pIIICTX a Predicted CTX Minor Coat Protein, Can Expand the Host Range of Coliphage fd To Include Vibrio cholerae
Division of Geographic Medicine/Infectious Disease, New England Medical Center and Tufts University School of Medicine, and Howard Hughes Medical Institute, Boston, Massachusetts 02111x}-z:3, 百拇医药
Received 19 August 2002/ Accepted 11 November 2002x}-z:3, 百拇医药
ABSTRACTx}-z:3, 百拇医药
CTX{phi} is a filamentous bacteriophage that encodes cholera toxin. CTX{phi} infection of its host bacterium, Vibrio cholerae, requires the toxin-coregulated pilus (TCP) and the products of the V. cholerae tolQRA genes. Here, we have explored the role of OrfU, a predicted CTX{phi} minor coat protein, in CTX{phi} infection. Prior to the discovery that it was part of a prophage, orfU was initially described as an open reading frame of unknown function that lacked similarity to known protein sequences. Based on its size and position in the CTX{phi} genome, we hypothesized that OrfU may function in a manner similar to that of the coliphage fd protein pIII and mediate CTX{phi} infection as well as playing a role in CTX{phi} assembly and release. Deletion of orfU from CTX{phi} dramatically reduced the number of CTX{phi} virions detected in supernatants from CTX{phi} -bearing cells. This defect was complemented by expression of orfU in trans, thereby confirming a role for this gene in CTX{phi} assembly and/or release. To evaluate the requirement for OrfU in CTX{phi} infection, we introduced fragments of orfU into gIII in an fd derivative to create OrfU-pIII fusions. While fd is ordinarily unable to infect V. cholerae, an fd phage displaying the N-terminal 274 amino acids of OrfU could infect V. cholerae in a TCP- and TolA-dependent fashion. Since our findings indicate that OrfU functions as the CTX{phi} pIII, we propose to rename OrfU as pIIICTX. Our data also provide new evidence for a conserved pathway for filamentous phage infection.
INTRODUCTIONd(9o, 百拇医药
The molecular interactions that mediate the entry of Escherichia coli-derived filamentous phages into their hosts have been studied in considerable detail. pIII, a phage minor coat protein found on one tip of the virion, accounts for host specificity in both N-pilus (IKE and I2-2)- and F-pilus (Ff)-specific phages (fd, f1, M13) (3, 8, 11, 24). The initial step in Ff phage infection is mediated by the binding of pIII to the tip of the F pilus (19). The Ff phages do not absolutely require the F pilus for infection; however, Ff infection of F- E. coli occurs at much reduced frequencies (32). The binding of pIII to F is thought to result in pilus retraction, which brings the phage particle in close proximity to and perhaps through the bacterium's outer membrane. Subsequent translocation of Ff DNA through the periplasm and inner membrane requires the products of the tolQRA genes both in F+ and in F- E. coli (32, 36, 37). Recent studies suggest that, in addition to binding to the pilus, pIII also binds to TolA, which serves as a coreceptor (4, 30). Although the exact role of TolQ and TolR in the infection process is unknown, they appear to interact with TolA through their inner membrane spanning regions (9, 21), and they may ordinarily constitute a channel for importing as yet unknown substrates across the inner membrane (42). Following the TolA-pIII interaction, pVIII, the Ff major coat protein, is inserted into the host cell inner membrane, and the phage DNA is translocated into the cytoplasm of the host cell (5, 33).
The 424-amino-acid Ff pIII is thought to consist of a leader sequence (2) and three domains, separated by glycine-rich regions, that serve distinct roles in phage entry and release. The first two pIII domains, designated g3p-D1 (residues 1 to 86) and g3p-D2 (residues 105 to 236), are required for Ff adsorption and entry (7, 30, 35), while the third domain, g3p-D3 (residues 270 to 406), is required for the assembly and release of Ff particles from host cells (6, 29). g3p-D1, the "penetration" domain, and g3p-D2, the "adsorption" domain, have been shown to bind to the Ff coreceptors TolA and F, respectively. Holliger and coworkers have proposed that following binding of g3p-D2 to F, the conformation of pIII changes and that this shift reveals the TolA binding site in g3p-D1. Thus, Ff infection is thought of as a two-step process in which different domains of pIII sequentially bind to F and then to TolA (16, 23).ns}(^, 百拇医药
The C terminus of pIII (g3p-D3) has been implicated in the assembly of Ff particles and in their release from the bacterial membrane. In the absence of pIII, Ff virions remain attached to the cell surface in long "polyphages" (6, 22, 28, 29). According to the model of Rakonjac et al. (28), pIII and pVI (another minor coat protein) are incorporated into assembling phage particles through interactions with pVIII to form a "pretermination complex." Only the last 83 amino acids of pIII are required to form this complex. A conformational change in g3p-D3 may disrupt the association of pIII with the inner membrane, enabling capping and release of virions. g3p-D3 is also required for the production of stable virions that are resistant to certain detergents (28).
CTX{phi} is a Vibrio cholerae-derived lysogenic filamentous phage that contains ctxAB, the operon encoding cholera toxin, in its 7-kb genome. Nontoxigenic V. cholerae strains can be converted to toxigenicity by CTX{phi} infection (41). As in Ff infection of E. coli, CTX{phi} infection of V. cholerae requires a cell surface pilus, toxin-coregulated pilus (TCP), and the periplasmic/inner membrane protein complex TolQRA (13). TCP is a type IV bundle-forming pilus that is also required for V. cholerae intestinal colonization (38). We and others previously found that inactivation of either tcpA or the tolQRA genes of V. cholerae reduced CTX{phi} infection by more than 3 orders of magnitude (13, 41). Given the similarities of the requirements for CTX{phi} infection of V. cholerae to those for Ff infection of E. coli, we hypothesized that CTX{phi} might contain a virion coat protein that functions analogously to pIII and interacts with TCP and the TolQRA complex to enable CTX{phi} infection of V. cholerae.
The CTX{phi} core region encodes proteins for virion morphogenesis, including the putative major coat protein (Cep), three putative minor coat proteins (Psh, OrfU, and Ace), and Zot (20), which is similar to the Ff protein pI and is required for phage assembly and secretion. Although several of the putative CTX{phi} gene products bear similarity to known filamentous phage proteins, none of the CTX{phi} genes encode a protein similar in sequence to pIII. However, since pIII and OrfU have similar sizes and positions within their respective genomes (immediately 3' of the gene encoding the major coat protein), we hypothesized that OrfU is the CTX{phi} pIII homologue. Recently, Holliger and Reichmann determined the solution structure of g3p-D1 by nuclear magnetic resonance spectroscopy. They predicted that residues 18 to 86 of OrfU may fold into a ß barrel similar to that of g3p-D1 (15). Their analysis lends additional credence to the hypothesis that OrfU is the CTX{phi} pIII equivalent and to introducing the idea that these particular OrfU residues are involved in the binding of OrfU to the V. cholerae TolA (15). To test the hypothesis that OrfU mediates CTX{phi} infection, we "displayed" OrfU on the surface of an Ff phage as an OrfU-pIII fusion protein. The resulting phage infected both F+ E. coli and TCP+ V. cholerae. Furthermore, OrfU was found to be required for CTX{phi} production by V. cholerae. These findings indicate that OrfU functions as the CTX{phi} pIII analogue, and we propose to rename OrfU as pIIICTX.
MATERIALS AND METHODS(7u, 百拇医药
Strains, media, and antibiotics. The bacterial strains used in this study are listed in Table 1. Bacteria were grown in Luria-Bertani (LB) broth at 37°C. To induce TCP production, the classical V. cholerae strain O395 was grown on a roller drum shaker at 30°C as previously described (13). To ensure F pilus production by E. coli strain TG1, the strain was maintained on M9 minimal medium plates containing glucose. Antibiotics were used at the following concentrations: ampicillin, 50 µg/ml (V. cholerae) and 100 µg/ml (E. coli); streptomycin, 200 µg/ml; kanamycin, 50 µg/ml; chloramphenicol, 15 µg/ml (E. coli) and 1 µg/ml (V. cholerae); tetracycline, 20 µg/ml (E. coli) and 1 µg/ml (V. cholerae). Arabinose (Ara) (0.02%) was added to LB broth to induce expression of genes under the control of the pBAD promoter.(7u, 百拇医药
fig.ommitted(7u, 百拇医药
List of strains and plasmids used in this study
Construction of pCTX-Kan{Delta} orfU and an orfU expression plasmid. An in-frame deletion of orfU in pCTX-Kan was constructed using the PCR technique of splicing by overlapping extension (18). The two outside primers were DelUA (5' GCGATAAGGGGCTTTATTGGTATG 3') and DelUD (5' CATTCGTGATGATGTGACGGC 3'). The two internal primers were DelUB2 (5' GAGCATGGCGGCCAGAAAAAAGGGCGAACG 3') and DelUC2 (5' CGTTCGCCCTTTTTTCTGGCCATGCTC 3'). After subcloning of the deletion PCR product into pCRII-TOPO, an AscI/PmlI fragment from this plasmid was ligated with AscI/PmlI-digested pCTX-Kn. The resulting plasmid lacks all but the last nine amino acids of OrfU. An orfU complementing plasmid, pDH5.52, was constructed by amplifying orfU with the forward and reverse primers OrfU1F (5' ATGCGCTATTTTCTACTGTTTTTG 3') and OrfU4R (5' TCTAGAAAATCACCTAAACAAAATGAGCATG 3'). After subcloning of the PCR product into pCRII-TOPO (Invitrogen, Carlsbad, Calif.), an XbaI/EcoRI fragment from this plasmid was ligated with XbaI/EcoRI-digested pBAD24.
Transduction assay. CTX-Kn{phi} and CTX-Kn{Delta} orfU{phi} transducing particles were enumerated using a previously described transduction assay (13). Briefly, filtered supernatants were prepared from cultures grown to an optical density at 600 nm (OD600) of approximately 0.8. A 75-µl volume of the filtered supernatants from these cultures was mixed with 75 µl of TCP+ O395 cells for 20 min at room temperature on a shaker. Each mixture was then plated on LB agar containing either streptomycin, to enumerate the number of recipient cells, or kanamycin, to enumerate the transductants. The transduction frequency was calculated by dividing the number of transductants by the number of recipients. Similar transduction assays were used to calculate the numbers of the fd-derived transducing phages we studied here. To prepare these phages, DH5pir harboring the different fd-derived phages was grown overnight in LB broth and tetracycline. Phage particles were then precipitated from filtered supernatants with 5% polyethylene glycol (PEG) in 2.5 M NaCl overnight at 4°C, followed by centrifugation at 10,000 x g for 30 min. The particles were then resuspended in LB broth. To enumerate transductants, cell mixtures were plated on LB agar plates containing tetracycline, and to enumerate recipients, the mixtures were plated on LB plates either alone (for TG1) or with streptomycin (for the V. cholerae strains). The transduction frequency was calculated as described above.
