LIU Lei, FANG Nan, LIU Wan Bing, HE Ying Yu, LIU Yan, and ZHANG Yi Quan

Type-VI secretion system (T6SS) is a widespread bacteriophage-like complex in bacteria that participates in multiple physiological processes,including metal ion uptake, bacterial competition,and biofilm formation[1].Yersinia pestisis the causative agent of plague. There are five T6SS gene clusters inY. pestisCO92, among which only theYPO0499-YPO0516locus was well investigated[1,2].The expression ofYPO0499-YPO0516is induced at 26 °C, suggesting that it plays a possible role in the natural cycle ofY. pestis[2]. Deletion ofYPO0499-YPO0516alters the uptake and intracellular growth ofY. pestisin macrophages, which is dependent on the pre-grown temperature[2]. However, it has no effect on the virulence in murine and on the ability of the pathogen to infect fleas[2]. The Hcp-like protein encoded byYPO0502was shown to be an autoagglutination factor involved in the bacterial interaction[3].

The transcriptional regulator RovA is required for the full virulence ofY. pestisthrough its regulatory activities on the virulence genes such aspsaEFABCandCUS-2loci[4]. In addition, RovA negatively regulates the biofilm formation byY. pestisviarepression ofhmsTtranscription and c-di-GMP production[5]. Moreover, a study showed that RovA positively regulates the transcription ofYPO0499-YPO0516genes but lacks detailed regulatory mechanisms[6]. TheYP3680-YP3663locus inY. pestis Microtusstrain 91001 is highly homologous to theYPO0499-YPO0516locus inY. pestisCO92(Supplementary Figure S1 available in www.besjournal.com). Therefore, in the present study,Y.pestis91001 (wild type, WT) and its non-polarrovAmutant (designatedΔrovA[5]) were employed to investigate the regulatory mechanisms of RovA onYP3680transcription.

The overnight cell cultures of WT andΔrovAwere diluted 20-fold into 18 mL of fresh LB broth [1%tryptone (Oxoid), 0.5% yeast extract (Oxoid), and 1%NaCl (Merck Millipore)], and then cultured at 26 °C by shaking at 230 rpm to reach an optical density at 620 nm (OD620) value of approximately 1.0. The bacterial cells were harvested, and total RNAs were extracted using TRIzol Reagent (Invitrogen, USA). The qPCR assay was applied to detect the mRNA levels ofYP3680in WT andΔrovAusing a Light-Cycler system(Roche, Switzerland) together with the SYBR Green master mix (the primers are listed in Supplementary Table S1 available in www.besjournal.com). As shown in Figure 1A, the mRNA level ofYP3680significantly decreased inΔrovAcompared to that in WT, indicating that transcription ofYP3680was activated by RovA. To detect whether RovA has regulatory action on theYP3680promoter activity,the promoter-proximal DNA region ofYP3680was cloned into the pRW50 plasmid harboring a promoterlesslacZreporter gene. The recombinant pRW50 was then transferred into WT andΔrovAto determine the β-galactosidase activities in each strain using a β-Galactosidase Enzyme Assay System(Promega, USA) according to the manufacturer’s instructions. As shown in Figure 1B, the promoter activity ofYP3680significantly decreased inΔrovArelative to that in WT, suggesting that RovA acted as a positive regulator ofYP3680expression. In addition, a RovA box-like sequence,TCATCGTGCTAA,was identified in the promoter-proximal DNA region ofYP3680(Supplementary Figure S2 available in www.besjournal.com)[7], indicating thatYP3680transcription was probably under the direct control of RovA inY. pesitis. Thus, the 487 bp upstream ofYP3680was amplified and then labeled with [γ-32P]-ATP at the 5′- ends used for the electrophoresis mobility shift assay (EMSA) DNA probe[5]. EMSA was performed as previously described[5], and the results showed that His-RovA was able to dose-dependently bind to the regulatory DNA region ofYP3680, but it could not bind to the DNA fragment of 16S rDNA as the negative control (Figure 1C). Taken together,these results suggested that transcription ofYP3680was directly activated by RovA inY. pestisbiovar Microtus.

Figure 1. Positive regulation of YP3680 by RovA. The minus and positive numbers in the brackets indicated nucleotide positions upstream and downstream of YP3680. (A) qPCR. The relative mRNA level of YP3680 was compared between ΔrovA and WT. (B) LacZ fusion. The YP3680: lacZ fusion vector was transferred into WT and ΔrovA to determine the β-galactosidase activity (miller units) in the cellular extracts. (C) EMSA. The entire promoter-proximal region of YP3680 was incubated with increasing amounts of purified His-RovA, and then subjected to 4% (w/v) polyacrylamide gel electrophoresis. Shown below the binding was the schematic representation of the EMSA design.

Figure 2. Negative regulation of YP3680 by RcsAB. The minus and positive numbers in the brackets indicated nucleotide positions upstream and downstream of YP3680. (A) Primer extension. Lanes C, T, A,and G represented Sanger sequencing reactions. An oligonucleotide primer was designed to be complementary to the RNA transcript of YP3680. The primer extension product was analyzed with an 8 M urea -6% acrylamide sequencing gel. The transcription start site was indicated by the arrow with nucleotide and position. The LacZ fusion (B) and EMSA (C) were performed as given in Figure 1. *Asterisks indicate statistically significant differences (P ＜ 0.01).

