mutants 26/27, is not found in the type strain DSM15868. Further, in this study, only M949_0169 gene from mutants 26 and 27 was ‘hit' twice. The proteins encoded by the 21 genes identified in this study were grouped into four different functional classes, as shown in Table 1. Of these, six proteins (28.57%, 6/21) were classified to categories related to information storage and processing (J, K, L). Cellular processes and signaling related categories (D, O, M, N, P and T) included six (28.57%, 6/21) proteins. Nine proteins (42.86%, 9/21) were represented in metabolism-related categories (C, G, E, F, H, I and Q). Poorly characterized clusters of orthologous groups (COGs) (R and S) contained one (4.76%, 1/21) protein. In addition, the other 7 proteins did not belong to any of the currently defined COGs. The subcellular locations of 28 proteins were predicted using PSORTb v.3.0 software. Of these, two were annotated as outer membrane (OM) proteins, two as cytoplasmic membrane proteins, 17 as cytoplasmic proteins (Table 1). The predicted localizations of the remaining 7 proteins were uncertain. 3.4. Characterization of the attenuated mutant strain M11 (Riean_0060:Tn4351) An insertion was found in the Riean_0060, a gene encoding transaldolase-like fructose-6-phosphate aldolase, of the mutant strain M11. The results of Real-time PCR suggested that inactivation of Riean_0060 did not have a polar effect on transcription of the upstream gene (Riean_0059) or the downstream gene (Riean_0061) (Fig. S2), and the growth of the mutant M11 was slightly slower than that of YZb1 (Fig. S3), but there was no significant difference between these two growth curves (p > 0.05). To investigate whether Riean_0060 is involved in the virulence of R. anatipestifer, the LD50 for the mutant strain M11, the complemented strain M11(pRES-Riean_0060), as well as the wild-type strain YZb1, were determined to evaluate the virulence

56

X. Ni et al. / Veterinary Microbiology 189 (2016) 52–58

of these strains. The results showed that the LD50 value for the mutant strain M11 was 2.20  109 CFU, which was a decrease in virulence of about 1700-fold than that of the wild-type strain YZb1 (1.29  106 CFU). Moreover, the LD50 for the complemented strain M11(pRES-Riean_0060) was 8.67  107 CFU, indicating that virulence was increased by about 25-fold as compared to that of mutant M11. In addition, as shown in Fig. 1, the blood bacterial loading of ducklings infected with the mutant strain M11 were significantly decreased as compared to that of ducklings infected with the wild-type strain YZb1 (P < 0.01), while that of the complemented strain M11(pRES-Riean_0060) was significantly restored. Together, these results provide evidence that Riean_0060 is involved in the pathogenesis of R. anatipestifer. 3.5. Characterization of the attenuated mutant strain M29 (Riean_1537:Tn4351) In the mutant strain M29, a 1047 bp ORF of a hypothetical protein (Riean_1537) was mutated by the insertion of transposon Tn4351. No putative conserved domains were detected in this protein (Riean_1537). The results of Real-time PCR suggested that inactivation of Riean_1537 did not have a polar effect on transcription of the upstream gene (Riean_1536) or the downstream gene (Riean_1538) (Fig. S4). The mutant strain M29 exhibited similar colony morphologies as the wild-type strain YZb1 and the complemented strain M29(pRES-Riean_1537) did on TSB agar (data not shown). The growth curve of the M29 mutant in TSB was also similar to those of strains YZb1 and M29(pRES- Riean_1537) (Fig. S5). The LD50 for the mutant strain M29, the complemented strain M29(pRES- Riean_1537), as well as the wild-type strain YZb1, were determined to evaluate the virulence of these strains. The results showed that the LD50 value for the mutant strain M29 was 1.86  108 CFU, which was a decrease in virulence of at least 210-fold than that of the wild-type strain YZb1 (8.86  105 CFU). Moreover, the LD50 for the complemented strain M29(pRES-Riean_1537) was 5.61 107 CFU, indicating that virulence was increased by at least 3-fold as compared to that of mutant M29. The blood bacterial loading of ducklings infected with the mutant strain M29 were significantly decreased as compared to that of ducklings infected with the wildtype strain YZb1 (P < 0.01, Fig. 2), while that of the complemented strain M29(pRES-Riean_1537) was significantly restored. Together,

Fig. 1. Blood bacterial loading in ducklings infected with the mutant strain M11, the complemented strain M11(pRES-Riean_0060), and wild-type strain YZb1. The ducklings (eight ducklings for each group) were challenged with 108 CFU of each bacteria respectively. The bacteria were counted at 12 and 24 h after infection with the respective bacteria. The error bars represent means  standard deviations from eight samples. Asterisks indicate statistically significant differences between two groups (**P < 0.01).

