Borrelia burgdorferi cp32 BpaB modulates expression of the prophage NucP nuclease and SsbP single-stranded DNA-binding protein

Abstract

The Borrelia burgdorferi BpaB proteins of the spirochete's ubiquitous cp32 prophages are DNA-binding proteins, required both for maintenance of the bacteriophage episomes and for transcriptional regulation of the cp32 erp operons. Through use of DNase I footprinting, we demonstrate that BpaB binds the erp operator initially at the sequence 5'-TTATA-3'. Electrophoretic mobility shift assays indicated that BpaB also binds with high affinity to sites located in the 5' noncoding regions of two additional cp32 genes. Characterization of the proteins encoded by those genes indicated that they are a single-stranded DNA-binding protein and a nuclease, which we named SsbP and NucP, respectively. Chromatin immunoprecipitation indicated that BpaB binds erp, ssbP, and nucP in live B. burgdorferi. A mutant bacterium that overexpressed BpaB produced significantly higher levels of ssbP and nucP transcript than did the wild-type parent.

Fig 1

Schematic of cp32-1 from B. burgdorferi type strain B31, a representative member of the prophage family. The boxes represent open reading frames, with arrows indicating directions of transcription. The sequences of ORFs shown in white are >90% identical between all known cp32s, while the shaded ORFs generally exhibit significant variation between replicons. The genes described in this report are identified. The large arrows indicate the locations of mapped BpaB-binding sites. A ChIP survey indicated additional in vivo BpaB-binding sites in the vicinities of blyA and the undefined ORFs PF-145, PF-146, PF-148, and PF-108.

Fig 2

Identification of a high-affinity BpaB-binding site by DNase I footprinting. The illustrated representative experiment was conducted with the addition of increasing concentrations of BpaB to samples containing 5 ng ³²P-labeled P50 erp operator noncoding dsDNA. Lanes 1 to 8 contained BpaB at final concentrations of 0, 4, 8, 10, 12, 15, 20, and 23 μM, respectively. The lower concentrations of BpaB protected the nucleotides TTATA from DNase1. Increasing protein concentrations led to expansion of protected nucleotides, consistent with EMSA data indicating the binding of additional BpaB molecules to DNA as the protein concentration increases (11, 30).

Fig 3

Alignment of the 5′ noncoding regions and the first 15 to 17 codons of the ssbP and nucP genes found on the native cp32 family members of B. burgdorferi type strain B31. Each gene's ribosome binding site (RBS) is indicated above the alignment, and initiation codons are indicated by three asterisks. Sequences found in the majority of loci are boxed and shaded. Sequences encompassed by labeled DNA probes and unlabeled competitors are indicated by colored lines below the alignment.

Fig 4

BpaB binds cp32 DNA 5′ of nucP and ssbP. (A) EMSA of BpaB with labeled DNA probes consisting of sequence 5′ of the cp32-1 nucP ORF (probe B-3.1), the cp32-1 erpAB operator, and the chromosomal flaB ORF. Lane 1 contained only DNAs. Lanes 2 to 5 also contained recombinant BpaB321 at concentrations of 0.03, 0.06, 0.09, 0.12, and 0.15 μM. Densitometric analyses indicated that BpaB binds the site 5′ of nucP with an approximately 1.4-fold-greater affinity than the site in the erpAB operator. (B) EMSA of BpaB, labeled nucP 5′ noncoding DNA (probe B-3.1), and unlabeled competitor DNAs. Lane 1 contained probe B-3.1 only, while the other lanes also included 0.1 μM BpaB321. Lanes 3 to 6, 7 to 10, and 11 to 14 additionally contained unlabeled probes at molar concentrations that were 63-, 125-, 250-, and 500-fold greater than that of the labeled probe. (C) EMSA indicating binding of BpaB to DNA 5′ of ssbP. Lane 1 contained only labeled ssbP DNA (probe B-4), while lanes 2 to 7 contained BpaB321 at concentrations of 0.3, 0.6, 0.9, 1.2, 1.4, and 1.5 μM.

Fig 5

Cellular levels of BpaB affect expression of ssbP and nucP. B. burgdorferi strains KS52 (which contains bpaB on a plasmid under the control of the ATc-inducible ostp promoter) and KS50 (which carries the parental, empty ostp vector) were incubated with ATc, and total RNAs were purified. Q-RT-PCR was used to determine the levels of ssbP, nucP, and erpA relative to the constitutively expressed housekeeping flaB gene. The illustrated results are representative of two independent analyses, each performed in triplicate using separately prepared batches of cDNA. Standard errors are indicated. Differences in expression levels for each gene with and without enhanced levels of BpaB were statistically significant (P < 0.05) by two-tailed Student's t tests.

Fig 6

Enhanced production of BpaB56 does not affect the relative copy numbers of cp32-1, cp32-3, and cp32-4. Oligonucleotide primer pairs specific for the resident cp32s of sibling strains KS50 and KS52 were utilized for limited PCR, followed by agarose gel electrophoresis and staining with ethidium bromide. Data for 15 PCR cycles are shown. Similar results were obtained after 18 or 21 cycles.

Fig 7

SsbP is a nonspecific ssDNA-binding protein. (A) SsbP binds ssDNA, but not dsDNA. Lanes 1 to 6 contained the labeled ssDNA probe ssBio104. Lanes 7 to 12 contained the labeled dsDNA probe dsBio104. Lanes 1 and 7 did not contain any added protein. SsbP was added to the DNAs at the following final concentrations: lanes 2 and 8, 0.083 ng/ml; lanes 3 and 9, 0.17 ng/ml; lanes 4 and 10, 0.25 ng/ml; lanes 5 and 11, 8.3 ng/ml; and lanes 6 and 12, 2.1 ng/ml. (B) SsbP binds nonspecifically. Lanes 1 to 7 contained the labeled ssDNA probe ssBio104. Lane 1 contained labeled probe only. Lanes 2 to 4 included SsbP added to final concentrations of 0.083 ng/ml, 0.83 ng/ml, or 1.7 ng/ml, respectively. Lanes 5 to 7 contained probe, 1.7 ng/ml SsbP, and 100× excesses of unlabeled ssDNA probes A42-R, A15-R, and A21-R, respectively.

Fig 8

NucP is a DNase. E. coli that produces polyhistidine-tagged NucP from an inducible lac promoter or that carries the expression vector with nucP inserted in the inverse orientation (E. coli control) was subjected to identical treatments for protein purification using immobilized nickel beads. A linearized DNA was incubated with purified NucP for 30 min or 4 or 8 h or with E. coli control extract for 24 h and then subjected to agarose gel electrophoresis and ethidium bromide staining. Rapid DNA degradation was repeatedly observed with NucP. E. coli control extracts possessed appreciably less nuclease activity, presumably due to residual native E. coli DNase(s).