Identification of proteins exhibiting in vitro RNA chaperone activity through gradient profiling in the Lyme disease spirochete, Borrelia burgdorferi.

Introduction

RNA-binding proteins (RBPs) are involved in cellular processes essential to bacterial survival and fitness; their activities range from enabling translation (ribosome-binding proteins) to regulating transcription, translation, and genome defense. RBPs execute these functions by serving as RNA matchmakers, scaffolds, recruiters, protectors, and structural remodelers [1,2]. ‘Conventional’ RBPs bind to sequence and structural motifs of RNA via well-defined RNA-binding domains, such as Sm and Sm-like domains, Fin-O domain, KH domains, double-stranded RNA-binding domains, S1 domain, cold-shock domain, PAZ/PIWI domains, and DEAD box helicase domains [3,4]. However, an increasing number of proteins have been identified that lack annotated RNA-binding domains but are capable of binding RNA and, so, have been termed ‘unconventional’, ‘enigma’, or ‘moonlighting’ RNA-binding proteins [5–10]. Several themes have emerged, including the recurrent identification of metabolic enzymes as RBPs and the frequent association of RNA with proteins containing intrinsically disordered regions (IDRs). Discovery of new and unconventional RBPs has accelerated due to advances in global RBP identification methods utilizing mass spectrometry-based proteomics. Several global RNA-centric approaches have recently been developed for bacterial RNA lacking poly(A) tails. Most of these methods rely on UV crosslinking RNA–protein complexes followed by organic phase extraction or silica-based purification [8–10]. In contrast, gradient fractionation followed by RNA sequencing (Grad-seq) and its iterations do not require UV crosslinking and rely on separating ribonucleoprotein complexes (RNPs) from cell lysate using a glycerol gradient or size-exclusion chromatography, followed by MS and RNA sequencing on fractions from the gradient [11–16]. Grad-seq identifies proteins that co-sediment with different classes of RNAs, but not all proteins identified are RNA binding proteins. Gradient sedimentation with RNase treatment (GradR), which includes an RNase treatment to identify RNase sensitive sedimentation profiles, improves the identification of bona fide RNA binding proteins. Grad-seq approaches have been applied to Gram-positive, Gram-negative and cyanobacteria leading to the discovery of novel RBPs and RNA-protein complexes [17].

RBPs can transiently interact or form a stable complex with RNA generating RNPs. Stable RNPs in bacteria have diverse functions, ranging from translation via the ribosome, to regulation and matchmaking involving Hfq, ProQ, and CRISPR complexes (recently reviewed by Gerovacet al. [17]). Stable RNP complexes are characterized by longer lifetimes, lower off-rates (kinetic stability), and higher affinities, with sub-micromolar disassociation constants (thermodynamic stability), compared to transient RNA-RBP interactions. Transient interactions between RNA and RBPs are inherently short-lived, enabling the RBPs to dynamically regulate diverse processes in response to cellular signals. Many mechanisms of RNA-based gene regulation rely on RNA–protein interactions to modulate RNA structures, facilitate RNA–RNA interactions, or occlude binding sites of other proteins or RNAs [1–3,18]. RNA chaperones are a group of RBPs capable of disrupting RNA secondary and tertiary structure (unwinding, unfolding, and strand displacing) and often accelerating base-pairing of RNAs (annealing) [19–22]. RNA chaperones perform their functions through both stable and/or transient interactions with RNA [20,23–27].

Borrelia (Borreliella) burgdorferiis a causative agent of Lyme disease and casts a wide shadow in the United States with an estimated 476,000 patients treated annually [28].B. burgdorfericycles in nature between a tick vector and vertebrate host, requiring phase-specific gene regulation. Transcriptomic studies have generated comprehensive maps of differentially regulated genes under various biologically relevant growth conditionsin vitroand, more recently, the transcriptome has also been definedin vivoduring tick feeding [29–31]. However, our understanding of posttranscriptional gene regulation remains limited. RBPs posttranscriptionally regulate gene expression by interacting with many, if not most, RNAs, including rRNAs, tRNAs, mRNAs, and regulatory RNAs such astrans-encoded small RNAs (sRNAs) and antisense RNAs (asRNAs). A large number of sRNAs have been identified inB. burgdorferivia transcriptomic studies [32–37], but only a few have been characterized [38–40]. A handful ofB. burgdorferiproteins have been reported to have RNA-binding capabilities, including BpuR, BosR, EbfC, SpoVG, CsrA, KhpA, KhpB, PlzA, and FlgV (previously annotated as HfqBb) [29,41–51]. However, the RNA chaperone activity, mechanisms of regulation and comprehensive RNA interactomes have not been identified for these proteins.

