Sequence analysis of TnphoA insertion sites in Vibrio cholerae mutants defective in rugose polysaccharide production

Abstract

Vibrio cholerae can switch from a smooth to a wrinkled or rugose colony phenotype characterized by the secretion of a polysaccharide that enables the bacteria to survive harsh environmental conditions. In order to understand the genetic basis of rugosity, we isolated TnphoA-induced stable, smooth mutants of two O1 El Tor rugose strains and mapped the insertion sites in several of the mutants using a modified Y-adapter PCR technique. One of the TnphoA insertions was mapped to the first gene of the vps region that was previously shown to encode the rugose polysaccharide biosynthesis cluster. Three insertions were mapped to a previously unknown hlyA-like gene, also in the vps region. Five other insertions were found in loci unlinked to the vps region: (i) in the epsD gene (encodes the "secretin" of the extracellular protein secretion apparatus), (ii) in a hydG-like gene (encodes a sigma(54)-dependent transcriptional activator similar to HydG involved in labile hydrogenase production in Escherichia coli, (iii) in a gene encoding malic acid transport protein upstream of a gene similar to yeiE of E. coli (encodes a protein with similarities to LysR-type transcriptional activators), (iv) in dxr (encodes 1-deoxy-D-xylulose 5-phosphate reductoisomerase), and (v) in the intergenic region of lpd and odp (encode enzymes involved in the pyruvate dehydrogenase complex formation). These data suggest the involvement of a complex regulatory network in rugose polysaccharide production and highlight the general utility of the Y-adapter PCR technique described here for rapid mapping of transposon insertion sites.

FIG. 1

Schematic of the Y-adapter PCR. In step 1, chromosomal DNAs of the TnphoA insertion containing strains are digested with restriction enzyme Sau3A1. In step 2, a Y-adapter linker (the Y region is noncomplementary) is prepared from oligonucleotides A1 and A2. In step 3, the Y-adapter is ligated to the restriction fragments. In step 4, the adapter-ligated DNAs are used as templates for PCRs with a Y-adapter (indicated as the 5′ PCR primer [P]) and a transposon-specific primer. Since the 5′ primer sequence is from the same strand as the Y region, it requires a first round of PCR with the Tn primer (step 4a). The resulting single-stranded DNA serves as the template for annealing of the “P primer” in the second round of PCR (step 4b). The resulting PCR products from subsequent cycles are cloned and sequenced to identify the transposon junction.

FIG. 2

Agarose gel electrophoresis of the Y-adapter PCR products. Size marker, 1-kb ladder. In four PCRs, the two PCR products resulting from amplification of the left and right TnphoA junctions can be seen. In three reactions, only one product is seen (S1, S13, and S20). The asterisk indicates the position of the primers.

FIG. 3

Genetic organization of the VPS region. The DNA sequence of the V. cholerae VPS region was obtained from the TIGR microbial genome database. The entire region in the map is 33,024 bp, and the vps region is 26,458 bp. Sequence analysis was done by the NCI BLAST server program, and the recently published annotation of the vps region (8) has been used to define the products of the ORFs. acrD (VC0914) is at the left junction of the vps cluster, and yitA (VC0940) and glyA (VC0941) are at the right junction. There is a 426-bp intergenic region containing the vps promoter. There are 22 ORFs arranged in three clusters: ORFs VC0916 to VC0927 and ORFs VC0934 to VC0937, separated by the cluster of ORFs from VC0927 to VC0933. The middle cluster genes apparently encode functions unrelated to polysaccharide biosynthesis. The arrowheads indicate the positions of the TnphoA insertions in the vps region. The three hly::TnphoA insertions were in close proximity to each other.

FIG. 4

Colony morphology of wild-type smooth and rugose variants of the N16961 strain. The left panel shows the streak plates, and the right panels show the magnified individual colonies on the respective plates. (Top panel) N16961 R, wild-type rugose; N16961 S, wild-type smooth. (Middle panel) NS25 and AA54, hydG::TnphoA and hydG::Kan^r mutants, respectively; AA10, epsD::Kan^r. Only AA54 and AA10 colonies are shown in the middle right panel. The NS25 colonies were similar to the AA54 colonies. (Bottom panel) AA10 complemented by the epsD⁺ plasmid pDSK-2 and NS25 and AA54 complemented by hydG⁺ plasmid. pAA435 is a pWSK29 derivative containing the hydG⁺ gene (SacII-BamHI) from pAA2311 (see the text).