Detection of phage DNA in supernatants. Southern blot analysis was used to detect phage DNA in filtered supernatants from mid-log-phase cultures of Bah3 harboring pCTX-Kn derivatives. The phage particles were PEG precipitated as described above and then resuspended in a sodium dodecyl sulfate (SDS)-containing buffer (1% SDS, 1x Tris-acetate-EDTA [TAE], 5% glycerol, 0.25% loading dye) at 70°C for 20 min to disrupt virions (29). These preparations were then subjected to 1% agarose electrophoresis, and CTX{phi} DNA was then detected by Southern analysis using the CTX{phi} rstB gene as a probe. The rstB gene was amplified with primers RstBF (5' ATGAAATTATGGGTGATTAAT 3') and RstBB (5' CCACCGCGAAATAGAGCGCCT 3'). For the nondenaturing Southern blots, the NaOH treatment of the gel was omitted prior to transfer.a/;g, 百拇医药
Construction of fd-DOG derivatives containing OrfU-pIII fusions. Our strategy was to amplify different regions of orfU and subclone them into the pcDNA3.1/Myc(-)B vector to add a myc epitope to the 3' end of the construct. We then inserted this downstream of the leader sequence of gIII in fd-DOG. The primers used for these constructs are listed in . After the first round of amplification, the PCR products were ligated into the TA cloning vector pCRII-TOPO according to the manufacturer's protocol. An ApaI/XbaI fragment from each of the resulting plasmids was then ligated to ApaI/XbaI-digested pcDNA3.1/Myc(-)B vector. These myc fusion constructs were PCR amplified with the original forward primer and the reverse primer MycR (5' GCGGCCGCCAGATCCTCTTC 3'). Following subcloning into pCRII-TOPO, an ApaLI/NotI fragment was ligated into ApaLI/NotI-digested fd-DOG. This scheme resulted in the insertion of orfU fragments, an extra CTA (leucine), and a myc tag immediately 3' of the pIII signal sequence.
fig.ommittedm-, http://www.100md.com
Primers used in construction of orfU-gIII fusionsm-, http://www.100md.com
fig.ommittedm-, http://www.100md.com
(A) Both pIII and OrfU contain two regions of low secondary structure according to the computer algorithm "Predictprotein" located at www.embl-heidelberg.de/predictprotein/predictprotein.html (44). The amino acids in red are these low-complexity regions (LCRs). The areas of LCR for OrfU are from residues 116 to 134 and 252 to 268, and those for pIII are from residues 86 to 104 and 246 to 272. These LCRs of pIII have been previously designated as the linkers that separate the three domains of pIII (15). (B) (Top) Schematic representations of three pIII domains and the three regions in OrfU separated by LCRs. The boundaries for the LCRs of OrfU were determined from the analysis described in the legend to panel A. The residues for the linkers of pIII are as previously described. (Bottom) Schematic of the six pIII-OrfU fusions used to investigate the regions of OrfU required to expand the host range of fd to include V. cholerae. Light blue boxes represent the myc tag added to the 3' end of orfU in each of the fusions. The transduction frequencies were calculated by dividing the number of transductants (tetracycline-resistant colonies) by the number of recipient cells. The TCP+ and TCP- V. cholerae recipients were O395 and TCP2, respectively. TG1 was the F+ E. coli recipient. The transduction frequencies represent the means from at least three experiments, and all standard deviations were less than 10%. UD, undetectable.
pIII-OrfU fusions that lacked the adsorption (g3p-D2) and the penetration (g3p-D1) domains of pIII were also generated. First, a NotI site was introduced into the pIII-OrfU fusion construct just upstream of the linker sequence which separates g3pD2 and g3pD3. The NotI site was added by using the QuikChange XL Site-Directed Mutagenesis kit and the NotIF primer (5' CCTCCTGTCAATGCGGCCGCAGGCTCTGGTGGTGG 3') and its reverse complement, NotIR. The resulting plasmids were digested with NotI to release the g3pD1D2 fragment and were then religated to form the fd{Delta} 1 constructs.2tr], 百拇医药
Molecular biology methods. Standard molecular biology methods were used in this study (1). Restriction enzymes and T4 DNA ligase were purchased from New England Biolabs (Beverly, Mass.). For the Western blot analyses, the anti-myc antibody was purchased from Invitrogen and the anti-pVIII antibody was purchased from Stratagene (La Jolla, Calif.). Southern hybridization was carried out with the ECL direct nucleic acid labeling and detection system (Amersham Pharmacia, Little Chalfont, Buckinghamshire, England) according to the manufacturer's instructions. The low-complexity regions of pIII and OrfU were analyzed by the "Predictprotein" program at the EMBL Heidelberg website (44).
RESULTS+h+0?, http://www.100md.com
Requirement for OrfU for CTX{phi} production. To begin to study the role of OrfU in CTX{phi} virion morphogenesis and infectivity, we made an in-frame orfU deletion in pCTX-Kn, the plasmid form of a kanamycin-marked CTX{phi} . This deletion removed all of orfU except for the sequence encoding the final nine amino acids. This deletion mutant was electroporated into the classical strain O395, which contains a defective prophage, and Bah3, which lacks a CTX prophage and a phage attachment site. Compared to Bah3(pCTX-Kn), no infectious particles were detected in filtered supernatants of Bah3(pCTX-Kn{Delta} orfU) ( third and fourth rows). This defect in CTX{phi} production could be complemented in trans by producing OrfU from a defective CTX prophage (in the classical strain O395) or from a plasmid, pDH5.52 ( second and sixth rows). For the latter experiments, we expressed orfU from the arabinose-inducible promoter pBAD on pDH5.52 in Bah3(pCTX-Kn{Delta} orfU). When arabinose was added to Bah3(pCTX-Kn{Delta} orfU/pDH5.52), infectious CTX-Kn{phi} was produced at levels comparable to those observed for wild-type CTX{phi} virions ( compare sixth row with third row). As expected, virions produced following introduction of exogenous OrfU still carried the orfU deletion and continued to be dependent upon a complementing OrfU for virion production in new hosts infected with CTX-Kn{Delta} orfU{phi} (data not shown). These results indicate that OrfU is required for the production of infectious CTX{phi} s.
fig.ommitted2vmhv, 百拇医药
orfU is required for formation of CTX-Kn{phi} transducing particles2vmhv, 百拇医药
Since pIII is required for Ff phage assembly and release, the absence of Kn-transducing particles in supernatants from Bah3(pCTX-Kn{Delta} orfU) was potentially attributable to a lack of production of CTX-Kn{Delta} orfU virions by this strain. An alternate possibility was that Bah3(pCTX-Kn{Delta} orfU) produced virions that were incapable of infecting a recipient strain. To distinguish between these alternatives, we tested supernatants of Bah3 cells harboring either pCTX-Kn or pCTX-Kn{Delta} orfU for the presence of single-stranded phage DNA by Southern blot analysis. Bah3(pCTX-Kn) supernatants contained abundant CTX{phi} -hybridizing DNA that could be detected even if the DNA was not denatured with NaOH after electrophoresis, suggesting that the hybridizing species was single-stranded phage DNA In marked contrast, very little CTX{phi} -hybridizing DNA was detected in supernatants from cells harboring pCTX-Kn{Delta} orfU ( lanes 4 and 10). However, as in the previous experiment, expression of orfU in trans restored levels of secreted DNA to wild-type levels . The reduction in phage production by Bah3(pCTX-Kn{Delta} orfU) supernatants was not due to a replication defect of pCTX-Kn{Delta} orfU, since pCTX-Kn{Delta} orfU replicated as well as pCTX-Kn. Approximately equal amounts of both the single- and double-stranded forms of phage DNA were detected in plasmid preparations of Bah3(pCTX-Kn) and Bah3(pCTX-Kn{Delta} orfU) ( lanes 1 and 2). These data suggest that the lack of infectious CTX-Kn{phi} in supernatants from Bah3(pCTX-Kn{Delta} orfU) results primarily from a failure in the assembly and/or release of CTX{phi} particles. However, since the small number of virions secreted from Bah3(pCTX-Kn{Delta} orfU) did not yield any transductants, it appears that OrfU is also necessary for CTX{phi} to infect a new host.
fig.ommittedit{\, 百拇医药
Comparison of the amounts of phage and plasmid DNA present in orfU+ and orfU mutant CTX{phi} derivatives by Southern blot analysis using an rstB probe. The samples shown in panel A were denatured in NaOH prior to transfer to the nylon membrane; samples in panel B were not denatured. Lanes 1 and 7 and lanes 2 and 8 contain PmlI-digested plasmid DNA of pCTX-Kn and pCTX-Kn{Delta} orfU, respectively. The remaining lanes are preparations of phage DNA from filtered supernatants: lanes 3 and 9, Bah3(pCTX-Kn); lanes 4 and 10, Bah3(pCTX-Kn{Delta} orfU); lanes 5 and 11, Bah3(pCTX-Kn{Delta} orfU/pDH5.52 + 0.02% Ara); lanes 6 and 12, Bah3(pCTX-Kn {Delta} orfU/pDH5.52 without Ara). After a longer exposure, a faint band with an apparent molecular weight identical to that of the lower band seen in lane 2 was evident in lane 4.it{\, 百拇医药
The role of OrfU in CTX{phi} infection. To investigate whether OrfU functions in CTX{phi} infection, we introduced fragments of orfU into gIII in a tetracycline-marked fd derivative, fd-DOG, that was previously used as a phage display vector (17). This resulted in display of OrfU fused to pIII on the surface of fd-DOG. We adopted this approach for three reasons. First, this method circumvents the requirement for OrfU in CTX{phi} production by V. cholerae because the recombinant fd phages utilize the pIII portion of the fusion protein to be released from E. coli. Second, fd virions containing pIII fusion proteins often retain normal levels of infectivity for F+ E. coli, so phages that failed to infect V. cholerae could still be titered on E. coli (34). Third, we knew that although fd alone cannot infect V. cholerae, it is capable of replication in V. cholerae when introduced by electroporation (unpublished data). Thus, by producing different pIII-OrfU fusion proteins on the surface of fd, we tested whether various portions of OrfU could enable fd to infect V. cholerae.
