ABSTRACT
The nucleotide sequence of pNAD1, a plasmid from Haemophilus ducreyi identified on the basis of its ability to confer NAD independence on Actinobacillus pleuropneumoniae and H. influenzae, has been determined. In addition to containing the nadV gene, the plasmid contains homologues of the rstR and rstA genes, genes encoding repressor and replication proteins, respectively, in the Vibrio CTXφ and the Vibrio RS1 element, suggesting a single-stranded bacteriophage origin for pNAD1. Tandem copies of the plasmid are integrated into the H. ducreyi 35000HP genome.
Haemophilus ducreyi is the causative agent of chancroid, a sexually transmitted ulcerative disease. H. ducreyi, a member of the Pasteurellaceae family, requires exogenous heme for growth but does not require NAD (V factor). V-factor-independent strains among the Pasteurellaceae can be differentiated from V-factor-dependent strains by the ability to utilize nicotinamide (NAm) as a precursor for NAD biosynthesis (12). H. haemoglobinophilus, which is V factor independent, synthesizes the enzyme nicotinamide phosphoribosyltransferase, which converts NAm to nicotinamide mononucleotide and allows the use of NAm as a source of pyridine nucleotide (7).
For many species of Pasteurellaceae defined as V factor dependent, V-factor-independent variants have been identified. These include strains of Actinobacillus pleuropneumoniae, which causes pleuropneumonia in swine (17); H. paragallinarum, which causes fowl coryza (1, 10); and H. parainfluenzae, which can cause pneumonia and meningitis in humans (4). In H. parainfluenzae, H. paragallinarum, and certain H. ducreyi strains, the gene encoding V-factor independence has been shown to be present on a plasmid (1, 19, 20).
Martin et al. identified a plasmid (designated pNAD1) that conferred NAD independence on H. influenzae strain KW20 and A. pleuropneumoniae (8). One gene (designated nadV) carried on this plasmid was shown to encode nicotinamide phosphoribosyltransferase. In the present study, we determined the sequence of the remainder of pNAD1 and determined that it encodes homologues of bacteriophage genes responsible for the replication of the Vibrio CTX phage and the RS1 element. Further, we demonstrated that the plasmid is integrated in tandem copies in the H. ducreyi 35000HP genome (GenBank accession no. NC_002940 ).
Bacterial strains and culture conditions.
H. ducreyi strains 35000HP and ATCC 27722 were grown at 35°C with 5% CO2 on chocolate agar (Becton Dickinson). Chocolate agar plates supplemented with kanamycin at 20 μg/ml and chocolate plates lacking NAD were prepared as previously described (14). Escherichia coli strains were grown on Luria-Bertani (LB) plates or in LB broth supplemented with appropriate antibiotics. Where appropriate, kanamycin was used at 10 or 20 μg/ml and ampicillin was used at 50 μg/ml.
Recombinant DNA techniques.
All restriction enzymes were purchased from New England Biolabs. Calf intestine alkaline phosphatase and T4 DNA ligase were purchased from Gibco BRL. Plasmid isolations were performed using Qiagen purification kits (Qiagen, Valencia, Calif.). Electroporation of E. coli and H. ducreyi was performed as described previously (13). Standard recombinant DNA methods were performed as described previously (13). Southern blotting was performed by the ECL method according to the instructions of the manufacturer (Amersham Pharmacia Biotech, Piscataway, N.J.).
Plasmids were prepared (using a Qiagen Miniprep kit) from H. ducreyi ATCC 27722 and 35000HP. No plasmid DNA was observed on ethidium bromide-stained agarose gels or SYBR-Gold stained agarose gels, but plasmid DNA was detected in plasmid preparations by Southern hybridization (data not shown). Although the plasmid was present at very low concentrations, NAD-independent clones of A. pleuropneumoniae or H. influenzae strain Rd were readily identified after transformation with the Miniprep DNA prepared from H. ducreyi ATCC 27722. A plasmid containing the nadV gene was isolated (using a Qiagen Miniprep procedure) from these NAD-independent transformants. The plasmid isolated from an H. influenzae strain Rd transformant was sequenced using an ABI 3100 DNA automated sequencer and dye terminator chemistries. Sequencing primers were generated on the basis of the previously determined sequence of the nadV gene (GenBank accession no. AF273842 ), and additional primers were generated as necessary to sequence both strands of the complete plasmid. Contig assembly and sequence analysis were performed with DNASTAR software (Madison, Wis.) and with GCG software (Genetics Computer Group, Madison, Wis.). Blast homologies were identified using the National Center for Biotechnology Information server (http://www.ncbi.nlm.nih.gov/BLAST/ ).