Fig. 2. Blood bacterial loading in ducklings infected with the mutant strain M29, the complemented strain M29(pRES-Riean_1537), and wild-type strain YZb1. 10-dayold Cherry Valley ducklings were injected intramuscularly with 109 CFU of each bacteria respectively. The bacteria were counted at 12 and 24 h after infection with the respective bacteria. The error bars represent means  standard deviations from three ducks. Blood bacterial loading in ducklings infected with M29 were significantly lower than that of ducklings infected with the wild-type strain YZb1. Asterisks indicate statistically significant differences between two groups (**P < 0.01).

these results provide evidence that hypothetical protein Riean_1537 is involved in the pathogenesis of R. anatipestifer. 4. Discussion In this study, 3580 Tn4351 insertion mutants of the wild-type strain YZb1 were screened in ducklings, 29 of which were shown to have an attenuated phenotype and 28 disrupted genes were identified. Whether these genes are involved in the virulence of R. anatipestifer need to be confirmed further. R. anatipestifer has no flagella or pili. Hence, during infection, the OM proteins may play a direct role in cell–cell adherence, which may be involved in virulence. In this study, 2 (7.14%) of 28 genes (Riean_1748 and Riean_1284) were predicted to encode for the TonB-dependent OM receptor (plug), which are important components of the bacterial machinery for the uptake of substances from the environment (Hartney et al., 2011). The roles of TonB-dependent OM proteins as receptors for siderophores, vitamin B12, certain phages, sucrose, maltodextrins, nickel, sulfate, and other substrates have been reported (Blanvillain et al., 2007; Kahnert et al., 2002; Lohmiller et al., 2008; Postle and Kadner, 2003; Schauer et al., 2007). In addition, iron may serve as the “critical determinant” that decides the outcome of host–bacteria interactions (Banin et al., 2005) and metal acquisition transport systems were shown to be important during infection for Vibrio cholerae, Pseudomonas aeruginosa, Staphylococcus aureus and Mycobacterium tuberculosis (Lehoux and Levesque, 2000). In addition, in some Gram-negative bacteria, TonB-dependent receptors are very important to the systemic spread of bacteria (Hu et al., 2012b; Tauseef et al., 2011; Torres et al., 2001). Our previous results also showed that another TonB-dependent OM receptor of R. anatipestifer, TbdR1, was involved in iron acquisition and infection in vivo (Lu et al., 2013). The amino acid sequences of three OM proteins screened in this study were predicted to contain a CirA domain, which is a TonB-dependent transporter involved in the uptake of ferric iron (Fe3+) complexed with catechol siderophores (Hantke, 1990). It suggests that these three proteins may play roles in iron uptake by R. anatipestifer. Of the 28 screened genes, 16 (57.14%) were predicted to encode for enzymes and, among these, 15 (93.75%) encoded for cytoplasmic proteins and the other one (Riean_1210) may have multiple localization sites. Some enzymes or their homologs were