We compared the amino acid sequences of OrfU and pIII in order to choose segments of OrfU to fuse to pIII. The three functional domains of pIII are separated by stretches of glycine-rich residues that are thought to constitute regions of minimal structural complexity functioning as flexible linkers (16). No such glycine-rich stretches are found in the OrfU sequence; however, there are two regions rich in prolines and serines that could separate OrfU into functionally distinct domains similar in size to those found in pIII. According to the computer algorithm "Predictprotein," these regions (residues 116 to 134 and 252 to 269) in OrfU are unlikely to form a secondary structure and therefore may act as linkers between OrfU domains. This algorithm accurately defined the residues that constitute the linker within pIII r, 百拇医药
We initially fused the first two-thirds of the orfU coding region to gIII in fd-DOG. As shown at the top of this segment of OrfU included the amino acids through the end of the second putative low-complexity region (Fig. 2A). This region was chosen because only the N-terminal two-thirds of pIII, which includes g3p-D1 and g3p-D2, mediates Ff infection of E. coli (29, 30, 35). The orfU N-terminal fragment was introduced 3' of the gIII signal sequence and 5' of the complete gIII coding sequence in fd-DOG, generating fd-pIIICTX(15-274) row 2). A myc tag was inserted at the same time to facilitate detection of the fusion protein. The fd-pIIICTX(15-274) virion was predicted to contain part of OrfU linked to pIII on one tip of the fd virion (10, 25). The OrfU fragment fused to pIII in fd-pIIICTX(15-274) did not alter this phage's ability to infect E. coli; fd-pIIICTX(15-274) infected the F+ E. coli strain TG1 as well as fd-DOG did ( compare rows 1 and 2). However, unlike supernatants from a strain producing fd-DOG , filtered supernatants from an E. coli strain harboring fd-pIIICTX(15-274) could be used to infect V. cholerae . Thus, the N-terminal two-thirds of the OrfU polypeptide fused to pIII is sufficient for expanding the host range of a phage that was previously restricted to F+ E. coli. The ability of fd-pIIICTX(15-274) to infect V. cholerae was dependent on TCP and TolA, the known CTX{phi} coreceptors, as the efficiency of infection by fd-pIIICTX(15-274) of a V. cholerae strain lacking TCP was reduced by 3 orders of magnitude . The infection of a tolA mutant strain (O395 tolA::pGP704) was below the levels of detection for this assay (<10-9). Taken together, these data suggest that OrfU mediates CTX{phi} infection of V. cholerae and that amino acids 15 to 274 of OrfU contain regions that interact with TCP and TolA.
To be certain that the pIII portion of the OrfU-pIII fusion protein did not contribute to infection of V. cholerae by fd-pIIICTX(15-274), we constructed a derivative of this phage, fd{Delta} 1-pIIICTX(15-274), that lacks g3p-D1 and g3p-D2 , row 3). Elimination of these pIII domains did not prevent fd{Delta} 1-pIIICTX(15-274) from infecting V. cholerae . In fact, this deletion appears to enhance the infectivity of fd{Delta} 1-pIIICTX(15-274) relative to fd-pIIICTX(15-274) for both TCP+ and TCP- V. cholerae ( compare rows 2 and 3). Thus, the pIII adsorption (g3p-D2) and penetration (g3p-D1) domains do not contribute to the ability of fd-pIIICTX(15-274) to infect V. cholerae. The reason why fd{Delta} 1-pIIICTX(15-274) infected V. cholerae at a greater efficiency than fd-pIIICTX(15-274) is not clear. It is not likely that unequal numbers of virions were added to the V. cholerae recipients. Western blot analysis of phage proteins with anti-pVIII antisera suggested that approximately equivalent numbers of virions were used in each infection experiment bottom panel, first and fifth lanes; also data not shown). Perhaps residues in g3pD1D2 in the pIII-OrfU(15-274) fusion protein sterically hinder OrfU residues from interacting with TCP and/or TolA.
fig.ommitted4%.?dhk, http://www.100md.com
Comparison of the amounts of pVIII phage protein and pIII-OrfU fusion proteins in phage preparations by Western blot analysis using an anti-myc (top) or anti-pVIII (bottom) antiserum. Phage preparations were resuspended in LB medium, prepared in sample buffer as previously described (1), run on a 10% Bis-Tris gel, and transferred to a nitrocellulose membrane.4%.?dhk, http://www.100md.com
The data presented above suggest that residues 15 to 274 of OrfU contain a region(s) that interacts with both TCP and TolA. Based on the sequence analysis shown in and on experimental studies of pIII, we investigated whether these 259 amino acids could be subdivided into an adsorption domain, for binding the pilus, and a penetration domain, for binding TolA. To test whether the segments flanked by the low-complexity regions can function as autonomous domains, we constructed recombinant fd phages containing OrfU residues 15 to 138 [fd-pIIICTX(15-138)] and 117 to 274 [fd-pIIICTX(117-274)] fused to pIII , respectively) by using the same cloning strategy that we followed to construct fd-pIIICTX(15-274). Derivatives of these new phages that lacked g3p-D1 and g3p-D2 were also generated and designated fd{Delta} 1. Both fd-pIIICTX(15-138) and fd{Delta} 1-pIIICTX(15-138) were capable of infecting TCP+ V. cholerae, although at lower efficiencies than fd-pIIICTX(15-274) and fd{Delta} 1-pIIICTX(15-274). However, unlike the fd derivative that displayed amino acids 15 to 274 of OrfU (, these new derivatives infected TCP+ and TCP- V. cholerae at similar frequencies ( rows 4 and 5). Neither fd-pIIICTX(15-138) nor fd{Delta} 1-pIIICTX(15-138) could infect V. cholerae harboring a mutation in tolA. Together with previous data, these results suggest that a TolA binding domain of OrfU lies within amino acids 15 to 138, since this region is sufficient to mediate TolA-dependent phage infection. They also suggest that amino acids 15 to 138 in and of themselves do not contain a TCP binding domain, since TCP does not influence the infectivity of the fd derivatives displaying these amino acids , rows 4 and 5). Interestingly, fd-pIIICTX(15-138) and fd{Delta} 1-pIIICTX(15-138) infect TCP- V. cholerae better (approximately 10-fold) than comparable phages that contain a fusion protein containing OrfU residues 15 to 274 ( compare rows 4 and 5 with rows 2 and 3). These findings suggest the possibility that OrfU amino acids 139 to 274 can inhibit infection of TCP- V. cholerae.
In contrast to the fd derivatives bearing OrfU residues 15 to 138, those fd derivatives that displayed only residues 117 to 274 of OrfU could not infect V. cholerae ( rows 6 and 7). As judged from the Western blots in, the failure of fd-pIIICTX(117-274) and fd{Delta} 1-pIIICTX(117-274) to infect V. cholerae was not due to low viral titers or degradation of the pIII-OrfU fusion protein. Phages displaying OrfU residues 117 to 274 appear to be as abundant as the other fd derivatives, and they contain a gIII fusion protein of the expected size. It is therefore likely that these fd derivatives were not able to infect V. cholerae because they do not contain OrfU residues 15 to 139 and consequently cannot bind to TolA. Ff derivatives that lack the pIII TolA binding domain are similarly unable to infect E. coli. It is not clear whether fd-pIIICTX(117-274) and fd{Delta} 1-pIIICTX(117-274) are able to bind to TCP. The experiments presented above suggest that OrfU amino acids 117 to 274 are necessary for binding to TCP, but we have not yet determined whether these amino acids are sufficient.
DISCUSSIONl, 百拇医药
We found that OrfU is required for assembly and/or release of CTX virions by V. cholerae and that display of a part of OrfU on fd is sufficient to enable fd to enter V. cholerae. These results suggest that OrfU functions as the pIII equivalent for CTX{phi} . orfU was first described as the open reading frame immediately 5' of ace several years prior to the discovery of CTX{phi} (40); it was designated orfU because its function was unknown. Given our findings, we propose to rename OrfU as pIIICTX and orfU as gIIICTX in order to remain consistent with the Ff nomenclature.l, 百拇医药
As with pIIIfd, different segments of pIIICTX appear to mediate distinct steps in phage infection. pIIICTX residues 15 to 138 were sufficient to allow fd entry into V. cholerae in a TolA-dependent and TCP-independent manner, suggesting that, as with pIIIfd, the amino-terminal end of pIIICTX constitutes its penetration domain. Inclusion of pIIICTX residues 139 to 274 as well as residues 15 to 138 increased infection of V. cholerae by up to 3 orders of magnitude in a TCP-dependent fashion, suggesting that this pIIICTX region includes residues that interact with TCP. However, we were unable to demonstrate a direct interaction between fd-pIIICTX(117-274) and TCP (data not shown). It is not clear whether residues 139 to 274 can function alone as an adsorption domain or whether additional amino acids from pIIICTX are needed as well. If residues 139 to 274 are sufficient to bind TCP, it is possible that proper folding of this region requires additional OrfU residues. In pIIIfd, Deng et al. were able to show functional pilus binding activity for isolated g3p-D2 only after expressing this domain with g3p-D1 and then separating the two domains by protease treatment (7). Thus, the TCP binding activity of pIIICTX residues 139 to 274 may be detectable only when pIIICTX residues 17 to 274 are produced together.
It has been proposed that pIIIfd interacts sequentially with F, then TolA, and that binding to F induces a conformational change in pIII that reveals the TolA binding site (16). Some of our data suggest that a similar multistep process may occur during pIIICTX-mediated infection of V. cholerae. We observed that fd-pIIICTX(15-138) and fd{Delta} 1-pIIICTX(15-138) infected TCP- V. cholerae nearly 1 order of magnitude more efficiently than did fd-pIIICTX(15-274) and fd{Delta} 1-pIIICTX(15-274), which both appear to contain a TCP binding domain. Thus, in the absence of TCP, the putative pIIICTX adsorption domain (residues 139 to 274) may inhibit the penetration domain (amino acids 15 to 138) from binding to TolA. If this model is correct, it suggests that the underlying mechanisms of CTX{phi} and Ff infection of their respective hosts are conserved. However, given the lack of similarity in the amino acid sequences of pIIICTX and pIIIfd, such conservation would probably be based on similarities in the molecular structures of these virion minor coat proteins.
pIIICTX is usually required for CTX{phi} assembly and release, as well as for CTX{phi} infection; however, we were unable to localize the assembly and release functions to a particular segment of the protein. While production of full-length pIIICTX from a plasmid could facilitate release of CTX-Kn{Delta} orfU particles from a strain harboring pCTX-Kn{Delta} orfU, production of the C-terminal two-thirds of pIIICTX did not (data not shown). Therefore, most or even all of pIIICTX seems to be required for assembly and release of CTX{phi} particles; in contrast, only the C-terminal 80 amino acids of pIII are required for fd assembly and release (28).5!6jcx, http://www.100md.com
Since the C-terminal domain of pIII (g3p-D3) is required for fd virion assembly and release, investigation of the role of g3p-D3 in fd infection has not been carried out. Our finding that fd{Delta} 1-pIIICTX(15-274), which contains the adsorption and penetration domains of pIIICTX fused to g3p-D3, infects V. cholerae suggests that the C terminus of pIIICTX does not play a role in infection. However, it is also possible that g3p-D3 substituted for this part of pIIICTX in fd{Delta} 1-pIIICTX(15-274) to allow infection of V. cholerae. Model and colleagues have proposed that g3p-D3 may mediate fusion of the virion with the membrane during entry (28). Similarly, the role of pVIII in fd infection is not clear. pVIII is synthesized as an inner membrane protein (31), and during infection, this virion protein has been shown to be "recycled" into the inner membrane of the recipient cell (33, 39). Our data demonstrate that besides pIIICTX, other CTX{phi} proteins are not required for fd infection of V. cholerae. These data could suggest that the major coat protein is not required for infection, or that pVIII's role in fd infection of E. coli is conserved in infection of V. cholerae.
