Previous Article | Next Article ![]()
Journal of Bacteriology, April 2009, p. 2622-2629, Vol. 191, No. 8
0021-9193/09/$08.00+0 doi:10.1128/JB.01370-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Wei Li,
,
Qian Zhang,
Hongxia Wang,
Xiao Xu,
Baowei Diao,
Lijuan Zhang, and
Biao Kan*
National Institute for Communicable Disease Control and Prevention and State Key Laboratory for Infectious Disease Prevention and Control, Chinese Center for Disease Control and Prevention, P.O. Box 5, Changping, Beijing 102206, Peoples Republic of China
Received 1 October 2008/ Accepted 22 January 2009
|
|
|---|
|
|
|---|
The first step of phage infection is the specific binding of phages to receptors on the host cell surface, triggering the ejection of phage DNA into the cell and initiating the propagation process (1). Some outer membrane proteins serve as receptors for phage adsorption, such as LamB (19), OmpA (8), and OmpC (52) of Escherichia coli and OmpK of Vibrio parahaemolyticus (23). Lipopolysaccharide (LPS) is a common phage receptor. Cytotoxin-converting
CTX from Pseudomonas aeruginosa (51) and FC3-10 from Klebsiella spp. (4) bind to the core region of LPS, while vibriophage CP-T1 (17) and
YeO3-12, specific for Yersinia enterocolitica serotype O:3 (35), bind to the O side chain. Other structures that can be used as receptors include flagella (32) and pili (49). Besides the bacterial receptors, there are some other host components necessary for the propagation of virulent phages, such as the host RNA polymerase involved in the transcription of phage genes and the components participating in the assembly of phage particles.
VP3 is one of the five typing phages in the phage-biotyping scheme (14). The complete genome sequence comprises 39,481 bp with an overall G+C content of 42.62%, and 52 open reading frames (ORFs) were predicted (our unpublished data). It is a T7-like phage by morphology and genome sequence comparison. Here, we have identified the receptor and other host genes of V. cholerae that are necessary for the lytic propagation of VP3. In addition, VP3 tail fiber protein was also predicted and proved to function as a receptor-binding protein.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Strains and plasmids used in this study
|
Transposon mutagenesis and the selection of VP3-resistant mutants.
Plasmid pUTkm2 (10) was transformed into E. coli SM10(
pir) (44) to obtain SM10-km2 (Kanr). Conjugation was performed between N16961-Smr as the recipient strain and SM10-km2 as the donor strain. The donor and recipient strain (optical density at 600 nm [OD600]
0.8) cultures were mixed 1:9 in a 1.5-ml microcentrifuge tube, washed twice with 1 ml of LB, resuspended in 100 µl of LB, and transferred onto 0.45-µm-pore-size membranes overlaid on LB plates. After being incubated at 37°C for 4 to 6 h, the membranes were washed with LB and the collected bacteria were diluted and plated onto the LB agar plates (Kan, 50 µg/ml, and Sm, 100 µg/ml) to select mutants.
Separated transconjugants were picked and inoculated into 150 µl of LB (Kan, 50 µg/ml, and Sm, 100 µg/ml) contained in 96-well plates (Corning Costar 3599) and incubated for 3 to 5 h until the OD was 0.5 to 0.8. This was monitored with a Tecan Genios Basic Microplate Reader at 395 nm. Then, 10 µl of the cultures was inoculated into 140 µl of LB (Kan, 50 µg/ml, and Sm, 100 µg/ml) with phage VP3 (1 x 108 PFU/ml) contained in new 96-well plates and incubated for 3 h. The wells with an OD395 of 0.5 or more were selected as the candidates for VP3-resistant mutants, which were confirmed subsequently by double-layer plaque assay. Briefly, 4 ml of melted 0.7% LB agar was mixed with 100 µl of cell cultures and poured onto an LB agar plate, and 10 µl of VP3 10-fold serial dilutions was dropped onto the plate when the upper layer solidified. After overnight incubation, mutants with no plaque formation were confirmed as VP3 resistant. The Kan resistance gene (kan) was detected by PCR with the primers pkan-low/pkan-up, and the transposase gene tnpA was detected with the primers ptnpA-low/ptnpA-up (Table 2). Strains with kan and without tnpA had the transposon inserted into chromosomes.
|
View this table: [in a new window] |
TABLE 2. Primers used in this study
|
Construction of deletion mutants and complementation experiments.
