A chemosensory-like histidine kinase is dispensable for chemotaxis in vitro but regulates the virulence of Borrelia burgdorferi through modulating the stability of RpoS

As an enzootic pathogen, the Lyme disease bacterium Borrelia burgdorferi possesses multiple copies of chemotaxis proteins, including two chemotaxis histidine kinases (CHK), CheA1 and CheA2. Our previous study showed that CheA2 is a genuine CHK that is required for chemotaxis; however, the role of CheA1 remains mysterious. This report first compares the structural features that differentiate CheA1 and CheA2 and then provides evidence to show that CheA1 is an atypical CHK that controls the virulence of B. burgdorferi through modulating the stability of RpoS, a key transcriptional regulator of the spirochete. First, microscopic analyses using green-fluorescence-protein (GFP) tags reveal that CheA1 has a unique and dynamic cellular localization. Second, loss-of-function studies indicate that CheA1 is not required for chemotaxis in vitro despite sharing a high sequence and structural similarity to its counterparts from other bacteria. Third, mouse infection studies using needle inoculations show that a deletion mutant of CheA1 (cheA1mut) is able to establish systemic infection in immune-deficient mice but fails to do so in immune-competent mice albeit the mutant can survive at the inoculation site for up to 28 days. Tick and mouse infection studies further demonstrate that CheA1 is dispensable for tick colonization and acquisition but essential for tick transmission. Lastly, mechanistic studies combining immunoblotting, protein turnover, mutagenesis, and RNA-seq analyses reveal that depletion of CheA1 affects RpoS stability, leading to reduced expression of several RpoS-regulated virulence factors (i.e., OspC, BBK32, and DbpA), likely due to dysregulated clpX and lon protease expression. Bulk RNA-seq analysis of infected mouse skin tissues further show that cheA1mut fails to elicit mouse tnf-α, il-10, il-1β, and ccl2 expression, four important cytokines for Lyme disease development and B. burgdorferi transmigration. Collectively, these results reveal a unique role and regulatory mechanism of CheA1 in modulating virulence factor expression and add new insights into understanding the regulatory network of B. burgdorferi.


Material and Methods
Multiple sequence alignment analysis (MSA) of spirochete CheA.MSA analysis was performed on CheA sequences for spirochetes whose genomes have been sequenced and annotated.Sequences were retrieved using Annotree (1) with the following search parameters: KEGG: K03407, percent identity: 30, E-value: 0.00001, percent subject alignment: 70, percent query alignment: 70.Sequences were obtained for all bacterial species where only spirochetes were analyzed.Within the spirochete CheA sequences, individual sequences were filtered and redundant sequences (both in sequence and species) were removed.From this list, the number of CheA isoforms per species were calculated and segregated into groups according to the total number of CheA isoforms per species.From the single CheA isoform group, we chose to analyze only the members from the order Treponematales (34 total sequences).For the two CheA isoform group, we chose to analyze only those members from the family Borreliaceae (i.e.Borrelia spp.and Borreliella spp.), totaling 40 individual unique sequences (i.e.20 CheA1 and 20 CheA2).Fasta files were fed into Clutal Omega7 using B. burgdorferi CheA1 and CheA2 as query sequences.

Construction of GFP reporter plasmids for the localization of CheA1 and CheA2 in B. burgdorferi.
For the localization of CheA1, primer pair P41/P42 was used to amplify 417 bp upstream of cheW2 encompassing the putative promoter region of the cheW2 operon (2) with engineered PstI and NdeI cut site at the 5' and 3' end, respectively; primer pair P43/P44 for cheA1 gene with engineered NdeI and NruI cut site at the 5' and 3' end, respectively; and lastly primer pair P45/P46 for the amplification of gfp gene containing a 5 x Gly linker, with engineered NruI and PstI cut site, respectively.The obtained 3 amplicons were cloned into pGEM-T-easy vector and step wise ligation using the engineered restriction sites was performed to obtain the full fusion product PW2-cheA1-gfp which was then cloned into pBSV2G vector at the PstI cut site (3), yielding cheA1-gfp/pBSV2G plasmid (Fig. S6A).Similarly, for cheA2-gfp/pBSV2G construct, flgB promoter (4) was PCR amplified using primer pair P47/P48 and fused to cheA2 orf amplified with primer pair P49/P50 followed by fusion to gfp gene using the same restriction sites as cheA1-gfp/pBSV2G.All primers used are listed in Table S1.
RNA isolation using Qiagen RNeasy mini kit.To prepare RNA for RNA-seq, 50 ml of mid-log phase B. burgdorferi cells cultured at 34°C, pH 7.4 were harvested and RNA was extracted using RNeasy mini kit following manufacturer's instruction.Briefly, bacterial pellet was resuspended in 700 μl Buffer RLT containing 1 % β-mercaptoethanol and vortexed vigorously for 5-10 seconds.Suspension was transferred to cell homogenizer tube and centrifuged at maximum speed for 1 minute.Equal volume of 70% ethanol was added to the flow through and mixed well by pipetting.The suspension was transferred to RNeasy Mini spin column placed in a 2 ml collection tube and centrifuged for 15 seconds at ≥ 8,000 x g. 350 μl of Buffer RW1 was added to the RNeasy spin column, and centrifuged for 15 seconds at ≥ 8,000 x g to wash the spin column membrane followed by on-column DNase I digestion at room temperature for 15 minutes.350 μl of Buffer RW1 was then added to the RNeasy spin column, incubated for 5 minutes followed by centrifugation for 15 seconds at ≥ 8,000 x g.The RNeasy Mini spin column was washed twice with 500 μl of Buffer RW2 before elution using 50 μl RNase-free water.RNA concentration was quantified using a NanoDrop and stored in -80°C freezer before subjecting to RNAseq analysis.
RNA-seq analysis.Extracted RNA was subjected to ribosomal depletion prior to sequencing run using Illumina MiSeq Reagent kit v3 for 600 cycles.Reads were mapped to Borrelia burgdorferi genome (GCA_000008685.2_ASM868v2)using Bowtie2 software.Genes that were differentially expressed between the wild type and mutant cells were determined using DEseq2 pipeline with the Benjamini-Hochberg (BH) adjustment to calculate for each gene an adjusted p (padj) to determine the fraction of false positive rate (FDR) among them.Differentially expressed genes identified using a 5% FDR (padj < 0.05) and a two-fold expression difference were considered to identify genes that are differentially regulated in the mutant relative to the wild type.Significantly differentially expressed genes (DEG) were then plotted on a Volcano plot.

a
The underlined sequences are the engineered restriction cut sites for DNA cloning; F, forward; R, reverse.