Our findings that the host range of fd can be altered to include a new genus by the relatively simple technique of either adding or substituting an alternative filamentous phage infection protein for pIIIfd may prove useful in the development of bacteriophage as therapeutic agents. For example, it may be possible to create a library of fd-based phages that are able to infect a broad range of bacterial pathogens. CTX{phi} may also be a useful starting platform from which to engineer such libraries. Furthermore, in addition to using CTX{phi} -based phages as antimicrobial agents, introduction of antigen-encoding DNA sequences into gIIICTX may enable display of antigenic polypeptides in CTX{phi} virions. Production of such recombinant CTX{phi} s during human infection by live attenuated V. cholerae vaccine strains could prove to be an effective method for delivery of heterologous antigens to the gut immune system.y%!, 百拇医药
ACKNOWLEDGMENTSy%!, 百拇医药
We thank B. Davis for critical reading of the manuscript. We thank A. Kane and the NEMC GRASP Center Intestinal Microbiology Core for preparation of media and J. Mekalanos for suggesting that fd would replicate in V. cholerae.
This work was supported by funds from NIH grant AI42347, the Pew Foundation, and the Howard Hughes Medical Institute (to M.K.W.).z@+f, http://www.100md.com
REFERENCESz@+f, http://www.100md.com
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1990. Current protocols in molecular biology. Greene Publishing and Wiley-Interscience, New York, N.Y.z@+f, http://www.100md.com
Beck, E., and B. Zink. 1981. Nucleotide sequence and genome organisation of filamentous bacteriophages fl and fd. Gene 16:35-58.z@+f, http://www.100md.com
Bross, P., K. Bussmann, W. Keppner, and I. Rasched. 1988. Functional analysis of the adsorption protein of two filamentous phages with different host specificities. J. Gen. Microbiol. 134:461-471.z@+f, http://www.100md.com
Click, E. M., and R. E. Webster. 1997. Filamentous phage infection: required interactions with the TolA protein. J. Bacteriol. 179:6464-6471.z@+f, http://www.100md.com
Click, E. M., and R. E. Webster. 1998. The TolQRA proteins are required for membrane insertion of the major capsid protein of the filamentous phage f1 during infection. J. Bacteriol. 180:1723-1728.
Crissman, J. W., and G. P. Smith. 1984. Gene-III protein of filamentous phages: evidence for a carboxyl-terminal domain with a role in morphogenesis. Virology 132:445-455.nu;1/q, http://www.100md.com
Deng, L. W., P. Malik, and R. N. Perham. 1999. Interaction of the globular domains of pIII protein of filamentous bacteriophage fd with the F-pilus of Escherichia coli. Virology 253:271-277.nu;1/q, http://www.100md.com
Endemann, H., P. Bross, and I. Rasched. 1992. The adsorption protein of phage IKe. Localization by deletion mutagenesis of domains involved in infectivity. Mol. Microbiol. 6:471-478.nu;1/q, http://www.100md.com
Germon, P., T. Clavel, A. Vianney, R. Portalier, and J. C. Lazzaroni. 1998. Mutational analysis of the Escherichia coli K-12 TolA N-terminal region and characterization of its TolQ-interacting domain by genetic suppression. J. Bacteriol. 180:6433-6439.nu;1/q, http://www.100md.com
Goldsmith, M. E., and W. H. Konigsberg. 1977. Adsorption protein of the bacteriophage fd: isolation, molecular properties, and location in the virus. Biochemistry 16:2686-2694.
Gray, C. W., R. S. Brown, and D. A. Marvin. 1981. Adsorption complex of filamentous fd virus. J. Mol. Biol. 146:621-627.p7g, 百拇医药
Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.p7g, 百拇医药
Heilpern, A. J., and M. K. Waldor. 2000. CTX{phi} infection of Vibrio cholerae requires the tolQRA gene products. J. Bacteriol. 182:1739-1747.p7g, 百拇医药
Herrington, D. A., R. H. Hall, G. Losonsky, J. J. Mekalanos, R. K. Taylor, and M. M. Levine. 1988. Toxin, toxin-coregulated pili, and the toxR regulon are essential for Vibrio cholerae pathogenesis in humans. J. Exp. Med. 168:1487-1492.p7g, 百拇医药
Holliger, P., and L. Riechmann. 1997. A conserved infection pathway for filamentous bacteriophages is suggested by the structure of the membrane penetration domain of the minor coat protein g3p from phage fd. Structure 5:265-275.p7g, 百拇医药
Holliger, P., L. Riechmann, and R. L. Williams. 1999. Crystal structure of the two N-terminal domains of g3p from filamentous phage fd at 1.9 Å: evidence for conformational lability. J. Mol. Biol. 288:649-657.
Hoogenboom, H. R., A. D. Griffiths, K. S. Johnson, D. J. Chiswell, P. Hudson, and G. Winter. 1991. Multi-subunit proteins on the surface of filamentous phage: methodologies for displaying antibody (Fab) heavy and light chains. Nucleic Acids Res. 19:4133-4137.dv, http://www.100md.com
Horton, R. M., H. D. Hunt, S. N. Ho, J. K. Pullen, and L. R. Pease. 1989. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77:61-68.dv, http://www.100md.com
Jacobson, A. 1972. Role of F pili in the penetration of bacteriophage fl. J. Virol. 10:835-843.dv, http://www.100md.com
Koonin, E. V. 1992. The second cholera toxin, Zot, and its plasmid-encoded and phage-encoded homologues constitute a group of putative ATPases with an altered purine NTP-binding motif. FEBS Lett. 312:3-6.dv, http://www.100md.com
Lazzaroni, J. C., A. Vianney, J. L. Popot, H. Benedetti, F. Samatey, C. Lazdunski, R. Portalier, and V. Geli. 1995. Transmembrane alpha-helix interactions are required for the functional assembly of the Escherichia coli Tol complex. J. Mol. Biol. 246:1-7. rphogenesis of filamentous bacteriophage f1: orientation of extrusion and production of polyphage. Virology 127:177-193.
Lubkowski, J., F. Hennecke, A. Pluckthun, and A. Wlodawer. 1999. Filamentous phage infection: crystal structure of g3p in complex with its coreceptor, the C-terminal domain of TolA. Structure Fold Des. 7:711-722.&g, 百拇医药
Marzari, R., D. Sblattero, M. Righi, and A. Bradbury. 1997. Extending filamentous phage host range by the grafting of a heterologous receptor binding domain. Gene 185:27-33.&g, 百拇医药
McCafferty, J., A. D. Griffiths, G. Winter, and D. J. Chiswell. 1990. Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348:552-554.&g, 百拇医药
Miller, V. L., and J. J. Mekalanos. 1988. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170:2575-2583.&g, 百拇医药
Pearson, G. D., A. Woods, S. L. Chiang, and J. J. Mekalanos. 1993. CTX genetic element encodes a site-specific recombination system and an intestinal colonization factor. Proc. Natl. Acad. Sci. USA 90:3750-3754.
Rakonjac, J., J. Feng, and P. Model. 1999. Filamentous phage are released from the bacterial membrane by a two-step mechanism involving a short C-terminal fragment of pIII. J. Mol. Biol. 289:1253-1265.f#rh, 百拇医药
Rakonjac, J., and P. Model. 1998. Roles of pIII in filamentous phage assembly. J. Mol. Biol. 282:25-41.f#rh, 百拇医药
Riechmann, L., and P. Holliger. 1997. The C-terminal domain of TolA is the coreceptor for filamentous phage infection of E. coli. Cell 90:351-360.f#rh, 百拇医药
Russel, M., and P. Model. 1982. Filamentous phage pre-coat is an integral membrane protein: analysis by a new method of membrane preparation. Cell 28:177-184.f#rh, 百拇医药
Russel, M., H. Whirlow, T. P. Sun, and R. E. Webster. 1988. Low-frequency infection of F- bacteria by transducing particles of filamentous bacteriophages. J. Bacteriol. 170:5312-5316.f#rh, 百拇医药
Smilowitz, H. 1974. Bacteriophage f1 infection: fate of the parental major coat protein. J. Virol. 13:94-99.f#rh, 百拇医药
Smith, G. P. 1985. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228:1315-1317.
Stengele, I., P. Bross, X. Garces, J. Giray, and I. Rasched. 1990. Dissection of functional domains in phage fd adsorption protein. Discrimination between attachment and penetration sites. J. Mol. Biol. 212:143-149.o99, 百拇医药
Sun, T. P., and R. E. Webster. 1986. fii, a bacterial locus required for filamentous phage infection and its relation to colicin-tolerant tolA and tolB. J. Bacteriol. 165:107-115.o99, 百拇医药
Sun, T. P., and R. E. Webster. 1987. Nucleotide sequence of a gene cluster involved in entry of E colicins and single-stranded DNA of infecting filamentous bacteriophages into Escherichia coli. J. Bacteriol. 169:2667-2674.o99, 百拇医药
Taylor, R. K., V. L. Miller, D. B. Furlong, and J. J. Mekalanos. 1987. Use of phoA gene fusions to identify a pilus colonization factor coordinately regulated with cholera toxin. Proc. Natl. Acad. Sci. USA 84:2833-2837.o99, 百拇医药
Trenkner, E., F. Bonhoeffer, and A. Gierer. 1967. The fate of the protein component of bacteriophage fd during infection. Biochem. Biophys. Res. Commun. 28:932-939.