An in-frame VC0231 deletion mutant of V. cholerae N16961 was constructed by homologous recombination using the suicide plasmid pDS132 (37). The flanking sequences on the 5' and 3' sides of the fragment that was to be deleted were PCR amplified from N16961 chromosomal DNA using the primers m0231U-1/m0231U-2 and m0231L-1/m0231L-2 (Table 2), respectively. Then, the fragments were cloned into pUC19, generating the plasmid p0231UL. The fragment containing the in-frame deletion pattern of VC0231 was obtained by PCR from a p0231UL template using primers m0231U-3/m0231L-2 (Table 2) and cloned into pDS132 to construct the plasmid pDS-0231, which was conjugally transferred into N16961 from the donor strain E. coli SM10(
pir). Transconjugants were selected on thiosulfate-citrate-bile-sucrose (TCBS) agar with 15 µg/ml Cm and streaked onto LB agar plates with 20% sucrose and without NaCl. Colonies from sucrose selection medium were screened by PCR using the primers inner0231-1/inner0231-2 (Table 2), which were specific for the deleted fragment. Sequencing confirmed that the negatively PCR-amplified colonies were mutants.
A trxA deletion mutant of N16961 was constructed using the suicide plasmid pCVD442 (11) in a similar way. The flanking sequences on the 5' and 3' sides of trxA were PCR amplified using the primers m0306U-1/m0306U-2 and m0306L-1/m0306L-2 (Table 2), respectively, and were subsequently cloned into pTcat (29), generating the plasmid p0306U-cat-L. The plasmid pTcat was constructed previously in our laboratory by inserting a cat gene, which confers Cm resistance, on pMD18-T (TaKaRa). The plasmid p0306U-cat-L was restricted with SacI and SphI, and the 2-kb band containing the in-frame deletion pattern of trxA was purified and cloned into pCVD442 to construct the plasmid pCVD-0306. The following procedures were similar to those for the construction of the VC0231 deletion mutant, except that the sucrose selection medium contained Cm (15 µg/ml). Mutants were screened using the primers inner0306-1/inner0306-2 (Table 2) and were confirmed by sequencing.
When the complementary plasmid pBR0231c was constructed, the fragment containing VC0231 was PCR amplified from chromosomal DNA of N16961 with the primers VC0231cU/VC0231cL and inserted into pBR322. In the same way, the plasmids pBR0229c, pBR0305c, pBR0306c, and pBR0305-6c were constructed with the primers VC0229cU/VC0229cL, VC0305cU/VC0305cL, VC0306cU/VC0306cL, and VC0305cU/VC0306cL (Table 2), respectively. The fragment inserted into pBR0305-6c covered the genes rhlB and trxA. The complementary plasmids were transformed into different mutants, as mentioned in Results. The plasmid pBR322 was transformed into the mutants as a control. The sensitivity of the complemented mutants was examined by double-layer plaque assay.
Detection of gene transcription. The transcription of rhlB and trxA was detected by reverse transcription (RT)-PCR. Total RNA was extracted with a Qiagen RNeasy Mini Kit from N16961, N60C6, and N60C6 complemented with different plasmids and reverse transcribed with SuperScript III Reverse Transcriptase (Invitrogen). The resulting cDNAs were used as PCR templates amplified with gene-specific primers.
Expression, purification, and FITC labeling of the predicted VP3 tail fiber protein. A DNA fragment containing gene 44, which was predicted to encode VP3 tail fiber protein, was PCR amplified from VP3 DNA with the primers gp44NB1 (5'-TATACATATGCACCATCATCATCATCATTCAGGCACTCGTGCTCCT-3') and gp44NB2 (5'-CGCACTCGAGTTAATTTAAAGGGATAGT-3'). The former contained the coding sequence of a hexahistidine tag and the restriction site for NdeI, and the latter contained the restriction site for XhoI (both underlined). The PCR product was digested with NdeI and XhoI and inserted into NdeI/XhoI-digested pET30-a (+). The resulting plasmid was transformed into the host strain E. coli BL21(DE3). Expression of His6-gp44 was induced with 0.4 mmol/liter isopropyl-β-D-thiogalactopyranoside (IPTG) at 28°C for 2 to 3 h. The total protein of the induced strain was separated by SDS-polyacrylamide gel electrophoresis (PAGE) and detected by Western blotting using mouse anti-His antibody at a concentration of 1:1,000 and horseradish peroxidase-conjugated goat anti-mouse secondary antibody at a concentration of 1:500. His6-gp44 was purified with a His·Bind Purification Kit (Novagen) and labeled with a fluorescein isothiocyanate (FITC) labeling kit (Guangzhou Chang Rui Biotech, China) according to the manufacturer's instructions. His6-gp44 labeled with FITC is referred to as FITC-gp44 below.