In this study, we employed gradient fractionation of RNPs followed by mass spectrometry to identify putative RNA binding proteins. We then evaluated thein vitroRNA chaperone activity of a selected subset as a functional pre-screening strategy to prioritize high-confidence candidates for futurein vivocharacterization, given the technical challenges associated with genetic manipulation inB. burgdorferi. We also chose to examine thein vitroRNA chaperone activity of a handful of proteins that were annotated or reported as nucleic acid-binding proteins. Most of the nucleic acid-binding proteins demonstrated RNA chaperone activity, except for the single-stranded DNA binding protein, SSB. The four putative RBPs (BB0749, BB0713, BB0796, and CheY2) also demonstrated strong RNA chaperone activity. Co-immunoprecipitation (co-IP) assays show that BB0749 binds RNA inB. burgdorferi. Future studies will focus on the identification of binding partners and mechanisms of the newly identified RBPs.

Results and discussion

Glycerol gradient fractionation ofB.burgdorferiRNA-protein complexes

To identify potential RBPs and RNA chaperones, we fractionatedB. burgdorfericell lysate on a glycerol gradient to separate RNP complexes by sedimentation as seminally described by Smirnovet al. [11] in gradient sequencing (Grad-seq) (Figure 1A). Following sedimentation, 20 fractions were collected, and each fraction was analysed for both RNA and protein constituents on denaturing gels (Figure S1). The fractionation pattern of RNAs and proteins in the gradient demonstrates successful sedimentation and separation of RNP complexes. Consistent with previous reports, sRNAs, including tRNAs, were detected in fractions 1–5, rRNA was found in fractions 9–18 (Figure S1), and RNAP subunits co-fractionated with 6S RNA primarily in fractions 4–7 (Figure 1B) [11–17]. In Grad-seq and its iterations, the RNA and proteins found in each Grad-seq fraction were identified by RNA-seq and mass spectrometry, respectively [11,15,16]. However, we were particularly interested in identifying proteins that bind sRNAs and nascent mRNAs and proceeded with LC MS/MS of fractions 1–10. One hundred and forty-five proteins were identified in fractions 1–10 and were manually classified by annotated and predicted functions using UniProt (Figure S2 and Table S1) [52]. The fractionation pattern of a subset of these proteins identified by LC MS/MS, including annotated or reported RNA and/or DNA-binding proteins (SSB, RRF, BosR, BpuR, SpoVG, KhpA, KhpB, and EbfC) are depicted inFigure 1B. Most of the recognized nucleic acid-binding proteins, BosR, BpuR, EbfC, RRF, and SSB, along with uncharacterized proteins BB0749 and BB0796, are found predominantly in fractions 1–5, which co-sediment with sRNAs. Uncharacterized protein BB0713, KhpB and SpoVG resolved in fractions 4–8 together with the RNAP subunits, while KhpA was identified in all fractions, but was more abundant in fractions 3–8. Western blot analyses of BosR, BB0749, and RNAP β-subunit in fractions 1–10 validate mass spectrometry results (Figure 1B).

Glycerol gradient fractionatingB. burgdorferiRNA-protein complexes. (A) Schematic of glycerol gradient experimental strategy.B. burgdorfericell lysate was fractionated in a glycerol gradient. Low molecular weight (LMW) complexes resolve towards the top of the gradient and high molecular weight (HMW) complexes resolve towards the bottom. 20 fractions were collected and analyzed for protein and RNA content. Mass spectrometry was performed on fractions 1–10. (B) Heat map of a subset of proteins identified in fractions 1–10. A 6S RNA northern blot and RNAP β-subunit Western blot demonstrates successful fractionation of RNPs and validate MS analyses. Western blots of BosR and BB0749 also validate MS gradient data.

