Insights on the regulation and function of the CRISPR/Cas transposition system located in the pathogenicity island VpaI-7 from Vibrio parahaemolyticus RIMD2210633

Abstract

CRISPR/Cas-mediated transposition is a recently recognized strategy for horizontal gene transfer in a variety of bacterial species. However, our understanding of the factors that control their function in their natural hosts is still limited. In this work, we report our initial genetic characterization of the elements associated with the CRISPR/Cas-transposition machinery (CASTm) from Vibrio parahaemolyticus (VpaCASTm), which are encoded within the pathogenicity island VpaI-7. Our results revealed that the components of the VpaCASTm and their associated CRISPR arrays (VpaCAST system) are transcriptionally active in their native genetic context. Furthermore, we were able to detect the presence of polycistrons and several internal promoters within the loci that compose the VpaCAST system. Our results also suggest that the activity of the promoter of the atypical CRISPR array is not repressed by the baseline activity of its known regulator VPA1391 in V. parahaemolyticus. In addition, we found that the activity of the promoter of tniQ was modulated by a regulatory cascade involving ToxR, LeuO, and H-NS. Since it was previously reported that the activity of the VpaCAST system was less efficient than that of the VchCAST system at promoting transposition of a miniaturized CRISPR-associated transposon (mini-CAST) in Escherichia coli, we analyzed if the transposition efficiency mediated by the VpaCAST system could be enhanced inside its natural host V. parahaemolyticus. We provide evidence that this might be the case, suggesting that there could be host induction factors in V. parahaemolyticus that could enable more efficient transposition of CASTs.IMPORTANCEMobile genetic elements such as transposons play important roles in the evolutionary trajectories of bacterial genomes. The success of transposon dissemination depends on their ability to carry selectable markers that improve the fitness of the host cell or loci with addictive traits such as the toxin-antitoxin systems. Here we aimed to characterize a transposon from Vibrio parahaemolyticus that carries and could disseminate multiple virulence factors. This transposon belongs to a recently discovered family of transposons whose transposition is guided by crRNA. We showed that the transposition machinery of this transposon is transcribed in V. parahaemolyticus and that there are likely host-associated factors that favor transposition in the natural host V. parahaemolyticus over transposition in Escherichia coli.

Keywords: CRISPR/Cas; Vibrio parahaemolyticus; pathogenicity islands; transcriptional regulation; transposons.

Fig 1

The pathogenicity island VpaI-7 from V. parahaemolyticus RIMD2210633 is within a CRISPR/Cas-transposon. (A) Diagram depicting genes encoding relevant components of the VpaI-7 island including virulence factors and the elements of a Type IF3 CRISPR/Cas transposition system. (B) Sequence alignment of the predicted TnsB-binding sites (TnsB-box) in the L- and R-terminal regions of the VchCAST and VpaCAST systems. The TnsB-boxes are delimited by rectangles. (C) Sequence alignment of the direct repeats flanking the spacer sequences of the typical and atypical CRISPR arrays from the VchCAST and VpaCAST systems.

Fig 2

The genetic clusters tniQ-cas876 and tnsABC are transcribed as polycistrons in V. parahaemolyticus RIMD2210633. (A) Diagram illustrating the arrangement of the tniQcas876 and tnsABC genetic clusters. The Tn7-like transposition-associated genes are colored in green, and the CRISPR/Cas-associated genes are colored in blue. The lines shown above the genetic clusters indicate the location of the regions amplified by PCR, and the size of the amplification product is also indicated. (B, C) Image of the electrophoretic migration in an agarose gel of PCR products obtained with primer pairs specific to (B) monocistrons or (C) polycistrons. (D) Image of a representative RT-PCR analysis of the transcript abundance of rpoB, tniQ, cas8, cas7, and cas6 in the WT strain (W) or the P_tniQ::FRT-aacC1 mutant strain (F). Total RNA was used as a template in negative controls (CT^-). For panel (D), a primer pair used to amplify a region of rpoB was used to assess the contamination of total RNA with gDNA. Complementary DNA (cDNA) obtained by RT-PCR was used as a template to detect the transcripts of interest. Genomic DNA from V. parahaemolyticus RIMD2210633 was used as a template in positive controls (CT⁺). (E) Bar graph representing the mean and standard deviation of the relative abundance (P_tniQ::FRTaacC1/WT) of transcripts for rpoB, tniQ, cas8, cas7, and cas6 obtained through a semiquantitative RT-PCR analysis. Means were compared using an ordinary one-way ANOVA test followed by a Dunnett’s test to compare each mean to the mean of the control rpoB. Mean differences with a P value ≤ 0.05 were interpreted as significant. * Adjusted P value ≤ 0.05, *** ≤0.001.