Characterization of pNAD1.
Plasmid pNAD1 is a circular 5,568-bp plasmid (Fig. 1). In addition to containing the previously reported nadV gene, the sequence contains several potential coding regions. One open reading frame (ORF) (designated plpR [phage-like protein]) encodes a putative protein that is 55% identical to the product of the rstR gene of the cholera toxin phage (CTXφ) from Vibrio cholerae O139 Calcutta, a repressor that regulates the transcription of the rstA gene (Fig. 2A) (3, 11). Transcribed in the opposite direction is a putative gene (designated plpA) whose product is 51% identical to ORF320 of bacteriophage If1 and 36% identical to the rstA gene product, a protein required for replication of CTXφ DNA (Fig. 2B) (18). Downstream of plpA is a small ORF designated plpD that is 31% identical to the product of gene V of enterobacterial phage I2-2, a putative single-stranded DNA binding protein (accession no. NP_039617 ). A fourth ORF (designated ORF167) has homology to several hypothetical proteins. Several other small ORFs have been noted but are of unknown significance, as they have no homologues in the GenBank database (Fig. 1). The Vibrio RS1 element has a third gene (designated rstB) (18). The product of the rstB gene is thought to play a role in integration of the RS1 element. No homologue of this gene is present in pNAD1. The sequence of pNAD1 has been deposited in GenBank (accession no. AY434675 ).
Circular map of pNAD1. The nadV gene encodes nicotinamide phosphoribosyl transferase as described by Martin and coworkers (GenBank accession no. AF273842 ) (8). The plpR and plpA genes, as well as the gene encoding a single-stranded DNA binding protein (designated plpD), are shown. Other ORFs are designated by the number of residues in the putative protein products. Plasmid pNAD1 has been assigned GenBank accession no. AY434675 .
(A) ClustalW alignment of the derived amino acid sequences of the plpR gene and the rstR gene from V. cholerae O139 Calcutta (accession no. AAF07851 and AAF43270 ). (B) ClustalW alignment of the derived amino acid sequences of the plpA gene, the rstA1 gene from V. cholerae strain SCE264 (accession no. AAK07627 ) (11), and ORF320 from bacteriophage If1 (accession no. NP_047361 ).
The pNAD1 element replicates as a plasmid in H. influenzae Rd. We were interested in determining whether it would replicate in E. coli. The ΩKan-2 interposon (16) was cloned into a unique BglII site in the nadV gene, and the ligation mixture was transformed into H. influenzae. Clones were selected on chocolate agar containing kanamycin, and a plasmid with the correct restriction map was saved as pNAD1::ΩKan-2. Plasmid pNAD1::ΩKan-2 was transformed into E. coli DH5α, and clones were selected on LB agar containing kanamycin at 10 μg/ml. Transformants were not obtained, indicating that pNAD1::ΩKan-2 could not become established in E. coli (data not shown).