Not all proteins that co-sediment with specific classes of RNAs are RBPs or components of RNA complexes. Proteins identified as putative RBPs through Grad-seq are often validated as bona fide RBPs using co-IPs of RNA with the protein or other orthogonal approaches. Co-IP assays require epitope-tagged proteins or highly specific antibodies, both of which are labour intensive and present certain limitations. Epitope-tagging can alter the native structure and function of the protein and generating antibodies with sufficient specificity is not always successful. Moreover, genetic manipulation ofB. burgdorferitakes longer than in most bacteria due to its slow growth, further complicating generation of epitope-tagged proteins. Therefore, we choose to biochemically analyze a subset of proteins identified through gradient profiling to assess theirin vitroRNA chaperone activity before assayingin vivoRNA binding by RNA co-IP. The goal was to generate a focused list of strong candidate proteins based onin vitroactivity, which can serve as a foundation for future in depthin vivocharacterization. Selection criteria of proteins included their co-sedimentation with sRNAs and/or RNAP, as well as their predicted structures and/or annotated domains (Table S2). BB0749 is an uncharacterized protein that has a 40-residue IDR. Despite their lack of stable structure, IDRs can function as the sole RNA binding domain of an RBP and can transition to an ordered state upon binding to RNA [5]. BB0749 co-sedimented in fractions 1–3 but does not contain any annotated domains known to bind RNA, other than an IDR. Uncharacterized protein BB0713 has an annotatedN-terminal coiled-coil domain and aC-terminal C4-type zinc-ribbon domain. The zinc-ribbon domain is rich in aromatic and positively charged amino acids and consists of two Zn knuckles scaffolded by two β-strands [52]. Aromatic and positively charged amino acids are often residues important for RNA binding [4]. In addition, the Zn knuckle is a zinc finger subtype known to interact with RNA [4]. AHelicobacter pyloriprotein with the same Zn ribbon 9 domain bindsflaAmRNA [53]. CheY2 is annotated as a chemotaxis response regulator and is necessary for vertebrate infection [54]. TheB. burgdorferigenome encodes three CheY response regulators (CheY1-3) but only CheY3 was found to be essential for motility and chemotaxis [55]. Xuet al. [54] hypothesized that CheY2 functions as a transcriptional or posttranscriptional regulator of virulence genes. None of the CheY response regulators inB. burgdorfericontain annotated RNA-binding or DNA-binding motifs. We also chose to examine the RNA chaperone activity of several known or predicted nucleic acid-binding proteins (BosR, BpuR, FlgV, KhpA, EbfC, RRF, and SSB).

RNA annealing and strand displacement activities of putative RBPs

RNA chaperones are RBPs that restructure RNA by disrupting secondary and tertiary structures (unwinding, unfolding, and/or strand displacement) and often accelerate base-pairing (annealing) [19–21,26,56]. We employed an in-solution coupled RNA annealing and strand displacement assay to determine RNA chaperone activities of a subset of the putative RBPs identified in gradient profiling. The assay was originally developed by the Schroeder lab and utilized fluorescence resonance energy transfer (FRET) to measure the annealing and strand displacement rates of RNAs labeled with Cy3 and Cy5 [25,57–61]. We modified the assay to measure a fluorescent signal from a fluorophore that is modulated by a quencher, rather than FRET (Figure 2A). Two previously published fully complimentary 21-mer unstructured RNAs were used in the assay [50,61]. M1 RNA was 5’ labelled with a 6-FAM fluorophore (FAM-M1) and J1 was 3’ labeled with Dabcyl (J1-Dab) (Table S3). RNA annealing (phase I) was initiated by injecting J1-Dab into a reaction mixture containing equimolar amounts of FAM-M1, either in the presence or absence of protein at 37°C. The fluorescent signal was quenched over time as J1-Dab RNA base-paired with FAM-M1. Strand displacement (phase II) was initiated by injection of tenfold excess unlabeled M1 (uM1) into the reaction and the fluorescent signal increased over time as uM1 replaced FAM-M1 in the duplex with J1-Dab, releasing the fluorophore from the quencher.

