Inhibition of SypG-induced biofilms and host colonization by the negative regulator SypE in Vibrio fischeri

Abstract

Vibrio fischeri produces a specific biofilm to promote colonization of its eukaryotic host, the squid Euprymna scolopes. Formation of this biofilm is induced by the sensor kinase RscS, which functions upstream of the response regulator SypG to regulate transcription of the symbiosis polysaccharide (syp) locus. Biofilm formation is also controlled by SypE, a multi-domain response regulator that consists of a central regulatory receiver (REC) domain flanked by an N-terminal serine kinase domain and a C-terminal serine phosphatase domain. SypE permits biofilm formation under rscS overexpression conditions, but inhibits biofilms induced by overexpression of sypG. We previously investigated the function of SypE in controlling biofilm formation induced by RscS. Here, we examined the molecular mechanism by which SypE naturally inhibits SypG-induced biofilms. We found that SypE's N-terminal kinase domain was both required and sufficient to inhibit SypG-induced biofilms. This effect did not occur at the level of syp transcription. Instead, under sypG-overexpressing conditions, SypE inhibited biofilms by promoting the phosphorylation of another syp regulator, SypA, a putative anti-sigma factor antagonist. Inhibition by SypE of SypG-induced biofilm formation could be overcome by the expression of a non-phosphorylatable SypA mutant, indicating that SypE functions primarily if not exclusively to control SypA activity via phosphorylation. Finally, the presence of SypE was detrimental to colonization under sypG-overexpressing conditions, as cells deleted for sypE outcompeted wild-type cells for colonization when both strains overexpressed sypG. These results provide further evidence that biofilm formation is critical to symbiotic colonization, and support a model in which SypE naturally functions to restrict biofilm formation, and thus host colonization, to the appropriate environmental conditions.

Figure 1. Model of inhibition of SypG-induced biofilms by SypE.

(A) Schematic of the domain structure of SypE and select SypE mutants. SypE contains a central regulatory receiver (REC) domain flanked by a N-terminal HPK-like serine kinase domain and a C-terminal PP2C-like serine phosphatase domain. The black lines represent select SypE mutants containing the indicated protein domains. (B) When sypG is overexpressed, transcription of the syp locus is activated, resulting in the production of Syp structural proteins necessary for polysaccharide production and biofilm formation, as well as the production of regulatory proteins SypA and SypE. Our data here show that, under these conditions, SypE functions as a serine kinase and phosphorylates the downstream target protein, the putative anti-sigma factor antagonist protein SypA. Phosphorylated SypA is inactive to promote biofilm formation and host colonization. In contrast, when rscS is overexpressed, the kinase activity of SypE is inactivated (presumably through phosphorylation of SypE’s REC domain); instead, it functions as a serine phosphatase to dephosphorylate SypA, which promotes biofilm formation and colonization through an unknown mechanism. Co-overexpression of sypG and sypA (not shown) leads to biofilm formation, presumably because high levels of SypA permit some SypA to escape phosphorylation and inactivation by SypE.

Figure 2. Regulation of SypG-induced wrinkled colony formation by SypE.

The SypG expression plasmid (pCLD56) was introduced into either wild-type cells [A] or ΔsypE mutant cells [KV3299] carrying either empty vector (pVSV105) [B] or the indicated SypE-complementing plasmids: full-length SypE (pCLD48)[C], SypE^ΔNTD (pARM3)[D], SypE^CTD (pCLD67)[E], SypE^NTD (pCLD64)[F], SypE^N52A (pARM4)[G], SypE^{NTD, N52A} (pCLD65)[H]. Cultures were spotted onto agar plates and wrinkled colony morphology was assessed at 48 h post-spotting. Images are representative of at least three independent experiments. [I] Expression of FLAG epitope-tagged SypE mutant proteins was assessed by western blot analysis. Plasmids expressing FLAG-sypE ^CTD (pARM111; lane 1) or sypE ^ΔNTD (pARM162; lane 2) alleles were introduced into the ΔsypE strain containing the sypG overexpression plasmid (pCLD56). Whole-cell lysates were resolved using SDS-PAGE and the FLAG-tagged proteins were detected by western blot analysis as described in the Materials and Methods.

Figure 3. Inhibition of SypG-induced pellicle formation by SypE.

The sypG expression plasmid (pCLD56) was introduced into either wild-type cells [A] or ΔsypE mutant cells [KV3299] carrying either empty vector (pVSV105) [B] or the indicated SypE-complementation plasmids: full-length SypE (pCLD48)[C], SypE^ΔNTD (pARM3)[D], SypE^CTD (pCLD67)[E], SypE^NTD (pCLD64)[F], SypE^N52A (pARM4)[G], SypE^{NTD, N52A} (pCLD65)[H]. Strains were cultured statically in LBS medium and pellicle formation was assessed 48 h post-inoculation. A pipette tip was dragged over the surface of the air-liquid interface to visualize the pellicle. (–) denotes a weak, easily disrupted pellicle. (+) denotes a strong, detectable pellicle. Images are representative of at least three independent experiments.