Adsorption of FITC-gp44 to strains. Strains were cultured in LB at 37°C until the OD600 was 0.3 to 0.4. Cells in 1 ml of the culture were collected by centrifugation at 4,000 x g for 4 min, washed once with phosphate-buffered saline (PBS), incubated for 1 h in FITC-gp44 solution, washed twice with PBS to remove the FITC-gp44 not adsorbed by the cells, and finally suspended in PBS. Two negative controls were made; one was E. coli BL21(DE3) treated with FITC-gp44 in parallel to the test strains, and the other was N16961 treated with PBS instead of the FITC-gp44 solution. To detect the dose dependence of agglutination of N16961 by FITC-gp44, cells of N16961 were incubated with twofold serial dilutions (1:2 to 1:32) of FITC-gp44 in PBS. All procedures were carried out at 4°C. After the FITC-gp44 solution was added, the samples were kept in the dark.
Each sample was mixed with an equal volume of 0.5% agarose preheated to 50°C, and 4 µl of the mixture was immediately dropped onto a slide, covered with a coverslip, and then observed with a fluorescence microscope (Olympus; BX51) and by confocal laser scanning microscopy (CLSM) (Olympus; FV500).
Nucleotide sequence accession number. The GenBank accession number for VP3 gene 44 is FJ217960.
|
|
|---|
![]() View larger version (37K): [in a new window] |
FIG. 2. trxA and VP3 infection. (A) In mutant N60C6, a transposon was inserted into the region between rhlB (VC0305) and trxA (VC0306) in the 5' noncoding regions of both genes. nt., nucleotides. (B) Three complementary plasmids were constructed, each carrying a fragment containing rhlB, trxA, or both. The lines show the regions covered by the cloned fragments. (C) Response to VP3 infection by double-layer plaque assay. N60C6 became VP3 sensitive when it was transformed by pBR0306c (generating N60C6-vc0306), but not when the transforming plasmid was pBR0305c (generating N60C6-vc0305). (D) The transcription of rhlB and trxA was detected by RT-PCR. The lanes marked with a plus are positive controls.
|
![]() View larger version (22K): [in a new window] |
FIG. 1. Transposon insertion in the wav gene cluster and VP3 infection of the mutants. (A) Organization of the wav gene cluster of N16961 and the transposon insertion sites of the five VP3-resistant mutants. The large black arrows represent wav genes that were assumed to be essential. The white arrows represent genes that interfere with O-antigen attachment (34, 41). The grey arrows represent genes in the wav gene cluster. The transposon insertion positions are marked with small black arrows. The numbers after the mutant names stand for the nucleotide (nt) site just following the transposon in chromosome I of V. cholerae N16961. (B) Double-layer plaque assay of mutants (N67B1, N77C6, and C29) and complemented mutants (N67B1-vc0231, N77C6-vc0229, and C29-vc0231).
|
V. cholerae trxA encoding thioredoxin is relevant for VP3 infection. In mutant N60C6, the transposon lies in the 5' noncoding regions of rhlB and trxA. To determine which gene (or both) is responsible for resistance to VP3 infection, three complementary plasmids were constructed and used to transform N60C6: pBR0305c, carrying rhlB; pBR0306c, carrying trxA; and pBR0305-6c, carrying both genes (Fig. 2B). N60C6 regained VP3 sensitivity when it was transformed with pBR0306c or pBR0305-6c, but not pBR0305c (Fig. 2C). We then tested whether both genes were transcribed in N60C6. When tested by RT-PCR, rhlB was still transcribed while trxA was not. The transcription of trxA was detected when N60C6 was complemented with pBR0306c (Fig. 2D). These results indicate that trxA was inactivated in N60C6, which led to the VP3 resistance. The trxA deletion mutant (N16961-d0306) resisted VP3 infection, and it became VP3 sensitive when transformed with pBR0306c (Fig. 2C), further confirming that trxA is indispensable for VP3 propagation.
gp44 of VP3 can adsorb to VP3-sensitive strains. In order to find out whether the mutations influenced the adsorption of VP3 or the subsequent processes, the gene for VP3 tail fiber protein was predicted and expressed in vitro. The predicted gene 44 of VP3 encodes a protein (designated gp44) of 753 amino acid residues (aa), the N-terminal portion (aa 8 to 244) of which showed an identity of 48% with the N-terminal part of T7 gp17 (tail fiber protein; 553 aa) and showed high similarity to tail fiber proteins of other T7 family phages (Fig. 3). By searching in the Pfam database (13), VP3 gp44 was found to contain two Pfam-A (Phage_T7_tail and Collar) and one Pfam-B (Pfam-B_57397) domains. These domains are all common in phage tail fiber proteins, and domain Phage_T7_tail (corresponding to aa 15 to 244 of VP3 gp44) attaches the fibers to phage particles in phage T7 (45). Therefore, gp44 was predicted to be a tail fiber protein of VP3 and probably capable of binding to receptors with its C terminus.