B. burgdorferiproteins with in-solution RNA annealing and strand displacement assay. (A) Schematic of in-solution annealing and strand displacement experimental strategy. The fluorescence of the 5’ FAM-labeled M1 RNA (FAM-M1), with or without protein, is measured for 40 s. The RNA annealing reaction is started by injection of 3’ Dabcyl-labeled J1 (J1-Dab) RNA. Fluorescent measurements were taken every 10 s for 200 s. Base-pairing of FAM-M1 and J1-Dab results in quenching of the FAM-fluorophore (M1J1). Unlabeled-M1 (uM1) is injected at 240 s and fluorescence is measured for another 200 s, with 10 s intervals. Strand displacement is observed as uM1 replaces FAM-M1 in the duplex (uM1J1), releasing FAM-M1 and increasing fluorescence. (B and C) time-resolved curves were normalized to 1 at 40 s and 0 at 240 s. Each vertical dotted line represents the time that RNA was injected into each sample. RNA without protein (red circle) or GrpE (teal triangle) served as negative controls. StpA, an RNA chaperone fromE. coli, was used as a positive control (purple square).B. burgdorferiproteins BpuR (open pink triangle), FlgV (green hexagon), RRF (open red square), BB0713 (open lavender diamond), BB0749 (blue diamond) and BB0796 (black diamond) with annealing rates at least 2-fold higher than RNA only are displayed in (B). WhereasB. burgdorferiproteins BosR (open magenta upside-down triangle), CheY2 (blue upside-down triangle), EbfC (black square), KhpA (open teal square) and SSB (open orange circle) with annealing rates less than 2-fold compared to RNA only are displayed in (C). Error bars represent the standard error of the mean (SEM) of three replicates.

The fluorescent signal of the reaction at the start of the assay is variable depending on the protein added to the reaction (Figure S3A and B). Therefore, this assay focuses on the kinetics of the reactions. To compare and visualize the results the fluorescent signal was normalized to 0 at the time point (240 s) directly before uM1 injection (Figure S3C and D) and normalized to 1 at the time point (40 s) direct prior to J1-Dab injection (Figure S3E and F). In the absence of protein (RNA only), the short fully complementary unstructured RNAs used in the assay anneal with an observed annealing rate constant of 3.15 × 10−6s−1(Figure 2and Table S4).E. coliStpA, an RNA chaperone with known annealing and strand displacement activity [19,25,26,58–62], was used as a positive control andB. burgdorferiGrpE was used as a negative control, as previously described [50]. As expected, StpA accelerated the rate of RNA annealing 17.3-fold compared to RNA alone, while GrpE demonstrated minimal to no RNA annealing compared to RNA alone (Figure 2B,C).B. burgdorferiBpuR, FlgV, RRF, BB0713, BB0749, and BB0796 proteins accelerated the rate of RNA annealing by over twofold compared to RNA only (5.4-fold, 5.4-fold, 2.7-fold, 2.5-fold, 16.5-fold, and 21.7-fold) (Figure 2Band Table S4). While BosR, CheY2, KhpA, SSB and EbfC accelerated RNA annealing by less than twofold compared to RNA alone (1.5-fold, 1.6-fold, 1.2-fold, 1.2-fold, and 1.1-fold) (Figure 2Cand Table S4).

Strand displacement (phase II) of the reaction is initiated by the injection of tenfold molar excess of uM1 (Figure 2). If strand displacement occurs, then uM1 replaces FAM-M1 in the duplex with J1-Dab, resulting in increased fluorescence. However, simultaneous annealing of remaining single-stranded J1-Dab with FAM-M1 or uM1 can continue, resulting in transient fluorescent quenching until the remainder of J1-dab is fully annealed. The extent and detection of annealing during strand displacement depends on the amount of single-stranded FAM-M1 at the initiation of strand displacement as well as the presence of proteins with strand displacement and/or annealing activity. For instance, in a reaction with a protein that has both annealing and strand displacement activity (StpA), FAM-M1 is largely duplexed with J1-Dab at the end of phase I, and the fluorescence is quenched and plateaued (Figure 2and S3). Upon uM1 injection, strand displacement is initiated and FAM-M1 is replaced in the duplex with uM1 and fluorescence increases. Any remaining FAM-M1 and J1-Dab annealing is not detected during strand displacement reaction (Figure 2and S3). In a reaction with a protein with strong annealing activity and minimal strand displacement activity (e.g. RRF), fluorescence is quenched and plateaued at the end of phase I. Again, most of FAM-M1 is duplexed with J1-Dab and continued annealing is not detected after initiation of phase II. However, the fluorescent signal only increases slightly and remains relatively stable, suggesting little to no strand displacement activity (Figure 2Band S3). In the absence of protein (RNA only) or a protein lacking annealing or strand displacement activity (e.g. GrpE), RNA annealing continues after initiation of strand displacement (Figure 2Band S3). The rate of annealing is much slower in these reactions and more single-stranded FAM-M1 and J1-Dab remain at the time of uM1 injection. However, after uM1 injection the ratio of uM1:FAM-M1 is high and most of J1-Dab anneals with uM1, leading to reduced quenching and a slower apparent annealing rate, compared to phase I. Consequently, we cannot directly compare the strand displacement and negative control’s reaction rates. Therefore, a protein is considered to have strand displacement activity in this assay if the strand displacement rate is positive (increasing fluorescence), and the fluorescence must increase at least 1000 arbitrary units (AU) during phase II in the unnormalized data (Figure S3, S4 and Table S4). As anticipated, strand displacement does not occur in the RNA only and GrpE reactions, and StpA has robust strand displacement activity. BosR, BpuR, CheY2, FlgV, BB0713, BB0749, and BB0796 all demonstrate strand displacement activity, while RRF, SSB, KhpA, and EbfC do not (Figure 2, S4 and Table S4).