Trucksis, M., J. E. Galen, J. Michalski, A. Fasano, and J. B. Kaper. 1993. Accessory cholera enterotoxin (Ace), the third toxin of a Vibrio cholerae virulence cassette. Proc. Natl. Acad. Sci. USA 90:5267-5271.#, http://www.100md.com
Waldor, M. K., and J. J. Mekalanos. 1996. Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272:1910-1914.#, http://www.100md.com
Webster, R. E. 1991. The tol gene products and the import of macromolecules into Escherichia coli. Mol. Microbiol. 5:1005-1011.#, http://www.100md.com
Woodcock, D. M., P. J. Crowther, J. Doherty, S. Jefferson, E. DeCruz, M. Noyer-Weidner, S. S. Smith, M. Z. Michael, and M. W. Graham. 1989. Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17:3469-3478.#, http://www.100md.com
Wootton, J. C., and S. Federhen. 1996. Analysis of compositionally biased regions in sequence databases. Methods Enzymol. 266:554-571.(Andrew J. Heilpern and Matthew K. Waldor)
Received 19 August 2002/ Accepted 11 November 2002x}-z:3, 百拇医药
ABSTRACTx}-z:3, 百拇医药
CTX{phi} is a filamentous bacteriophage that encodes cholera toxin. CTX{phi} infection of its host bacterium, Vibrio cholerae, requires the toxin-coregulated pilus (TCP) and the products of the V. cholerae tolQRA genes. Here, we have explored the role of OrfU, a predicted CTX{phi} minor coat protein, in CTX{phi} infection. Prior to the discovery that it was part of a prophage, orfU was initially described as an open reading frame of unknown function that lacked similarity to known protein sequences. Based on its size and position in the CTX{phi} genome, we hypothesized that OrfU may function in a manner similar to that of the coliphage fd protein pIII and mediate CTX{phi} infection as well as playing a role in CTX{phi} assembly and release. Deletion of orfU from CTX{phi} dramatically reduced the number of CTX{phi} virions detected in supernatants from CTX{phi} -bearing cells. This defect was complemented by expression of orfU in trans, thereby confirming a role for this gene in CTX{phi} assembly and/or release. To evaluate the requirement for OrfU in CTX{phi} infection, we introduced fragments of orfU into gIII in an fd derivative to create OrfU-pIII fusions. While fd is ordinarily unable to infect V. cholerae, an fd phage displaying the N-terminal 274 amino acids of OrfU could infect V. cholerae in a TCP- and TolA-dependent fashion. Since our findings indicate that OrfU functions as the CTX{phi} pIII, we propose to rename OrfU as pIIICTX. Our data also provide new evidence for a conserved pathway for filamentous phage infection.
INTRODUCTIONd(9o, 百拇医药
The molecular interactions that mediate the entry of Escherichia coli-derived filamentous phages into their hosts have been studied in considerable detail. pIII, a phage minor coat protein found on one tip of the virion, accounts for host specificity in both N-pilus (IKE and I2-2)- and F-pilus (Ff)-specific phages (fd, f1, M13) (3, 8, 11, 24). The initial step in Ff phage infection is mediated by the binding of pIII to the tip of the F pilus (19). The Ff phages do not absolutely require the F pilus for infection; however, Ff infection of F- E. coli occurs at much reduced frequencies (32). The binding of pIII to F is thought to result in pilus retraction, which brings the phage particle in close proximity to and perhaps through the bacterium's outer membrane. Subsequent translocation of Ff DNA through the periplasm and inner membrane requires the products of the tolQRA genes both in F+ and in F- E. coli (32, 36, 37). Recent studies suggest that, in addition to binding to the pilus, pIII also binds to TolA, which serves as a coreceptor (4, 30). Although the exact role of TolQ and TolR in the infection process is unknown, they appear to interact with TolA through their inner membrane spanning regions (9, 21), and they may ordinarily constitute a channel for importing as yet unknown substrates across the inner membrane (42). Following the TolA-pIII interaction, pVIII, the Ff major coat protein, is inserted into the host cell inner membrane, and the phage DNA is translocated into the cytoplasm of the host cell (5, 33).
The 424-amino-acid Ff pIII is thought to consist of a leader sequence (2) and three domains, separated by glycine-rich regions, that serve distinct roles in phage entry and release. The first two pIII domains, designated g3p-D1 (residues 1 to 86) and g3p-D2 (residues 105 to 236), are required for Ff adsorption and entry (7, 30, 35), while the third domain, g3p-D3 (residues 270 to 406), is required for the assembly and release of Ff particles from host cells (6, 29). g3p-D1, the "penetration" domain, and g3p-D2, the "adsorption" domain, have been shown to bind to the Ff coreceptors TolA and F, respectively. Holliger and coworkers have proposed that following binding of g3p-D2 to F, the conformation of pIII changes and that this shift reveals the TolA binding site in g3p-D1. Thus, Ff infection is thought of as a two-step process in which different domains of pIII sequentially bind to F and then to TolA (16, 23).ns}(^, 百拇医药
The C terminus of pIII (g3p-D3) has been implicated in the assembly of Ff particles and in their release from the bacterial membrane. In the absence of pIII, Ff virions remain attached to the cell surface in long "polyphages" (6, 22, 28, 29). According to the model of Rakonjac et al. (28), pIII and pVI (another minor coat protein) are incorporated into assembling phage particles through interactions with pVIII to form a "pretermination complex." Only the last 83 amino acids of pIII are required to form this complex. A conformational change in g3p-D3 may disrupt the association of pIII with the inner membrane, enabling capping and release of virions. g3p-D3 is also required for the production of stable virions that are resistant to certain detergents (28).
CTX{phi} is a Vibrio cholerae-derived lysogenic filamentous phage that contains ctxAB, the operon encoding cholera toxin, in its 7-kb genome. Nontoxigenic V. cholerae strains can be converted to toxigenicity by CTX{phi} infection (41). As in Ff infection of E. coli, CTX{phi} infection of V. cholerae requires a cell surface pilus, toxin-coregulated pilus (TCP), and the periplasmic/inner membrane protein complex TolQRA (13). TCP is a type IV bundle-forming pilus that is also required for V. cholerae intestinal colonization (38). We and others previously found that inactivation of either tcpA or the tolQRA genes of V. cholerae reduced CTX{phi} infection by more than 3 orders of magnitude (13, 41). Given the similarities of the requirements for CTX{phi} infection of V. cholerae to those for Ff infection of E. coli, we hypothesized that CTX{phi} might contain a virion coat protein that functions analogously to pIII and interacts with TCP and the TolQRA complex to enable CTX{phi} infection of V. cholerae.
The CTX{phi} core region encodes proteins for virion morphogenesis, including the putative major coat protein (Cep), three putative minor coat proteins (Psh, OrfU, and Ace), and Zot (20), which is similar to the Ff protein pI and is required for phage assembly and secretion. Although several of the putative CTX{phi} gene products bear similarity to known filamentous phage proteins, none of the CTX{phi} genes encode a protein similar in sequence to pIII. However, since pIII and OrfU have similar sizes and positions within their respective genomes (immediately 3' of the gene encoding the major coat protein), we hypothesized that OrfU is the CTX{phi} pIII homologue. Recently, Holliger and Reichmann determined the solution structure of g3p-D1 by nuclear magnetic resonance spectroscopy. They predicted that residues 18 to 86 of OrfU may fold into a ß barrel similar to that of g3p-D1 (15). Their analysis lends additional credence to the hypothesis that OrfU is the CTX{phi} pIII equivalent and to introducing the idea that these particular OrfU residues are involved in the binding of OrfU to the V. cholerae TolA (15). To test the hypothesis that OrfU mediates CTX{phi} infection, we "displayed" OrfU on the surface of an Ff phage as an OrfU-pIII fusion protein. The resulting phage infected both F+ E. coli and TCP+ V. cholerae. Furthermore, OrfU was found to be required for CTX{phi} production by V. cholerae. These findings indicate that OrfU functions as the CTX{phi} pIII analogue, and we propose to rename OrfU as pIIICTX.
MATERIALS AND METHODS(7u, 百拇医药
Strains, media, and antibiotics. The bacterial strains used in this study are listed in Table 1. Bacteria were grown in Luria-Bertani (LB) broth at 37°C. To induce TCP production, the classical V. cholerae strain O395 was grown on a roller drum shaker at 30°C as previously described (13). To ensure F pilus production by E. coli strain TG1, the strain was maintained on M9 minimal medium plates containing glucose. Antibiotics were used at the following concentrations: ampicillin, 50 µg/ml (V. cholerae) and 100 µg/ml (E. coli); streptomycin, 200 µg/ml; kanamycin, 50 µg/ml; chloramphenicol, 15 µg/ml (E. coli) and 1 µg/ml (V. cholerae); tetracycline, 20 µg/ml (E. coli) and 1 µg/ml (V. cholerae). Arabinose (Ara) (0.02%) was added to LB broth to induce expression of genes under the control of the pBAD promoter.(7u, 百拇医药
fig.ommitted(7u, 百拇医药
List of strains and plasmids used in this study
Construction of pCTX-Kan{Delta} orfU and an orfU expression plasmid. An in-frame deletion of orfU in pCTX-Kan was constructed using the PCR technique of splicing by overlapping extension (18). The two outside primers were DelUA (5' GCGATAAGGGGCTTTATTGGTATG 3') and DelUD (5' CATTCGTGATGATGTGACGGC 3'). The two internal primers were DelUB2 (5' GAGCATGGCGGCCAGAAAAAAGGGCGAACG 3') and DelUC2 (5' CGTTCGCCCTTTTTTCTGGCCATGCTC 3'). After subcloning of the deletion PCR product into pCRII-TOPO, an AscI/PmlI fragment from this plasmid was ligated with AscI/PmlI-digested pCTX-Kn. The resulting plasmid lacks all but the last nine amino acids of OrfU. An orfU complementing plasmid, pDH5.52, was constructed by amplifying orfU with the forward and reverse primers OrfU1F (5' ATGCGCTATTTTCTACTGTTTTTG 3') and OrfU4R (5' TCTAGAAAATCACCTAAACAAAATGAGCATG 3'). After subcloning of the PCR product into pCRII-TOPO (Invitrogen, Carlsbad, Calif.), an XbaI/EcoRI fragment from this plasmid was ligated with XbaI/EcoRI-digested pBAD24.
Transduction assay. CTX-Kn{phi} and CTX-Kn{Delta} orfU{phi} transducing particles were enumerated using a previously described transduction assay (13). Briefly, filtered supernatants were prepared from cultures grown to an optical density at 600 nm (OD600) of approximately 0.8. A 75-µl volume of the filtered supernatants from these cultures was mixed with 75 µl of TCP+ O395 cells for 20 min at room temperature on a shaker. Each mixture was then plated on LB agar containing either streptomycin, to enumerate the number of recipient cells, or kanamycin, to enumerate the transductants. The transduction frequency was calculated by dividing the number of transductants by the number of recipients. Similar transduction assays were used to calculate the numbers of the fd-derived transducing phages we studied here. To prepare these phages, DH5pir harboring the different fd-derived phages was grown overnight in LB broth and tetracycline. Phage particles were then precipitated from filtered supernatants with 5% polyethylene glycol (PEG) in 2.5 M NaCl overnight at 4°C, followed by centrifugation at 10,000 x g for 30 min. The particles were then resuspended in LB broth. To enumerate transductants, cell mixtures were plated on LB agar plates containing tetracycline, and to enumerate recipients, the mixtures were plated on LB plates either alone (for TG1) or with streptomycin (for the V. cholerae strains). The transduction frequency was calculated as described above.