![]() View larger version (38K): [in a new window] |
FIG. 3. Comparison of tail fiber protein sequences of phage T7, T3, YeO3-12, gh-1, and VP3. T7, T3, YeO3-12, and gh-1 are all T7-like phages. At the top are shown the identities and conservation of the full-length proteins. Darker blue dots mean higher identities. Lighter and higher yellow columns mean more conservation. The box shows multiple alignment of the N termini of the tail fiber proteins, where the similarity is particularly high.
|
His6-gp44 was labeled with FITC and used to treat VP3-sensitive N16961. Upon observation with a fluorescence microscope, many green masses were found in the sample, and most bacteria were agglutinated. Green-fluorescent masses were not found for the two negative controls, N16961 treated with PBS and E. coli BL21(DE3) treated with FITC-gp44 (data not shown). A similar phenomenon was also found under CLSM. The agglutination of N16961 changed with the amount of FITC-gp44 added (Fig. 4). The less FITC-gp44 added, the weaker the fluorescence and the smaller the bacterial masses formed, which showed dose dependence of gp44 for agglutination of N16961. These results suggest that gp44 adheres to VP3-sensitive strains by binding to receptors on the cell surface, and the possible oligomerization of gp44 causes the cells to agglutinate. If so, instead of VP3 particles, gp44 can be used to determine whether a strain can adsorb VP3.
|
View larger version (19K): [in a new window] |
FIG. 4. Dose dependence analysis of FITC-gp44 for agglutination of N16961. Cells of N16961 were treated with twofold serial dilutions of prepared FITC-gp44 (1:2 to 1:32) and observed by CLSM. The dilution factors of FITC-gp44 are shown below.
|
![]() View larger version (98K): [in a new window] |
FIG. 5. Binding of FITC-gp44 on the surfaces of different strains observed by CLSM. E. coli BL21(DE3) treated with FITC-gp44 was used as a negative control. N60C6 is an N16961 mutant with a transposon inserted into the intergenic region between rhlB and trxA. N67B1 and N77C6 are N16961 mutants with a transposon inserted into VC0231 and VC0229, respectively. C29 is an N16961 mutant with an in-frame deletion in VC0231. N67B1-vc0231, N77C6-vc0229, and C29-vc0231 are mutants with complementary plasmids.
|
|
|
|---|
Many members of the T7 group use LPS as a receptor. The receptor of Y. enterocolitica phage
YeO3-12 is the O chain of serotype O:3, which consists of the rare sugar 6-deoxy-L-altropyranose (35); the receptor of lytic Pseudomonas putida bacteriophage gh-1 is proposed to be LPS, as well (27); and the LPS core is the binding site for T7 (38). The wav gene cluster of N16961, responsible for the synthesis of the OS region of LPS, lies in the 17.2-kb region between ORFs VC0223 and VC0240, including a group of glycosyl transferases and O-antigen ligase (Fig. 1A). It has been reported that waaA (VC0233), wavC (VC0227), and waaC (VC0225) are highly conserved and putatively essential core OS genes in V. cholerae (34). Among all the wav genes, only mutations in waaL (VC0237), wavA (VC0223), and wavL (VC0239) affect O antigen attachment (34, 41). Here, VC0229 and VC0231 were found to be relevant for N16961 sensitivity to phage VP3 in the process of phage attachment. Since genes related to O-antigen synthesis in these VC0229 and VC0231 insertion mutants are integral, O antigen is not proposed to function as the VP3 receptor. The exact functions of VC0229 and VC0231 are still unknown. VC0231 encodes a protein predicted to contain the catalytic domain of serine/threonine protein kinases. Genes encoding similar proteins exist in the V. parahaemolyticus O- and K-antigen biosynthesis gene cluster (GI193787939) and the Actinobacillus suis serotype K1 capsular gene cluster (GI29469153) ORF 7 (2). We predicted that the product of VC0231 possibly affects the activities of some enzymes needed in the process of OS synthesis. The OS was concluded to be the VP3 receptor on strain N16961.