Dipartite RNA gel annealing and strand displacement assays were also conducted to examine RNA chaperone activity of the proteins as previously described [50]. Cy3-labeled M1 and Cy5-labeled J1 RNAs were used in the gel assays to observe RNA annealing and strand displacement separately. To start the gel RNA annealing assay, Cy5-labeled J1 was added to a master mix containing Cy3-labeled M1 in the presence or absence of a protein. Samples were removed from the reaction over time, mixed with stop buffer containing an excess of unlabeled M1 RNA (uM1) and resolved by electrophoresis on a native polyacrylamide gel (Figure 3A). RNA annealing is detected as the formation of double-stranded J1M1 RNA (yellow band). Representative images for each protein are shown in Figure S5A. The formation of J1M1 (RNA annealing) over time was plotted and quantified using the ratio of dsRNA (J1M1) to total RNA (J1M1 and J1uM1) using only the Cy5 signal (Figure 3B). As expected, StpA increased the rate of RNA annealing by 8.8-fold compared to the RNA only control and the amount of dsRNA formed at the endpoint of the assay was significantly higher than RNA only (Figure 3B, S5B and Table S4). BosR, BpuR, CheY2, EbfC, FlgV, KhpA, RRF, BB0713, BB0749, and BB0796 all accelerated RNA annealing (5-fold, 2.7-fold, 7.7-fold, 3.0-fold, 18.4-fold, 2.1-fold, 4.7-fold, 4.3-fold, 8.3-fold and 2.1-fold) compared to RNA only and formed significantly more dsRNA at the endpoint of the assay compared to RNA alone. SSB and the negative control GrpE did not demonstrate RNA annealing activity in the gel assay (Figure 3B, S5B and Table S4). BpuR, FlgV, RRF, BB0713, BB0749, and BB0796 accelerated the rate of annealing by more than twofold compared to RNA only in both the in-solution and gel-based assay. However, BosR, CheY2, EbfC, and KhpA increased the rate of annealing less than twofold compared to RNA alone in the in-solution assay but performed better in the in-gel assay demonstrating greater than twofold increase in annealing rate. These differences may be attributed to the temperature conditions under which the assays were performed (23°C for the gel-based assay and 37°C for the in-solution assay), the difference in the length of time of the assays or to differences in the reaction milieu between the two assays, including the 5’and 3’ modifications of the RNAs.

B. burgdorferiproteins with gel RNA annealing activity. (A) Schematic of RNA annealing assay using Cy3-labeled M1 (green) and Cy5-labeled J1 (red) RNAs. The reaction is initiated by adding Cy5-labeled J1 to a mixture of Cy3-labeled M1 RNA and buffer with or without protein. Annealing is measured over a time course of 15 min resulting in an increasing yellow band (J1M1). Stop buffer and excess unlabeled-M1 (uM1) are used to halt the reaction. Excess uM1 binds to J1, yielding a dsRNA that appears as a red band (J1uM1). This band is slightly smaller than J1M1 due to the lack of the Cy3-fluorophore on uM1. Representative images of RNA annealing gels are shown in Figure S5A. (B) Normalized linear regression of the ratio of J1M1 dsRNA to total RNA at each timepoint produced the annealing curves for each sample. RNA without protein (red circle) and StpA (purple square) served as negative and positive controls, respectively. FlgV (green hexagon), BB0749 (blue diamond), CheY2 (blue upside-down triangle), BB0713 (open lavender diamond), RRF (open red square), BosR (open magenta upside-down triangle), BpuR (open pink triangle), EbfC (black square), KhpA (open teal square), and BB0796 (black diamond) displayed gel annealing activity. Error bars represent the SEM of three replicates.