The Rcs phosphorelay is a complex twocomponent system generally composed of three proteins, RcsB, RcsC, and RcsD[8]. The membrane protein RcsC is a sensor kinase catalyzing the phosphorylation of RcsD, which then transfers the phosphate group to RcsB[8]. Phosphorylated RcsB acts either alone or in combination with the RcsA to regulate the gene expression[8]. The gene encoding RcsA is nonfunctional inY. pestis[9]. However,replacement of thercsApseudogene with the functionalrcsAallele fromY. pseudotuberculosisto yield the RcsAB complex may strongly inhibit theY.pestisbiofilm formation, which is accomplished by directly repressing the transcription ofhmsCDE,hmsT,andhmsHFRS[9]. A recent study showed that expression of T6SS in pathogenicEscherichia coliwas strongly induced by the RcsB[10]. In this work, we detected whether the RcsAB complex has a regulatory activity on the transcription ofYP3680inY. pestis.

The mid-log phase (OD600≈1.0) bacterial cells of WT,ΔrcsB(rcsBmutant), c-rcsA(the WT strain transformed with the recombinant pACYC184 expressing RcsA ofY. pseudotuberculosis), andΔrcsB/c-rcsA(theΔrcsBstrain transformed with the recombinant pACYC184 plasmid)[9]were harvested,and then total RNAs were extracted. The primer extension assay, which was performed as described in the previous studies[5,9], was employed to map the transcription start sites ofYP3680and to investigate the relationship between RcsB regulation andYP3680transcription. As shown in Figure 2A, the assay detected a single transcription start site forYP3680located at 189 bp upstream of the translation start site, but both the -10 and -35 elements are a bad match with the consensus prokaryotic sequence (Supplementary Figure S2),suggesting that the promoter is relatively weak. The results of primer extension also showed that the mRNA levels ofYP3680significantly increased in theΔrcsBandΔrcsB/c-rscAthan those in WT and c-rcsA,whereas no significant differences were observed between theΔrcsBandΔrcsB/c-rscA, but it seemed to be enhanced in WT relative to the c-rcsA(Figure 2A). In addition, the results of LacZ fusion assay showed that the promoter activities ofYP3680were much higher inΔrcsBandΔrcsB/c-rscAthan those in WT and c-rcsA, whereas no significant differences were observed betweenΔrcsBandΔrcsB/c-rscAor WT and c-rcsA(Figure 2B). A RcsAB box-like sequence,TAAGTTTATCCCTA, was identified in the promoter-proximal DNA region ofYP3680[9],suggesting that transcription ofYP3680may be directly regulated by RcsAB. Indeed, the EMSA results showed that His-RcsB-P alone or mixed with the excess MBP-RcsA was able to specifically bind to the regulatory DNA region ofYP3680in a dosedependent mannerin-vitro(Figure 2C). The addition of excess MBP-RcsA improved the DNA-binding activity of His-RcsB-P (Figure 2C). Taken together,these results suggested that RcsB alone or upon binding with RcsA allele was able to specifically bind to the regulatory DNA region ofYP3680to repress its transcription.

Supplementary Figure S1. Organization of the T6SS gene cluster in Y. pestis. Genes connected by the lines indicate alleles in the two Yersinia strains.

Supplementary Figure S2. Promoter structure of YP3680. Shown were translation/transcription starts,predicted core promoter -10 and -35 elements, predicted Shine-Dalgarno (SD) sequences for ribosomal binding, and RovA and RcsAB box-like sequences.

Supplementary Table S1. Oligonucleotide primers designed in this study

Both RovA and RcsAB are involved in repressing the biofilm formation byY. pestis[5,9]. Biofilms formed byY. pestisin flea gut are essential for the transmission of plague[9]. Moreover, RovA has been demonstrated to be a protein thermometer that senses the temperature shift to change its conformation, thereby reducing its DNA-binding activity and enhancing its susceptibility to proteolytic degradation[11]. In addition, the temperature shift from 26 °C to 37 °C also triggered the downregulation ofrovAat the transcriptional level[4].According to the observations inY. pestisCO92, theYP3680-YP3663locus may play key roles in the flea cycle and bacterial interaction, and its expression also probably manifests in a temperature-dependent manner[2,3]. Thus, it is worth investigating the roles of theYP3680-YP3663locus in the flea cycle and the regulation of RovA and RcsAB in this process in the future.

Taken together, the data disclosed that RovA bound to the promoter-proximal DNA region ofYP3680on theYP3680-YP3663cluster to activate its transcription, whereas the RcsAB complex directly repressed the transcription ofYP3680. These results provided us with a deeper understanding of T6SS regulation inY. pestisbiovar Microtus.

&These authors contributed equally to this work.

#Correspondence should be addressed to ZHANG Yi Quan, E-mail: zhangyiquanq@163.com, Tel: 86-513-89093910

Biographical notes of the first authors: LIU Lei, male,born in 1989, PhD, Associate Chief Physician, majoring in pathogenic mechanisms of bacteria and clinical blood transfusion; FANG Nan, male, born in 1981, PhD, majoring in microbial physiology and next-generation sequencing technology.

Received: August 19, 2021;

Accepted: November 21, 2021