Detection of phage DNA in supernatants. Southern blot analysis was used to detect phage DNA in filtered supernatants from mid-log-phase cultures of Bah3 harboring pCTX-Kn derivatives. The phage particles were PEG precipitated as described above and then resuspended in a sodium dodecyl sulfate (SDS)-containing buffer (1% SDS, 1x Tris-acetate-EDTA [TAE], 5% glycerol, 0.25% loading dye) at 70°C for 20 min to disrupt virions (29). These preparations were then subjected to 1% agarose electrophoresis, and CTX{phi} DNA was then detected by Southern analysis using the CTX{phi} rstB gene as a probe. The rstB gene was amplified with primers RstBF (5' ATGAAATTATGGGTGATTAAT 3') and RstBB (5' CCACCGCGAAATAGAGCGCCT 3'). For the nondenaturing Southern blots, the NaOH treatment of the gel was omitted prior to transfer.a/;g, 百拇医药
Construction of fd-DOG derivatives containing OrfU-pIII fusions. Our strategy was to amplify different regions of orfU and subclone them into the pcDNA3.1/Myc(-)B vector to add a myc epitope to the 3' end of the construct. We then inserted this downstream of the leader sequence of gIII in fd-DOG. The primers used for these constructs are listed in . After the first round of amplification, the PCR products were ligated into the TA cloning vector pCRII-TOPO according to the manufacturer's protocol. An ApaI/XbaI fragment from each of the resulting plasmids was then ligated to ApaI/XbaI-digested pcDNA3.1/Myc(-)B vector. These myc fusion constructs were PCR amplified with the original forward primer and the reverse primer MycR (5' GCGGCCGCCAGATCCTCTTC 3'). Following subcloning into pCRII-TOPO, an ApaLI/NotI fragment was ligated into ApaLI/NotI-digested fd-DOG. This scheme resulted in the insertion of orfU fragments, an extra CTA (leucine), and a myc tag immediately 3' of the pIII signal sequence.
fig.ommittedm-, http://www.100md.com
Primers used in construction of orfU-gIII fusionsm-, http://www.100md.com
fig.ommittedm-, http://www.100md.com
(A) Both pIII and OrfU contain two regions of low secondary structure according to the computer algorithm "Predictprotein" located at www.embl-heidelberg.de/predictprotein/predictprotein.html (44). The amino acids in red are these low-complexity regions (LCRs). The areas of LCR for OrfU are from residues 116 to 134 and 252 to 268, and those for pIII are from residues 86 to 104 and 246 to 272. These LCRs of pIII have been previously designated as the linkers that separate the three domains of pIII (15). (B) (Top) Schematic representations of three pIII domains and the three regions in OrfU separated by LCRs. The boundaries for the LCRs of OrfU were determined from the analysis described in the legend to panel A. The residues for the linkers of pIII are as previously described. (Bottom) Schematic of the six pIII-OrfU fusions used to investigate the regions of OrfU required to expand the host range of fd to include V. cholerae. Light blue boxes represent the myc tag added to the 3' end of orfU in each of the fusions. The transduction frequencies were calculated by dividing the number of transductants (tetracycline-resistant colonies) by the number of recipient cells. The TCP+ and TCP- V. cholerae recipients were O395 and TCP2, respectively. TG1 was the F+ E. coli recipient. The transduction frequencies represent the means from at least three experiments, and all standard deviations were less than 10%. UD, undetectable.
pIII-OrfU fusions that lacked the adsorption (g3p-D2) and the penetration (g3p-D1) domains of pIII were also generated. First, a NotI site was introduced into the pIII-OrfU fusion construct just upstream of the linker sequence which separates g3pD2 and g3pD3. The NotI site was added by using the QuikChange XL Site-Directed Mutagenesis kit and the NotIF primer (5' CCTCCTGTCAATGCGGCCGCAGGCTCTGGTGGTGG 3') and its reverse complement, NotIR. The resulting plasmids were digested with NotI to release the g3pD1D2 fragment and were then religated to form the fd{Delta} 1 constructs.2tr], 百拇医药
Molecular biology methods. Standard molecular biology methods were used in this study (1). Restriction enzymes and T4 DNA ligase were purchased from New England Biolabs (Beverly, Mass.). For the Western blot analyses, the anti-myc antibody was purchased from Invitrogen and the anti-pVIII antibody was purchased from Stratagene (La Jolla, Calif.). Southern hybridization was carried out with the ECL direct nucleic acid labeling and detection system (Amersham Pharmacia, Little Chalfont, Buckinghamshire, England) according to the manufacturer's instructions. The low-complexity regions of pIII and OrfU were analyzed by the "Predictprotein" program at the EMBL Heidelberg website (44).
RESULTS+h+0?, http://www.100md.com
Requirement for OrfU for CTX{phi} production. To begin to study the role of OrfU in CTX{phi} virion morphogenesis and infectivity, we made an in-frame orfU deletion in pCTX-Kn, the plasmid form of a kanamycin-marked CTX{phi} . This deletion removed all of orfU except for the sequence encoding the final nine amino acids. This deletion mutant was electroporated into the classical strain O395, which contains a defective prophage, and Bah3, which lacks a CTX prophage and a phage attachment site. Compared to Bah3(pCTX-Kn), no infectious particles were detected in filtered supernatants of Bah3(pCTX-Kn{Delta} orfU) ( third and fourth rows). This defect in CTX{phi} production could be complemented in trans by producing OrfU from a defective CTX prophage (in the classical strain O395) or from a plasmid, pDH5.52 ( second and sixth rows). For the latter experiments, we expressed orfU from the arabinose-inducible promoter pBAD on pDH5.52 in Bah3(pCTX-Kn{Delta} orfU). When arabinose was added to Bah3(pCTX-Kn{Delta} orfU/pDH5.52), infectious CTX-Kn{phi} was produced at levels comparable to those observed for wild-type CTX{phi} virions ( compare sixth row with third row). As expected, virions produced following introduction of exogenous OrfU still carried the orfU deletion and continued to be dependent upon a complementing OrfU for virion production in new hosts infected with CTX-Kn{Delta} orfU{phi} (data not shown). These results indicate that OrfU is required for the production of infectious CTX{phi} s.
fig.ommitted2vmhv, 百拇医药
orfU is required for formation of CTX-Kn{phi} transducing particles2vmhv, 百拇医药
Since pIII is required for Ff phage assembly and release, the absence of Kn-transducing particles in supernatants from Bah3(pCTX-Kn{Delta} orfU) was potentially attributable to a lack of production of CTX-Kn{Delta} orfU virions by this strain. An alternate possibility was that Bah3(pCTX-Kn{Delta} orfU) produced virions that were incapable of infecting a recipient strain. To distinguish between these alternatives, we tested supernatants of Bah3 cells harboring either pCTX-Kn or pCTX-Kn{Delta} orfU for the presence of single-stranded phage DNA by Southern blot analysis. Bah3(pCTX-Kn) supernatants contained abundant CTX{phi} -hybridizing DNA that could be detected even if the DNA was not denatured with NaOH after electrophoresis, suggesting that the hybridizing species was single-stranded phage DNA In marked contrast, very little CTX{phi} -hybridizing DNA was detected in supernatants from cells harboring pCTX-Kn{Delta} orfU ( lanes 4 and 10). However, as in the previous experiment, expression of orfU in trans restored levels of secreted DNA to wild-type levels . The reduction in phage production by Bah3(pCTX-Kn{Delta} orfU) supernatants was not due to a replication defect of pCTX-Kn{Delta} orfU, since pCTX-Kn{Delta} orfU replicated as well as pCTX-Kn. Approximately equal amounts of both the single- and double-stranded forms of phage DNA were detected in plasmid preparations of Bah3(pCTX-Kn) and Bah3(pCTX-Kn{Delta} orfU) ( lanes 1 and 2). These data suggest that the lack of infectious CTX-Kn{phi} in supernatants from Bah3(pCTX-Kn{Delta} orfU) results primarily from a failure in the assembly and/or release of CTX{phi} particles. However, since the small number of virions secreted from Bah3(pCTX-Kn{Delta} orfU) did not yield any transductants, it appears that OrfU is also necessary for CTX{phi} to infect a new host.
fig.ommittedit{\, 百拇医药
Comparison of the amounts of phage and plasmid DNA present in orfU+ and orfU mutant CTX{phi} derivatives by Southern blot analysis using an rstB probe. The samples shown in panel A were denatured in NaOH prior to transfer to the nylon membrane; samples in panel B were not denatured. Lanes 1 and 7 and lanes 2 and 8 contain PmlI-digested plasmid DNA of pCTX-Kn and pCTX-Kn{Delta} orfU, respectively. The remaining lanes are preparations of phage DNA from filtered supernatants: lanes 3 and 9, Bah3(pCTX-Kn); lanes 4 and 10, Bah3(pCTX-Kn{Delta} orfU); lanes 5 and 11, Bah3(pCTX-Kn{Delta} orfU/pDH5.52 + 0.02% Ara); lanes 6 and 12, Bah3(pCTX-Kn {Delta} orfU/pDH5.52 without Ara). After a longer exposure, a faint band with an apparent molecular weight identical to that of the lower band seen in lane 2 was evident in lane 4.it{\, 百拇医药
The role of OrfU in CTX{phi} infection. To investigate whether OrfU functions in CTX{phi} infection, we introduced fragments of orfU into gIII in a tetracycline-marked fd derivative, fd-DOG, that was previously used as a phage display vector (17). This resulted in display of OrfU fused to pIII on the surface of fd-DOG. We adopted this approach for three reasons. First, this method circumvents the requirement for OrfU in CTX{phi} production by V. cholerae because the recombinant fd phages utilize the pIII portion of the fusion protein to be released from E. coli. Second, fd virions containing pIII fusion proteins often retain normal levels of infectivity for F+ E. coli, so phages that failed to infect V. cholerae could still be titered on E. coli (34). Third, we knew that although fd alone cannot infect V. cholerae, it is capable of replication in V. cholerae when introduced by electroporation (unpublished data). Thus, by producing different pIII-OrfU fusion proteins on the surface of fd, we tested whether various portions of OrfU could enable fd to infect V. cholerae.