X. Ni et al. / Veterinary Microbiology 189 (2016) 52–58

confirmed to play roles in the virulence of other bacterial species. Mutants 5 and 10 were found to have a transposon insertion in site-specific DNA methyltransferase and homocysteine S-methyltransferase, respectively. DNA methylation plays a role in providing epigenetic information that affects gene expression and other cellular events. In some bacterial species, mutation or overexpression of DNA adenine methylase has been linked to alterations in virulence (Banas et al., 2011; Heusipp et al., 2007; Low et al., 2001; Marinus and Casadesus, 2009). The potential mechanisms behind these effects have been postulated to be indirect, secondary to effects on basic cellular fitness, or direct, affecting transcriptional regulation (Heusipp et al., 2007). Two attenuated mutants, 14 and 21, have mutations in the genes coding for the prolyl oligopeptidase family of proteins and oligopeptidase B, respectively. Oligopeptidase B is a processing peptidase present in Gram-negative bacteria, protozoa, and plants, that is an important virulence factor and therapeutic target in animal trypanosomiasis (Coetzer et al., 2008) and is also involved in host cell invasion by the generation of a Ca2+-agonist necessary for the recruitment and fusion of host lysosomes at the site of parasite attachment (Motta et al., 2012). The attenuated mutant 19 had an interrupted aroC gene, which encodes for chorismate synthase. In Xanthomonas oryzae pathovar oryzae, chorismate synthase plays a crucial role in growth, virulence attenuation, and pigment production (Song et al., 2012). In addition, defined aroC and ssaV deletions were introduced into Salmonella enterica to develop an oral vaccine for humans (Hindle et al., 2002). In mutant 20, an insertion was made to relA, which codes for (p)ppGpp synthetase. Several relA null mutants exhibiting a large decrease in virulence have been observed in S. enterica serovar Typhimurium (Pizarro-Cerda and Tedin, 2004), Vibrio cholerae (Haralalka et al., 2003), Mycobacterium tuberculosis (Dahl et al., 2003), and Listeria monocytogenes (Taylor et al., 2002). In addition, mutants 9 and 18 were found to affect RluD and FtsA, respectively, which are also involved in biofilm formation by R. anatipestifer (Hu et al., 2012a). Disruption of rluD or ftsA is a common factor affecting biofilm formation and virulence of R. anatipestifer, although the exact mechanism remains unclear. Fructose-6-phosphate aldolase catalyzes the reversible formation of fructose 6-phosphate from dihydroxyacetone (DHA) and D-glyceraldehyde 3-phosphate via an aldolization reaction (Schurmann and Sprenger, 2001). In this study, mutant 11, which has a Tn4351 insertion mutation in Riean_0060 encoding transaldolase-like fructose-6-phosphate aldolase, was confirmed to be greatly attenuated to ducklings. In fact, it is not an isolated case. Transaldolase of Vibrio parahaemolyticus was identified as virulence-associated factors involved in the pathogenicity of the bacterium (He et al., 2015). In addition, we also identified nine genes encoding for hypothetical proteins. In this study, our results showed that hypothetical protein Riean_1537 was involved in the virulence of R. anatipestifer. Riean_1537 was predicted to be a cytoplasmic protein and no conserved domain was found according to its amino acid sequence. How does this protein affect the virulence of R. anatipestifer remains to be further investigated. In addition, whether the null mutant of other hypothetical proteins affected virulence remains to be clarified. Further studies on these potential virulence-associated factors will improve our understanding of the pathogenesis of R. anatipestifer.

Acknowledgement This work was supported by the National Natural Science Foundation of China (31272590 and 31472224).