H. ducreyi strain 35000HP has tandem copies of pNAD1 integrated into the chromosome.
Strain 35000HP is the prototype strain used by most investigators. This strain is thought to be plasmid free. The sequence of the genome of this strain was recently completed; again, there was no indication that a plasmid was present. When a Qiagen Miniprep was prepared and used to transform H. influenzae, however, a plasmid conferring NAD independence on H. influenzae was readily identified. This plasmid had the same restriction map as pNAD1. When the genome sequence was probed with the nadV gene, it was apparent that there were tandem integrated copies of pNAD1 in the H. ducreyi genome. The genomic region from strain 35000HP is shown in Fig. 3. The ompP2A and ompP2B genes encode porin-like proteins with homology to the P2 protein of H. influenzae. Downstream of the ompP2B gene are a number of putative genes, several of which have homology to putative phage proteins. Hd1436 is most closely related to a putative phage-related secreted protein from Yersinia pestis CO92 (YPO2280) (accession no. NP_405817 ; 48% identical over a 356-amino-acid [aa] overlap) (15). Hd1437 is 40% identical to a putative Y. pestis CO92 phage-related membrane protein (YPO2278) (accession no. NP_405815 ) (15). The C-terminal 17 aa of Hd1439 are 76% identical to YPO2277 (accession no. NP_405814 ) (15), but overall, the putative proteins are not closely related. The 5′ portion of the Hd1439 gene (Fig. 3) is located within the pNAD1 plasmid. The second copy of the plasmid contains only the 5′ portion of Hd1439 (Fig. 3) (annotated as ORF80b in Fig. 1). Downstream of the nadV2 gene is Hd1456, the ackA gene that encodes acetate kinase. The first 51 bases of the acetate kinase gene are located in the pNAD1 plasmid (ORF17) (Fig. 1).
Map of the chromosomal region of the H. ducreyi 35000HP genome containing pNAD1. The pNAD1 genes are color coded as shown in Fig. 1 for easy reference.
A gene with homology to the nadV gene is present in a number of organisms (8). The ORFs with the highest homology are from Shewanella oneidensis MR-1 (5), Cytophaga hutchinsonii (accession no. ZP_00118436 ), and Mycoplasma pneumoniae M129 (2, 6). Among the Pasteurellaceae, the gene is absent from H. influenzae but present in Pasteurella multocida and A. actinomycetemcomitans (9). In the S. oneidensis, P. multocida, and M. pneumoniae genomes, the gene order does not resemble that seen in H. ducreyi.
To experimentally demonstrate that pNAD1 was integrated into the H. ducreyi 35000HP genome, Southern hybridization analysis was performed using an enhanced chemiluminescence methodology. Genomic DNA from strain 35000HP was digested with BglII (cuts once in the pNAD1 plasmid) and BstEII, an enzyme that does not cut pNAD1. The positions of the restriction enzyme sites are shown in Fig. 3. The Southern blot was probed with pNAD1 (Fig. 4). BglII fragments of the anticipated sizes (12.8, 5.6, and 2.4 kb) were observed. Similarly, the expected 16.8-kb genomic BstEII fragment was observed, indicating that there are two tandem copies of pNAD1 integrated into the H. ducreyi chromosome.
Chromosomal DNA was prepared from H. ducreyi strain 35000HP and digested with BglII or BstEII, and fragments were separated on an agarose gel. After transfer, the blot was hybridized to a PCR product containing the majority of pNAD1 that had been labeled by an enhanced chemiluminescence procedure. The size of the restriction fragments confirms the chromosomal localization of pNAD1. The size of the BstEII genomic fragment is consistent with two tandem copies of pNAD1.
Miller and coworkers recently identified a T4-like bacteriophage (designated KVP40) that infects V. cholerae and other Vibrio species (E. S. Miller, Abstr. 103rd Gen. Meet. Am. Soc. Microbiol., abstr. M-020, 2003, and M. Pineda, B. Szczypinski, E. S. Miller, and D. M. Hinton, Abstr. 103rd Gen. Meet. Am. Soc. Microbiol., abstr. M-008, 2003). The genome of this bacteriophage encodes a complete NAD scavenging system including the nadV gene. Plasmid pNAD1 appears (by sequence homology) to have evolved from a single-stranded DNA CTXφ-like bacteriophage that acquired the nadV gene, possibly from a KVP40-like phage, and then underwent a duplication event. Phage-like sequences are recognizable adjacent to one copy of pNAD1 in the strain 35000HP genome, but only the 5′ portion of the Hd1439 gene is present in pNAD1. The existence of plasmid-mediated NAD independence in other members of the Pasteurellaceae suggests that pNAD1, or a related plasmid, has been horizontally transferred among members of this group of organisms.
ACKNOWLEDGMENTS
This work was supported by the National Institutes of Health (grants R01 AI45091 and R01 HD-96-003).
FOOTNOTES
- Received 30 June 2003.
- Returned for modification 30 August 2003.
- Accepted 11 November 2003.
- Copyright © 2004 American Society for Microbiology