We compared the amino acid sequences of OrfU and pIII in order to choose segments of OrfU to fuse to pIII. The three functional domains of pIII are separated by stretches of glycine-rich residues that are thought to constitute regions of minimal structural complexity functioning as flexible linkers (16). No such glycine-rich stretches are found in the OrfU sequence; however, there are two regions rich in prolines and serines that could separate OrfU into functionally distinct domains similar in size to those found in pIII. According to the computer algorithm "Predictprotein," these regions (residues 116 to 134 and 252 to 269) in OrfU are unlikely to form a secondary structure and therefore may act as linkers between OrfU domains. This algorithm accurately defined the residues that constitute the linker within pIII r, 百拇医药
We initially fused the first two-thirds of the orfU coding region to gIII in fd-DOG. As shown at the top of this segment of OrfU included the amino acids through the end of the second putative low-complexity region (Fig. 2A). This region was chosen because only the N-terminal two-thirds of pIII, which includes g3p-D1 and g3p-D2, mediates Ff infection of E. coli (29, 30, 35). The orfU N-terminal fragment was introduced 3' of the gIII signal sequence and 5' of the complete gIII coding sequence in fd-DOG, generating fd-pIIICTX(15-274) row 2). A myc tag was inserted at the same time to facilitate detection of the fusion protein. The fd-pIIICTX(15-274) virion was predicted to contain part of OrfU linked to pIII on one tip of the fd virion (10, 25). The OrfU fragment fused to pIII in fd-pIIICTX(15-274) did not alter this phage's ability to infect E. coli; fd-pIIICTX(15-274) infected the F+ E. coli strain TG1 as well as fd-DOG did ( compare rows 1 and 2). However, unlike supernatants from a strain producing fd-DOG , filtered supernatants from an E. coli strain harboring fd-pIIICTX(15-274) could be used to infect V. cholerae . Thus, the N-terminal two-thirds of the OrfU polypeptide fused to pIII is sufficient for expanding the host range of a phage that was previously restricted to F+ E. coli. The ability of fd-pIIICTX(15-274) to infect V. cholerae was dependent on TCP and TolA, the known CTX{phi} coreceptors, as the efficiency of infection by fd-pIIICTX(15-274) of a V. cholerae strain lacking TCP was reduced by 3 orders of magnitude . The infection of a tolA mutant strain (O395 tolA::pGP704) was below the levels of detection for this assay (<10-9). Taken together, these data suggest that OrfU mediates CTX{phi} infection of V. cholerae and that amino acids 15 to 274 of OrfU contain regions that interact with TCP and TolA.
To be certain that the pIII portion of the OrfU-pIII fusion protein did not contribute to infection of V. cholerae by fd-pIIICTX(15-274), we constructed a derivative of this phage, fd{Delta} 1-pIIICTX(15-274), that lacks g3p-D1 and g3p-D2 , row 3). Elimination of these pIII domains did not prevent fd{Delta} 1-pIIICTX(15-274) from infecting V. cholerae . In fact, this deletion appears to enhance the infectivity of fd{Delta} 1-pIIICTX(15-274) relative to fd-pIIICTX(15-274) for both TCP+ and TCP- V. cholerae ( compare rows 2 and 3). Thus, the pIII adsorption (g3p-D2) and penetration (g3p-D1) domains do not contribute to the ability of fd-pIIICTX(15-274) to infect V. cholerae. The reason why fd{Delta} 1-pIIICTX(15-274) infected V. cholerae at a greater efficiency than fd-pIIICTX(15-274) is not clear. It is not likely that unequal numbers of virions were added to the V. cholerae recipients. Western blot analysis of phage proteins with anti-pVIII antisera suggested that approximately equivalent numbers of virions were used in each infection experiment bottom panel, first and fifth lanes; also data not shown). Perhaps residues in g3pD1D2 in the pIII-OrfU(15-274) fusion protein sterically hinder OrfU residues from interacting with TCP and/or TolA.
fig.ommitted4%.?dhk, http://www.100md.com
Comparison of the amounts of pVIII phage protein and pIII-OrfU fusion proteins in phage preparations by Western blot analysis using an anti-myc (top) or anti-pVIII (bottom) antiserum. Phage preparations were resuspended in LB medium, prepared in sample buffer as previously described (1), run on a 10% Bis-Tris gel, and transferred to a nitrocellulose membrane.4%.?dhk, http://www.100md.com
The data presented above suggest that residues 15 to 274 of OrfU contain a region(s) that interacts with both TCP and TolA. Based on the sequence analysis shown in and on experimental studies of pIII, we investigated whether these 259 amino acids could be subdivided into an adsorption domain, for binding the pilus, and a penetration domain, for binding TolA. To test whether the segments flanked by the low-complexity regions can function as autonomous domains, we constructed recombinant fd phages containing OrfU residues 15 to 138 [fd-pIIICTX(15-138)] and 117 to 274 [fd-pIIICTX(117-274)] fused to pIII , respectively) by using the same cloning strategy that we followed to construct fd-pIIICTX(15-274). Derivatives of these new phages that lacked g3p-D1 and g3p-D2 were also generated and designated fd{Delta} 1. Both fd-pIIICTX(15-138) and fd{Delta} 1-pIIICTX(15-138) were capable of infecting TCP+ V. cholerae, although at lower efficiencies than fd-pIIICTX(15-274) and fd{Delta} 1-pIIICTX(15-274). However, unlike the fd derivative that displayed amino acids 15 to 274 of OrfU (, these new derivatives infected TCP+ and TCP- V. cholerae at similar frequencies ( rows 4 and 5). Neither fd-pIIICTX(15-138) nor fd{Delta} 1-pIIICTX(15-138) could infect V. cholerae harboring a mutation in tolA. Together with previous data, these results suggest that a TolA binding domain of OrfU lies within amino acids 15 to 138, since this region is sufficient to mediate TolA-dependent phage infection. They also suggest that amino acids 15 to 138 in and of themselves do not contain a TCP binding domain, since TCP does not influence the infectivity of the fd derivatives displaying these amino acids , rows 4 and 5). Interestingly, fd-pIIICTX(15-138) and fd{Delta} 1-pIIICTX(15-138) infect TCP- V. cholerae better (approximately 10-fold) than comparable phages that contain a fusion protein containing OrfU residues 15 to 274 ( compare rows 4 and 5 with rows 2 and 3). These findings suggest the possibility that OrfU amino acids 139 to 274 can inhibit infection of TCP- V. cholerae.
In contrast to the fd derivatives bearing OrfU residues 15 to 138, those fd derivatives that displayed only residues 117 to 274 of OrfU could not infect V. cholerae ( rows 6 and 7). As judged from the Western blots in, the failure of fd-pIIICTX(117-274) and fd{Delta} 1-pIIICTX(117-274) to infect V. cholerae was not due to low viral titers or degradation of the pIII-OrfU fusion protein. Phages displaying OrfU residues 117 to 274 appear to be as abundant as the other fd derivatives, and they contain a gIII fusion protein of the expected size. It is therefore likely that these fd derivatives were not able to infect V. cholerae because they do not contain OrfU residues 15 to 139 and consequently cannot bind to TolA. Ff derivatives that lack the pIII TolA binding domain are similarly unable to infect E. coli. It is not clear whether fd-pIIICTX(117-274) and fd{Delta} 1-pIIICTX(117-274) are able to bind to TCP. The experiments presented above suggest that OrfU amino acids 117 to 274 are necessary for binding to TCP, but we have not yet determined whether these amino acids are sufficient.
DISCUSSIONl, 百拇医药
We found that OrfU is required for assembly and/or release of CTX virions by V. cholerae and that display of a part of OrfU on fd is sufficient to enable fd to enter V. cholerae. These results suggest that OrfU functions as the pIII equivalent for CTX{phi} . orfU was first described as the open reading frame immediately 5' of ace several years prior to the discovery of CTX{phi} (40); it was designated orfU because its function was unknown. Given our findings, we propose to rename OrfU as pIIICTX and orfU as gIIICTX in order to remain consistent with the Ff nomenclature.l, 百拇医药
As with pIIIfd, different segments of pIIICTX appear to mediate distinct steps in phage infection. pIIICTX residues 15 to 138 were sufficient to allow fd entry into V. cholerae in a TolA-dependent and TCP-independent manner, suggesting that, as with pIIIfd, the amino-terminal end of pIIICTX constitutes its penetration domain. Inclusion of pIIICTX residues 139 to 274 as well as residues 15 to 138 increased infection of V. cholerae by up to 3 orders of magnitude in a TCP-dependent fashion, suggesting that this pIIICTX region includes residues that interact with TCP. However, we were unable to demonstrate a direct interaction between fd-pIIICTX(117-274) and TCP (data not shown). It is not clear whether residues 139 to 274 can function alone as an adsorption domain or whether additional amino acids from pIIICTX are needed as well. If residues 139 to 274 are sufficient to bind TCP, it is possible that proper folding of this region requires additional OrfU residues. In pIIIfd, Deng et al. were able to show functional pilus binding activity for isolated g3p-D2 only after expressing this domain with g3p-D1 and then separating the two domains by protease treatment (7). Thus, the TCP binding activity of pIIICTX residues 139 to 274 may be detectable only when pIIICTX residues 17 to 274 are produced together.
It has been proposed that pIIIfd interacts sequentially with F, then TolA, and that binding to F induces a conformational change in pIII that reveals the TolA binding site (16). Some of our data suggest that a similar multistep process may occur during pIIICTX-mediated infection of V. cholerae. We observed that fd-pIIICTX(15-138) and fd{Delta} 1-pIIICTX(15-138) infected TCP- V. cholerae nearly 1 order of magnitude more efficiently than did fd-pIIICTX(15-274) and fd{Delta} 1-pIIICTX(15-274), which both appear to contain a TCP binding domain. Thus, in the absence of TCP, the putative pIIICTX adsorption domain (residues 139 to 274) may inhibit the penetration domain (amino acids 15 to 138) from binding to TolA. If this model is correct, it suggests that the underlying mechanisms of CTX{phi} and Ff infection of their respective hosts are conserved. However, given the lack of similarity in the amino acid sequences of pIIICTX and pIIIfd, such conservation would probably be based on similarities in the molecular structures of these virion minor coat proteins.
pIIICTX is usually required for CTX{phi} assembly and release, as well as for CTX{phi} infection; however, we were unable to localize the assembly and release functions to a particular segment of the protein. While production of full-length pIIICTX from a plasmid could facilitate release of CTX-Kn{Delta} orfU particles from a strain harboring pCTX-Kn{Delta} orfU, production of the C-terminal two-thirds of pIIICTX did not (data not shown). Therefore, most or even all of pIIICTX seems to be required for assembly and release of CTX{phi} particles; in contrast, only the C-terminal 80 amino acids of pIII are required for fd assembly and release (28).5!6jcx, http://www.100md.com
Since the C-terminal domain of pIII (g3p-D3) is required for fd virion assembly and release, investigation of the role of g3p-D3 in fd infection has not been carried out. Our finding that fd{Delta} 1-pIIICTX(15-274), which contains the adsorption and penetration domains of pIIICTX fused to g3p-D3, infects V. cholerae suggests that the C terminus of pIIICTX does not play a role in infection. However, it is also possible that g3p-D3 substituted for this part of pIIICTX in fd{Delta} 1-pIIICTX(15-274) to allow infection of V. cholerae. Model and colleagues have proposed that g3p-D3 may mediate fusion of the virion with the membrane during entry (28). Similarly, the role of pVIII in fd infection is not clear. pVIII is synthesized as an inner membrane protein (31), and during infection, this virion protein has been shown to be "recycled" into the inner membrane of the recipient cell (33, 39). Our data demonstrate that besides pIIICTX, other CTX{phi} proteins are not required for fd infection of V. cholerae. These data could suggest that the major coat protein is not required for infection, or that pVIII's role in fd infection of E. coli is conserved in infection of V. cholerae.