Fig 3

Multiple independent promoters drive transcription of genes associated with the VpaCAST system. (A) Diagram illustrating the genetic arrangement of the components of the VpaCAST system, and the regions used to generate each transcriptional fusion with the luxCDABE gene cluster. (B) Box plots represent the values of light production expressed as relative luminescence units (RLU) obtained from six independent biological replicates of strains harboring the transcriptional fusions of interest. The label P_empty refers to the negative control of a strain harboring the promoterless plasmid pBBRlux. Means were compared using a Brown-Forsythe and Welch ANOVA test followed by a Dunnett’s T3 test to compare each mean to the mean of the negative control (P_empty). Mean differences with a P value ≤ 0.05 were interpreted as significant. ** Adjusted P value ≤ 0.01, *** ≤0.001.

Fig 4

The VpxR (VPA1391) regulator positively regulates the PvpxR promoter and negatively regulates the P_VPA1392 and P_crAtyp promoters. (A) Logo (Forward and reverse complement orientation) generated with frequency matrices of a conserved motif discovered by the Multiple Em for Motif Elicitation software. (B) Diagram illustrating the arrangement of the CRISPR arrays and the vpxR and VPA1392 genes. The sequence of the VpxR-box and putative −10 boxes located upstream of the atypical CRISPR array and in the intergenic region between vpxR and VPA1392 is also shown. The putative VpxR-binding site is indicated with a rectangular box. (C) Box plots represent the values of light production expressed as relative luminescence units (RLU) obtained from 12 independent biological replicates from three independent experiments of strains harboring the transcriptional fusions of interest and either an empty overexpression plasmid (pEMPTY) or a plasmid that overexpresses vpxR. Means were compared using a Brown-Forsythe and Welch ANOVA test followed by a Dunnett’s T3 test for multiple comparisons. Mean differences with a P value ≤ 0.05 were interpreted as significant. **** P value ≤ 0.0001, ** P value ≤ 0.01.

Fig 5

ToxR, LeuO, and H-NS modulate the expression of the P_tniQ and P_VPA1392 promoters. (A) Illustration depicting the sequence of the ToxR-box consensus and the putative ToxR-boxes located upstream of P_leuO and P_tniQ promoters. (B) Box plots represent the values of light production expressed as relative luminescence units (RLU) obtained from 12 independent biological replicates from three independent experiments of strains harboring the transcriptional fusions of interest. Means were compared using a Brown-Forsythe and Welch ANOVA test followed by a Dunnett’s T3 test for multiple comparisons. Mean differences with a P value ≤ 0.05 were interpreted as significant. **** P value ≤ 0.0001, *** P value ≤ 0.001, ** P value ≤ 0.01, * P value ≤ 0.05.

Fig 6

The CRISPR/Cas-transposition system from V. parahaemolyticus promotes transposition in E. coli and V. parahaemolyticus. (A) Diagram illustrating the screening assay used to analyze transposition events in the lacZ_Ec allele from E. coli GenR and VpaΔCAST_lacZ strains harboring either the pVchINT_lacZ or pVpaINT_lacZ plasmid. (B) Box plots of the proportion of white colonies over blue colonies from E. coli GenR and VpaΔCAST_lacZ strains harboring either the pVchINT_lacZ or pVpaINT_lacZ plasmid grown in selection plates with X-gal and IPTG. Experiments were done at least six times. The differences between means were analyzed with an unpaired t-test. Mean differences with a P value ≤ 0.05 were interpreted as significant. **** indicate a P value ≤ 0.0001. ** indicate a P value ≤ 0.01. (C) Diagram of the regions amplified by PCR to identify transposition events and intact lacZ_Ec copies. The solid lines represent the amplified regions and the region corresponding to the mini-CAST. (D) Image of the electrophoretic migration in an agarose gel of PCR products stained with ethidium bromide (EtBr) and obtained using as template the lysate of a blue colony from a V. parahaemolyticus strain harboring pVpaINT (CT^-), or blue and white colonies from a V. parahaemolyticus strain harboring pVpaINT_lacZ. (E, F) Image of the electrophoretic migration in an agarose gel of PCR products stained with EtBr and obtained using as template the lysate of 10 randomly selected colonies of (E) E. coli or (F) V. parahaemolyticus strains harboring pVpaINT_lacZ. All gels were loaded in the first lane with the Quick-Load Purple 1 kb Plus DNA Ladder from New England Biolabs. The relevant size range of the ladder is indicated on the gel from panel D.

Fig 7

The promoters that drive transcription of genes associated with the VpaCAST system are more active in V. parahaemolyticus than in E. coli. Box plots represent the values of light production expressed as relative luminescence units (RLU) obtained from 12 independent biological replicates of V. parahaemolyticus and E. coli strains harboring the transcriptional fusions of interest. The data were obtained from three independent experiments. Pairwise comparisons of means were done using an unpaired t-test. Mean differences with a P value ≤ 0.05 were interpreted as significant. **** P value ≤ 0.0001.