Gel strand displacement assays were initiated by adding unlabeled-M1 competitor RNA into a reaction that contained preformed dsRNA (Cy3-labeled M1 and Cy5-labeled J1). Samples were removed from the reaction over time and resolved by electrophoresis on a native polyacrylamide gel (Figure 4A). Strand displacement was detected by the loss of M1J1 dsRNA (yellow band) and the formation of uM1J1 dsRNA (red band); representative images are shown in Figure S6. The dissolution of M1J1 (yellow band) and the formation of uM1JI (red dsRNA band) was plotted over time and quantified using the ratio of dsRNA (J1M1) to total RNA (J1M1 and J1uM1) using only the Cy5 signal (Figure 4B). As noted earlier, strand displacement does not occur spontaneously at 37°C under these conditions and, consistent with this, strand displacement was not observed in the RNA only or GrpE reactions, whereas StpA increased the rate of strand displacement 16.5-fold compared to RNA alone and had significantly less of the dsRNA M1J1 at the end of the assay compared to RNA alone (Figure 4B, S6C and Table S4). BB0749 also demonstrated strand displacement activity, increasing the rate of strand displacement 20.6-fold compared to RNA alone, while CheY2 and FlgV exhibited modest strand displacement activity, increasing the rate of strand displacement 2.6- and 2.8-fold compared to RNA alone (Figure 4Band Table S4). BB0749, CheY2 and FlgV also finished the assay with significantly less of the dsRNA M1J1 compared to RNA alone (Figure S6C). All the other proteins (BpuR, BB0713, BB0796, BosR, KhpA, EbfC, RRF, and SSB) lacked strand displacement activity in the gel assay (Figure S6B).

B. burgdorferiproteins with gel RNA strand displacement activity. (A) Illustration of the gel RNA displacement assay. Preformed dsRNA (M1J1) was generated with 5’ Cy3-labeled M1 and 5’ Cy5-labeled J1. The strand displacement reaction is initiated by adding unlabeled M1 (uM1) to J1M1 in the presence or absence of protein. Samples are withdrawn from the reaction over time and analyzed by native polyacrylamide gel electrophoresis. The amount of preformed J1M1 dsRNA (yellow band) will decrease with a concomitant increase in the J1uM1 dsRNA (red band) formation over time if the protein has strand displacement activity. Representative images of the RNA strand displacement gels are shown in Figure S6A. (B) Normalized non-linear regression of the ratio of J1M1 dsRNA to total RNA for each time point yielded stand displacement curves. Only the Cy5 channel was quantified and used when graphing the curves. RNA only (red circle) and StpA (purple square) were used in the assay as negative and positive controls, respectively. FlgV (green hexagon), BB0749 (blue diamond), and CheY2 (blue upside-down triangle) displayed gel strand displacement activity. Error bars represent the SEM of three replicates.

KhpA, EbfC, RRF and SSB did not demonstrate strand displacement activity in either the in-solution or gel-based assay. In contrast, BB0713, BB0796, BosR, and BpuR demonstrated strand displacement activity in the in-solution assay but not in the gel-based strand displacement assay (Table S4). Strand displacement of one RNA in a duplex by another RNA is a concerted process of opening the initial double strand and annealing of the new duplex. RBPs with strand displacement activity destabilize double strands, either from the ends or bulges in base-paired regions. A third strand uses the destabilized regions to initiate invasion. RBPs that strand displace and exhibit RNA annealing activity can catalyze strand displacement by both destabilizing double stranded regions and accelerating the annealing reaction of the invading strand. The in-solution and gel-based assays use the same RNA sequence, but have different fluorophores conjugated to different ends of the RNA, which may explain the different results between the two assays for some proteins. The in-solution assay has the 6-FAM fluorophore on the 5ʹ end of the M1 RNA and the quencher on the 3ʹ end of the J1 RNA, while the gel-based assay has Cy3 and Cy5 fluorophores attached to the 5ʹ ends of M1 and J1. Cy3 and Cy5 dyes linked to the 5ʹ ends of a DNA duplex have been reported to stabilize the duplex [63,64], and this may explain why some proteins did not display strand displacement activity in the gel-based assay. Moreover, having the 6-FAM and Dabcyl quencher proximal to each other on the RNA duplex in the in-solution assay may destabilize the dsRNA duplex via steric hindrance.