Our findings that the host range of fd can be altered to include a new genus by the relatively simple technique of either adding or substituting an alternative filamentous phage infection protein for pIIIfd may prove useful in the development of bacteriophage as therapeutic agents. For example, it may be possible to create a library of fd-based phages that are able to infect a broad range of bacterial pathogens. CTX{phi} may also be a useful starting platform from which to engineer such libraries. Furthermore, in addition to using CTX{phi} -based phages as antimicrobial agents, introduction of antigen-encoding DNA sequences into gIIICTX may enable display of antigenic polypeptides in CTX{phi} virions. Production of such recombinant CTX{phi} s during human infection by live attenuated V. cholerae vaccine strains could prove to be an effective method for delivery of heterologous antigens to the gut immune system.y%!, 百拇医药
ACKNOWLEDGMENTSy%!, 百拇医药
We thank B. Davis for critical reading of the manuscript. We thank A. Kane and the NEMC GRASP Center Intestinal Microbiology Core for preparation of media and J. Mekalanos for suggesting that fd would replicate in V. cholerae.
This work was supported by funds from NIH grant AI42347, the Pew Foundation, and the Howard Hughes Medical Institute (to M.K.W.).z@+f, http://www.100md.com
REFERENCESz@+f, http://www.100md.com
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1990. Current protocols in molecular biology. Greene Publishing and Wiley-Interscience, New York, N.Y.z@+f, http://www.100md.com
Beck, E., and B. Zink. 1981. Nucleotide sequence and genome organisation of filamentous bacteriophages fl and fd. Gene 16:35-58.z@+f, http://www.100md.com
Bross, P., K. Bussmann, W. Keppner, and I. Rasched. 1988. Functional analysis of the adsorption protein of two filamentous phages with different host specificities. J. Gen. Microbiol. 134:461-471.z@+f, http://www.100md.com
Click, E. M., and R. E. Webster. 1997. Filamentous phage infection: required interactions with the TolA protein. J. Bacteriol. 179:6464-6471.z@+f, http://www.100md.com
Click, E. M., and R. E. Webster. 1998. The TolQRA proteins are required for membrane insertion of the major capsid protein of the filamentous phage f1 during infection. J. Bacteriol. 180:1723-1728.
Crissman, J. W., and G. P. Smith. 1984. Gene-III protein of filamentous phages: evidence for a carboxyl-terminal domain with a role in morphogenesis. Virology 132:445-455.nu;1/q, http://www.100md.com
Deng, L. W., P. Malik, and R. N. Perham. 1999. Interaction of the globular domains of pIII protein of filamentous bacteriophage fd with the F-pilus of Escherichia coli. Virology 253:271-277.nu;1/q, http://www.100md.com
Endemann, H., P. Bross, and I. Rasched. 1992. The adsorption protein of phage IKe. Localization by deletion mutagenesis of domains involved in infectivity. Mol. Microbiol. 6:471-478.nu;1/q, http://www.100md.com
Germon, P., T. Clavel, A. Vianney, R. Portalier, and J. C. Lazzaroni. 1998. Mutational analysis of the Escherichia coli K-12 TolA N-terminal region and characterization of its TolQ-interacting domain by genetic suppression. J. Bacteriol. 180:6433-6439.nu;1/q, http://www.100md.com
Goldsmith, M. E., and W. H. Konigsberg. 1977. Adsorption protein of the bacteriophage fd: isolation, molecular properties, and location in the virus. Biochemistry 16:2686-2694.
Gray, C. W., R. S. Brown, and D. A. Marvin. 1981. Adsorption complex of filamentous fd virus. J. Mol. Biol. 146:621-627.p7g, 百拇医药
Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.p7g, 百拇医药
Heilpern, A. J., and M. K. Waldor. 2000. CTX{phi} infection of Vibrio cholerae requires the tolQRA gene products. J. Bacteriol. 182:1739-1747.p7g, 百拇医药
Herrington, D. A., R. H. Hall, G. Losonsky, J. J. Mekalanos, R. K. Taylor, and M. M. Levine. 1988. Toxin, toxin-coregulated pili, and the toxR regulon are essential for Vibrio cholerae pathogenesis in humans. J. Exp. Med. 168:1487-1492.p7g, 百拇医药
Holliger, P., and L. Riechmann. 1997. A conserved infection pathway for filamentous bacteriophages is suggested by the structure of the membrane penetration domain of the minor coat protein g3p from phage fd. Structure 5:265-275.p7g, 百拇医药
Holliger, P., L. Riechmann, and R. L. Williams. 1999. Crystal structure of the two N-terminal domains of g3p from filamentous phage fd at 1.9 Å: evidence for conformational lability. J. Mol. Biol. 288:649-657.
Hoogenboom, H. R., A. D. Griffiths, K. S. Johnson, D. J. Chiswell, P. Hudson, and G. Winter. 1991. Multi-subunit proteins on the surface of filamentous phage: methodologies for displaying antibody (Fab) heavy and light chains. Nucleic Acids Res. 19:4133-4137.dv, http://www.100md.com
Horton, R. M., H. D. Hunt, S. N. Ho, J. K. Pullen, and L. R. Pease. 1989. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77:61-68.dv, http://www.100md.com
Jacobson, A. 1972. Role of F pili in the penetration of bacteriophage fl. J. Virol. 10:835-843.dv, http://www.100md.com
Koonin, E. V. 1992. The second cholera toxin, Zot, and its plasmid-encoded and phage-encoded homologues constitute a group of putative ATPases with an altered purine NTP-binding motif. FEBS Lett. 312:3-6.dv, http://www.100md.com
Lazzaroni, J. C., A. Vianney, J. L. Popot, H. Benedetti, F. Samatey, C. Lazdunski, R. Portalier, and V. Geli. 1995. Transmembrane alpha-helix interactions are required for the functional assembly of the Escherichia coli Tol complex. J. Mol. Biol. 246:1-7. rphogenesis of filamentous bacteriophage f1: orientation of extrusion and production of polyphage. Virology 127:177-193.
Lubkowski, J., F. Hennecke, A. Pluckthun, and A. Wlodawer. 1999. Filamentous phage infection: crystal structure of g3p in complex with its coreceptor, the C-terminal domain of TolA. Structure Fold Des. 7:711-722.&g, 百拇医药
Marzari, R., D. Sblattero, M. Righi, and A. Bradbury. 1997. Extending filamentous phage host range by the grafting of a heterologous receptor binding domain. Gene 185:27-33.&g, 百拇医药
McCafferty, J., A. D. Griffiths, G. Winter, and D. J. Chiswell. 1990. Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348:552-554.&g, 百拇医药
Miller, V. L., and J. J. Mekalanos. 1988. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170:2575-2583.&g, 百拇医药
Pearson, G. D., A. Woods, S. L. Chiang, and J. J. Mekalanos. 1993. CTX genetic element encodes a site-specific recombination system and an intestinal colonization factor. Proc. Natl. Acad. Sci. USA 90:3750-3754.
Rakonjac, J., J. Feng, and P. Model. 1999. Filamentous phage are released from the bacterial membrane by a two-step mechanism involving a short C-terminal fragment of pIII. J. Mol. Biol. 289:1253-1265.f#rh, 百拇医药
Rakonjac, J., and P. Model. 1998. Roles of pIII in filamentous phage assembly. J. Mol. Biol. 282:25-41.f#rh, 百拇医药
Riechmann, L., and P. Holliger. 1997. The C-terminal domain of TolA is the coreceptor for filamentous phage infection of E. coli. Cell 90:351-360.f#rh, 百拇医药
Russel, M., and P. Model. 1982. Filamentous phage pre-coat is an integral membrane protein: analysis by a new method of membrane preparation. Cell 28:177-184.f#rh, 百拇医药
Russel, M., H. Whirlow, T. P. Sun, and R. E. Webster. 1988. Low-frequency infection of F- bacteria by transducing particles of filamentous bacteriophages. J. Bacteriol. 170:5312-5316.f#rh, 百拇医药
Smilowitz, H. 1974. Bacteriophage f1 infection: fate of the parental major coat protein. J. Virol. 13:94-99.f#rh, 百拇医药
Smith, G. P. 1985. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228:1315-1317.
Stengele, I., P. Bross, X. Garces, J. Giray, and I. Rasched. 1990. Dissection of functional domains in phage fd adsorption protein. Discrimination between attachment and penetration sites. J. Mol. Biol. 212:143-149.o99, 百拇医药
Sun, T. P., and R. E. Webster. 1986. fii, a bacterial locus required for filamentous phage infection and its relation to colicin-tolerant tolA and tolB. J. Bacteriol. 165:107-115.o99, 百拇医药
Sun, T. P., and R. E. Webster. 1987. Nucleotide sequence of a gene cluster involved in entry of E colicins and single-stranded DNA of infecting filamentous bacteriophages into Escherichia coli. J. Bacteriol. 169:2667-2674.o99, 百拇医药
Taylor, R. K., V. L. Miller, D. B. Furlong, and J. J. Mekalanos. 1987. Use of phoA gene fusions to identify a pilus colonization factor coordinately regulated with cholera toxin. Proc. Natl. Acad. Sci. USA 84:2833-2837.o99, 百拇医药
Trenkner, E., F. Bonhoeffer, and A. Gierer. 1967. The fate of the protein component of bacteriophage fd during infection. Biochem. Biophys. Res. Commun. 28:932-939.
Trucksis, M., J. E. Galen, J. Michalski, A. Fasano, and J. B. Kaper. 1993. Accessory cholera enterotoxin (Ace), the third toxin of a Vibrio cholerae virulence cassette. Proc. Natl. Acad. Sci. USA 90:5267-5271.#, http://www.100md.com
Waldor, M. K., and J. J. Mekalanos. 1996. Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272:1910-1914.#, http://www.100md.com
Webster, R. E. 1991. The tol gene products and the import of macromolecules into Escherichia coli. Mol. Microbiol. 5:1005-1011.#, http://www.100md.com
Woodcock, D. M., P. J. Crowther, J. Doherty, S. Jefferson, E. DeCruz, M. Noyer-Weidner, S. S. Smith, M. Z. Michael, and M. W. Graham. 1989. Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17:3469-3478.#, http://www.100md.com
Wootton, J. C., and S. Federhen. 1996. Analysis of compositionally biased regions in sequence databases. Methods Enzymol. 266:554-571.(Andrew J. Heilpern and Matthew K. Waldor)