Figure 4. Co-overexpression of sypG and sypA permits biofilm formation.

Biofilm formation by wild-type V. fischeri cells carrying either the sypG overexpression plasmid (pARM9) [A], the sypA overexpression plasmid (pARM13) [B], or both [C]. For A and B, the indicated vectors are pKV282 and pVSV105, respectively. The strains were cultured in LBS broth containing Tet and Cm. Cultures were spotted onto agar plates and wrinkled colony morphology was assessed at 48 h post-spotting. Images are representative of at least three independent experiments.

Figure 5. A sypA ^S56A mutant permits SypG-induced biofilm formation.

Assessment of SypG-induced wrinkled colony formation by sypG-overexpressing (pCLD56) wild-type cells [A], and sypG-overexpressing ΔsypA cells complemented with either wild-type sypA ⁺ (KV5479) [B] or the sypA ^S56A allele (KV5481) [C]. Cultures were spotted onto LBS medium containing Cm at 28°C and wrinkled colony formation was assessed at 48 h post spotting. Images are representative of at least three independent experiments.

Figure 6. Assessment of SypA phosphorylation in vivo.

Soluble lysates from the indicated V. fischeri strains were resolved by SDS-PAGE on 30 µM Phos-tag™ acrylamide gels and the proteins were detected by western blot analysis using anti-HA antibody (A) or anti-FLAG antibody (B). (A) Phos-tag™ analysis of soluble cell lysates from V. fischeri strains expressing HA-tagged sypA in single-copy. SypG-expressing (pCLD56) ΔsypA cells containing untagged sypA [KV5479] [lane 1] or HA-tagged wild-type sypA [KV6578] [lane 2]; ΔsypA ΔsypE cells expressing HA-tagged wild-type sypA ⁺ [KV6580] and carrying pCLD56[lane 3]; ΔsypA ΔsypE expressing HA-tagged wild-type sypA ⁺ [KV6580] and carrying pCLD56 and pSypE plasmid (pCLD48)[lane 4]; ΔsypA cells expressing HA-tagged sypA ^S56A [KV6579] and carrying pCLD56 [lane 5]. (+) indicates cells expressing wild-type sypA-HA and/or sypE. (S56A) indicates cells expressing sypA ^S56A. (–) indicates cells expressing untagged sypA or deleted for sypE. (B) Phos-tag™ analysis of soluble cell lysates from sypA- and sypG-overexpressing V. fischeri strains. ΔsypA cells [KV4715] carrying the SypG plasmid (pARM9) and either untagged SypA plasmid (pARM13) [lane 1] or FLAG-tagged SypA plasmid (pARM35) [lane 2]; ΔsypA ΔsypE cells [KV4716] carrying pARM9 and pARM35 [lane 3]; ΔsypA ΔsypE cells complemented with wild-type sypE [KV5649] carrying pARM9 and pARM35 [lane 4]. (–) indicates cells expressing untagged sypA or deleted for sypE. Images are representative of at least three independent experiments.

Figure 7. Impact of SypE on syp locus activation.

Transcription of the syp locus was monitored using a β-galactosidase activity assay. A transcriptional reporter construct consisting of the sypA promoter region fused upstream of a promoterless lacZ gene was inserted at the chromosomal Tn7 site of wild-type [KV3246] or ΔsypE [KV4926] cells carrying the pSypG plasmid (pCLD56) and indicated SypE expression plasmids: wild-type SypE (pCLD48), SypE^ΔNTD (pARM3), and SypE^NTD (pCLD64). Vector corresponds to pVSV105 or, in the case of wild-type carrying two vectors, pVSV105 and pKV282. Cells were grown in LBS containing Tc and Cm for 24 h. Results shown are representative of at least 3 independent experiments. Error bars indicated the standard deviation.

Figure 8. Deletion of sypE promotes host colonization.

Competitive colonization assay with sypG-overexpressing wild-type (WT) and sypE mutant strains. Newly hatched squid were exposed to a mixed inoculum of WT carrying the pSypG plasmid (pCLD56) and either ΔsypE cells [KV4390] (A) or ΔsypE cells complemented with wild-type sypE ⁺ [KV4819] (B) and carrying pCLD56. The Log RCI is plotted on the x-axis. The position of the circles on the y-axis is merely for spacing. Each circle represents a single animal. Open symbols indicate animals containing no WT cells. The black diamond and error bars indicate the average Log RCI and standard deviation for the indicated data set. Data shown are representative of at least three independent experiments.