57

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vetmic.2016. 04.014. References Alvarez, B., Secades, P., Prieto, M., McBride, M.J., Guijarro, J.A., 2006. A mutation in Flavobacterium psychrophilum tlpB inhibits gliding motility and induces biofilm formation. Appl. Environ. Microbiol. 72, 4044–4053. Banas, J.A., Biswas, S., Zhu, M., 2011. Effects of DNA methylation on expression of virulence genes in Streptococcus mutans. Appl. Environ. Microbiol. 77, 7236– 7242. Banin, E., Vasil, M.L., Greenberg, E.P., 2005. Iron and Pseudomonas aeruginosa biofilm formation. Proc. Natl. Acad. Sci. U. S. A. 102, 11076–11081. Blanvillain, S., Meyer, D., Boulanger, A., Lautier, M., Guynet, C., Denance, N., Vasse, J., Lauber, E., Arlat, M., 2007. Plant carbohydrate scavenging through tonBdependent receptors: a feature shared by phytopathogenic and aquatic bacteria. PLoS One 2, e224. Coetzer, T.H., Goldring, J.P., Huson, L.E., 2008. Oligopeptidase B: a processing peptidase involved in pathogenesis. Biochimie 90, 336–344. Dahl, J.L., Kraus, C.N., Boshoff, H.I., Doan, B., Foley, K., Avarbock, D., Kaplan, G., Mizrahi, V., Rubin, H., Barry, C.E., 2003. The role of RelMtb-mediated adaptation to stationary phase in long-term persistence of Mycobacterium tuberculosis in mice. Proc. Natl. Acad. Sci. U. S. A. 100, 10026–10031. Hantke, K., 1990. Dihydroxybenzoylserine–a siderophore for E. coli. FEMS Microbiol. Lett. 55, 5–8. Haralalka, S., Nandi, S., Bhadra, R.K., 2003. Mutation in the relA gene of Vibrio cholerae affects in vitro and in vivo expression of virulence factors. J. Bacteriol. 185, 4672–4682. Hartney, S.L., Mazurier, S., Kidarsa, T.A., Quecine, M.C., Lemanceau, P., Loper, J.E., 2011. TonB-dependent outer-membrane proteins and siderophore utilization in Pseudomonas fluorescens Pf-5. Biometals 24, 193–213. He, Y., Wang, H., Chen, L., 2015. Comparative secretomics reveals novel virulenceassociated factors of Vibrio parahaemolyticus. Front. Microbiol. 6, 707. Heusipp, G., Falker, S., Schmidt, M.A., 2007. DNA adenine methylation and bacterial pathogenesis. Int. J. Med. Microbiol. 297, 1–7. Hindle, Z., Chatfield, S.N., Phillimore, J., Bentley, M., Johnson, J., Cosgrove, C.A., Ghaem-Maghami, M., Sexton, A., Khan, M., Brennan, F.R., Everest, P., Wu, T., Pickard, D., Holden, D.W., Dougan, G., Griffin, G.E., House, D., Santangelo, J.D., Khan, S.A., Shea, J.E., Feldman, R.G., Lewis, D.J., 2002. Characterization of Salmonella enterica derivatives harboring defined aroC and Salmonella pathogenicity island 2 type III secretion system (ssaV) mutations by immunization of healthy volunteers. Infect. Immun. 70, 3457–3467. Hu, Q., Chen, H., Liu, X., Zhan, M., Zhang, Z., Deen, S., Zhang, Y., 2002. Determination of growth curve of Riemerella anatipestifer. Anim. Husb. Vet. Med. 34, 8–9. Hu, Q., Han, X., Zhou, X., Ding, S., Ding, C., Yu, S., 2010. Characterization of biofilm formation by Riemerella anatipestifer. Vet. Microbiol. 144, 429–436. Hu, Q., Han, X., Zhou, X., Ding, C., Zhu, Y., Yu, S., 2011. OmpA is a virulence factor of Riemerella anatipestifer. Vet. Microbiol. 150, 278–283. Hu, Q., Zhu, Y., Tu, J., Yin, Y., Wang, X., Han, X., Ding, C., Zhang, B., Yu, S., 2012a. Identification of the genes involved in Riemerella anatipestifer biofilm formation by random transposon mutagenesis. PLoS One 7, e39805. Hu, Y.H., Dang, W., Sun, L., 2012b. A TonB-dependent outer membrane receptor of Pseudomonas fluorescens: virulence and vaccine potential. Arch. Microbiol. 194, 795–802. Hu, Q., Miao, S., Ni, X., Lu, F., Yu, H., Xing, L., Jiang, P., 2013. Construction of a shuttle vector for use in Riemerella anatipestifer. J. Microbiol. Methods 95, 262–267. Kahnert, A., Mirleau, P., Wait, R., Kertesz, M.A., 2002. The LysR-type regulator SftR is involved in soil survival and sulphate ester metabolism in Pseudomonas putida. Environ. Microbiol. 4, 225–237. Lehoux, D.E., Levesque, R.C., 2000. Detection of genes essential in specific niches by signature-tagged mutagenesis. Curr. Opin. Biotechnol. 11, 434–439. Lohmiller, S., Hantke, K., Patzer, S.I., Braun, V., 2008. TonB-dependent maltose transport by Caulobacter crescentus. Microbiology 154, 1748–1754. Low, D.A., Weyand, N.J., Mahan, M.J., 2001. Roles of DNA adenine methylation in regulating bacterial gene expression and virulence. Infect. Immun. 69, 7197– 7204. Lu, F., Miao, S., Tu, J., Ni, X., Xing, L., Yu, H., Pan, L., Hu, Q., 2013. The role of TonBdependent receptor TbdR1 in Riemerella anatipestifer in iron acquisition and virulence. Vet. Microbiol. 167, 713–718. Marinus, M.G., Casadesus, J., 2009. Roles of DNA adenine methylation in hostpathogen interactions: mismatch repair, transcriptional regulation, and more. FEMS Microbiol. Rev. 33, 488–503. Mavromatis, K., Lu, M., Misra, M., Lapidus, A., Nolan, M., Lucas, S., Hammon, N., Deshpande, S., Cheng, J.F., Tapia, R., Han, C., Goodwin, L., Pitluck, S., Liolios, K., Pagani, I., Ivanova, N., Mikhailova, N., Pati, A., Chen, A., Palaniappan, K., Land, M., Hauser, L., Jeffries, C.D., Detter, J.C., Brambilla, E.M., Rohde, M., Goker, M., Gronow, S., Woyke, T., Bristow, J., Eisen, J.A., Markowitz, V., Hugenholtz, P., Klenk, H.P., Kyrpides, N.C., 2011. Complete genome sequence of Riemerella anatipestifer type strain (ATCC 11845). Stand. Genomic Sci. 4, 144–153. Motta, F.N., Bastos, I.M., Faudry, E., Ebel, C., Lima, M.M., Neves, D., Ragno, M., Barbosa, J.A., de Freitas, S.M., Santana, J.M., 2012. The Trypanosoma cruzi virulence factor