In vivoantitermination andinvitroRNA unwinding activities of putative RBPs

Antitermination activity of proteins can be examinedin vivousing the RL211 strain ofE. coli[50,65–68]. The chloramphenicol (Cm) acetyltransferase gene (cat) is preceded by the Rho-independenttrpLtranscriptional terminator stem-loop that efficiently terminates transcription of thecatgene resulting in Cm susceptibility. Expression of a protein with antitermination activity that can unwind or remodel the terminator stem loop results in transcription of thecatgene and resistance to Cm (Figure 5A). We assayed the antitermination activity of our cohort of putative RBPsin vivousing the RL211 strain. The RL211 strain was transformed with plasmids harbouring the genes of the RBPs under control of thelacpromoter. Cultures were grown overnight in the absence of Cm and then spotted onto plates with and without Cm plus IPTG (Figure 5B). All strains grew similarly in the absence of Cm and RL211 did not grow on Cm. Substantial growth on Cm was observed in RL211 strains expressingstpA,bosR,cheY2,bb0713, andflgVdemonstrating theirin vivoantitermination activity. RL211 strains expressingebfC,frr,bb0796, andbpurgrew modestly on Cm plates, while no growth was observed in strains expressingkhpA,bb0749andgrpE.

In vivoRNA anti-termination andin vitroRNA unwinding assays. (A) Illustration of thein vivo trpLterminator assay. ThetrpLterminator stem loop is fused to the chloramphenicol (Cm) acetyltransferase (cat) gene. In the RL211 strain, the terminator stem loop is formed, transcription is terminated, and thecatgene is not expressed resulting in Cm susceptibility and no growth on Cm. However, an RNA chaperone (orange circle) expressed in RL211 that unwinds the terminator stem loop will allow transcription elongation to proceed and thecatgene to be expressed, resulting in Cm resistance and growth on Cm [50]. (B)E. coliRL211 strains expressing denoted proteins or the positive control StpA were grown in LB overnight to an OD600 of ~1. The overnight cultures were plated on LB (no antibiotic) or Cm plus 1 mM IPTG, and growth was assessed. (C) schematic of thein vitrounwinding assay. A truncatedtrpLterminator RNA was 5’ and 3’ labeled with 6-FAM and Dabcyl, respectively. Protein was added to thetrpLfluorescent reporter RNA and incubated for 10 min. Florescence was measured to quantify RNA unwinding. (D) a bar graph illustrating the relative fluorescence units of each reaction after subtracting the background levels of unwinding measured in the RNA only reaction. Error bars represent the SEM of three replicates. Relative fluorescent units of unnormalized data were analyzed by a two-way ANOVA with multiple comparisons (p-value ≥ 0.050, ns,p-value, ≤0.050,,p-value ≤ 0.0001, ***).

The antitermination activity observed in the RL211 strain is likely due to the expressed protein remodeling, unwinding or melting thetrpLterminator stem loop. We developed a fluorescent reporter to assay the ability of the proteins to remodel or unwind thetrpLterminatorin vitro. ThetrpLterminator was 5ʹ and 3ʹ labeled with 6-FAM and Dabcyl, respectively. The labeledtrpLterminator RNA was incubated with the putative RBPs, and fluorescence was measured to assay RNA unwinding or remodeling (Figure 5C). BpuR, FlgV, KhpA, BB0713, BB0749, and BB0796 exhibited statistically significant stem-loop remodelingin vitro(Figure 5D). In contrast, BosR, CheY2, EbfC, RRF, SSB, and GrpE did not demonstrate significant RNA remodeling activity compared to RNA alone under the same conditions (Figure 5D).