Our results also implicate the gene trxA, encoding thioredoxin, in VP3 infection. In phage T7, E. coli thioredoxin is the processivity factor of T7 DNA polymerase (the gene 5 product of T7), increasing the processivity of nucleotide polymerization (31, 46), and influences the binding of T7 helicase to a DNA polymerase-thioredoxin complex (15). The situation may be similar for DNA polymerase of phage T3 (9). E. coli mutants with inactivated trxA become resistant to T7 (21, 38) but adsorb T7 normally (6). Similarly, thioredoxin is probably involved in the life process of VP3 as a subunit of VP3 DNA polymerase but is not involved in the binding of VP3. This supposition was also supported by the binding of FITC-gp44 on the surface of N60C6, which suggested that the inactivation of trxA did not alter the adsorption sites of phage VP3.
The tail fiber protein of a tailed phage has the receptor-binding function and decides the host specificity of the phage (16, 47, 50). VP3 gp44 is a homologue of tail fiber protein gp17 of phage T7, especially the proximal part. Three domains are predicted in gp44, and the proximal domain, Phage_T7_tail, is common in T7 family members, such as bacteriophage T7, T3 (36), K1F (43), and Pseudomonas phage gh-1 (27), and it links tail fibers to phage particles (45). The other two domains are also common in phage tail proteins. Domain Collar exists in the short tail fiber protein gp12 and the long tail fiber protein gp37 responsible for the receptor recognition of bacteriophage T4 (33, 48), while PfamB_PB057397 exists in ORF35 (tail fiber protein) of Vibrio phage K139 and some K139-like phages, such as Ch457 and E8498 (25). Therefore, gp44 of phage VP3 was predicted to be the tail fiber protein and to have a receptor-binding function; this was validated by the binding of gp44 to the surface of wild-type N16961. N16961 mutants with mutations in the wav gene cluster did not bind gp44, further proving the interaction of gp44 with the OS region of LPS.
We found that the expressed His6-gp44 tended to form oligomers, and strong denaturing conditions promoted the dissociation of the oligomer. In addition, fluorescence microscopy and CLSM showed that when FITC-gp44 was added, cells of VP3-sensitive strains agglutinated while cells of VP3-resistant wav gene mutants were dispersed. When N16961 and N60C6 were blended with a His6-gp44 solution dropped onto a slide, the cells were agglutinated, just like the reaction with V. cholerae antiserum, while E. coli BL21(DE3) and the wav gene mutants were not agglutinated (data not shown). These results implied that His6-gp44 is oligomeric in solution, which is the common form for phage tail fiber proteins. For example, T4 short tail fibers are trimeric gp12 (30), and tail fibers of T3 (26) and T7 (45) are trimers of the gp17 monomer. The agglutination of VP3-sensitive strains could be caused by the oligomerization of His6-gp44.
Conclusion. Some bacteria cannot be lysed by a phage because they do not adsorb the phage particles, owing either to a lack of receptors on the cell surface or to the fact that the receptors are not accessible due to other cell surface components (42). Some bacteria are able to adsorb phages but fail to serve as host cells for phage multiplication if they lack any components necessary for phage propagation (1). We confirmed that mutants with an altered wav gene cluster resisted the virulent infection of VP3 because they could not adsorb the phage particles through the interaction between VP3 gp44 and the OS of V. cholerae. Thioredoxin, encoded by trxA, is necessary for VP3 proliferation and may be a subunit of VP3 DNA polymerase. These results can help us to understand the mechanism of T7-like phage infection of V. cholerae.
Published ahead of print on 6 February 2009. ![]()
J.Z. and W.L. contributed equally to this work. ![]()
Present address: Laboratory of Nucleic Acid Vaccines, Department of Medicine, University of Massachusetts Medical School, Worcester, MA. ![]()
|
|
|---|
, a novel filamentous phage of Vibrio cholerae, integrates into the same chromosomal site as CTX
. J. Bacteriol. 185:5685-5696.
of Vibrio cholerae and its induction. Zhonghua Wei Sheng Wu Xue He Mian Yi Xue Za Zhi 24:253-257.
YeO3-12, specific for Yersinia enterocolitica serotype O:3, is related to coliphages T3 and T7. J. Bacteriol. 182:5114-5120.
CTX, of Pseudomonas aeruginosa. J. Bacteriol. 176:5262-5269.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»