58

X. Ni et al. / Veterinary Microbiology 189 (2016) 52–58

oligopeptidase B (OPBTc) assembles into an active and stable dimer. PLoS One 7, e30431. Ochman, H., Gerber, A.S., Hartl, D.L., 1988. Genetic applications of an inverse polymerase chain reaction. Genetics 120, 621–623. Pizarro-Cerda, J., Tedin, K., 2004. The bacterial signal molecule, ppGpp, regulates Salmonella virulence gene expression. Mol. Microbiol. 52, 1827–1844. Postle, K., Kadner, R.J., 2003. Touch and go: tying TonB to transport. Mol. Microbiol. 49, 869–882. Reed, M.J., Muench, H., 1938. A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27, 493–497. Schauer, K., Gouget, B., Carriere, M., Labigne, A., de Reuse, H., 2007. Novel nickel transport mechanism across the bacterial outer membrane energized by the TonB/ExbB/ExbD machinery. Mol. Microbiol. 63, 1054–1068. Schurmann, M., Sprenger, G.A., 2001. Fructose-6-phosphate aldolase is a novel class I aldolase from Escherichia coli and is related to a novel group of bacterial transaldolases. J. Biol. Chem. 276, 11055–11061. Song, E.S., Park, Y.J., Noh, T.H., Kim, Y.T., Kim, J.G., Cho, H., Lee, B.M., 2012. Functional analysis of the aroC gene encoding chorismate synthase from Xanthomonas oryzae pathovar oryzae. Microbiol. Res. 167, 326–331. Tauseef, I., Harrison, O.B., Wooldridge, K.G., Feavers, I.M., Neal, K.R., Gray, S.J., Kriz, P., Turner, D.P., Ala’Aldeen, D.A., Maiden, M.C., Bayliss, C.D., 2011. Influence of the

combination and phase variation status of the haemoglobin receptors HmbR and HpuAB on meningococcal virulence. Microbiology 157, 1446–1456. Taylor, C.M., Beresford, M., Epton, H.A., Sigee, D.C., Shama, G., Andrew, P.W., Roberts, I.S., 2002. Listeria monocytogenes relA and hpt mutants are impaired in surfaceattached growth and virulence. J. Bacteriol. 184, 621–628. Torres, A.G., Redford, P., Welch, R.A., Payne, S.M., 2001. TonB-dependent systems of uropathogenic Escherichia coli: aerobactin and heme transport and TonB are required for virulence in the mouse. Infect. Immun. 69, 6179–6185. Tu, J., Lu, F., Miao, S., Ni, X., Jiang, P., Yu, H., Xing, L., Yu, S., Ding, C., Hu, Q., 2014. The siderophore-interacting protein is involved in iron acquisition and virulence of Riemerella anatipestifer strain CH3. Vet. Microbiol. 168, 395–402. Wang, X., Zhu, D., Wang, M., Cheng, A., Jia, R., Zhou, Y., Chen, Z., Luo, Q., Liu, F., Wang, Y., Chen, X.Y., 2012. Complete genome sequence of Riemerella anatipestifer reference strain. J. Bacteriol. 194, 3270–3271. Yuan, J., Liu, W., Sun, M., Song, S., Cai, J., Hu, S., 2011. Complete genome sequence of the pathogenic bacterium Riemerella anatipestifer strain RA-GD. J. Bacteriol. 193, 2896–2897. Yuan, J., Li, L., Sun, M., Dong, J., Hu, Q., 2013. Genome sequence of avirulent Riemerella anatipestifer strain RA-SG. Genome Announcements 1, e0021812.