Thein vivoandin vitrodata correlate well for BpuR, FlgV, BB0796, and StpA. StpA and FlgV exhibited robust growth on the Cm plates and efficiently remodeled thetrpLterminatorin vitro. BpuR and BB0796 showed more modest growth on Cm plates consistent with their comparatively weaker remodeling activity of thetrpLterminatorin vitro. In addition, our negative control GrpE did not demonstratein vivoantitermination activity orin vitrounwinding activity (Figure 5). BB0713, BosR and CheY2 exhibited strongin vivoantitermination activity as demonstrated by substantial growth on Cm plates. However, thein vitroassay revealed that BB0713 only moderately unwound thetrpLterminator, while BosR and CheY2 showed no statistically significant unwinding of thetrpLterminator (Figure 5B-D). BB0713 and BosR both contain two CXXC motifs in theirC-terminal domains, which often coordinate Zn2+and can be involved oligomerization of proteins. BosR has been reported to bind both DNA and RNA [44,69,70], and DNA binding requires dimerization [71]; however, the role of the dimer in RNA binding has not yet been investigated. The discrepancy between the activity of BosR in the two assays may be explained by the presence of DTT in thein vitroassay buffer, which may influence folding, oligomerization and/or Zn2+binding of BosR, resulting in inhibition of RNA binding and potentially explaining the discrepancies. The RNA annealing and strand displacement assays also utilize the same buffer containing DTT, and the modest activity demonstrated in these assays by BosR and BB0713 may be due to inhibition of oligomerization or Zn2+coordination and proper folding of the proteins. Prior studies demonstrated that a loss of the cysteine residues in the CXXC motifs alters the oligomeric structure of BosR [71]. In addition, the CXXC domains may serve as redox sensors that affect the binding activity of BosR to both RNA and DNA [71]. A deeper understanding of the role of tertiary and quaternary structure in these activities awaits further investigation.

Posttranslational modifications (PTMs) of RBPs, including phosphorylation, acetylation, and methylation, regulate their interactions and functions [4,5,72,73]. InB. burgdorferi, CheY2 is phosphorylated by CheA and is postulated to act as a transcriptional or posttranscriptional regulator [54]. Thus, phosphorylation may act as a molecular switch regulating the RNA chaperone activity and RNA binding of CheY2. CheY2 displayed strongin vivoantitermination activity, but did not unwind thetrpLterminatorin vitro. We postulate that the phosphorylation state of CheY2 might differ between the purified recombinant protein and thein vivoexpressed protein in RL211, leading to the observed discrepancy. EbfC and RRF exhibited modest growth on Cm plates, but did not unwind thetrpLterminatorin vitro, suggesting that thein vitrounwinding assay conditions may be missing important factors foundin vivo. Notably, thetrpLterminator we designed for the assay does not include any single-stranded regions other than the stem loop and thus the structure differs from the stem loopin vivo. The proteins that performed well in thein vivoassay and poorly in thein vitroassay may require a toehold or single stranded region before the stem loop in order to bind and unwind the RNA.

RL211 strains expressing BB0749 and KhpA did not grow on Cm plates but unwind thetrpLterminator robustlyin vitro. Unfortunately, we cannot determine if the proteins are being properly expressed in the RL211 strain while growing on the plates and the lack ofin vivoantitermination activity of these proteins may be due to poor protein expression or toxicity due to overexpression of the protein.

BB0749 binds RNA inB.burgdorferi

Evidence ofin vivoRNA binding by newly identified RBPs is typically required to classify them as bona fide RBPs. Crosslinking co-immunoprecipitation (CLIP) of the protein bound to RNA is often used to demonstratein vivoRNA binding. However, antibody to the protein or an epitope-tagged version of protein is required for CLIP, therefore high throughput screening by CLIP is not feasible. AnN-terminal 3×FLAG tagged BB0749 fusion protein was constructed and inserted into thebb0749genomic location of the linear chromosome. Thebb0749:3×flagstrain was grown to mid-log phase (5 × 107cells ml−1) in one-liter batches and split in two; half of the culture was UV crosslinked while the other half was left untreated as a control. BB0749-RNA complexes were immunoprecipitated with anti-FLAG antibodies and washed several times to remove non-crosslinked RNA from the protein. Western blot analyses demonstrate that BB0749 was enriched and pulled down in the IP in both the crosslinked and non-crosslinked samples (Figure 6A). However, substantially more RNA was immunoprecipitated with BB0749 in the crosslinked samples compared to the non-crosslinked samples demonstrating that BB0749 binds RNAin vivoinB. burgdorferigrown under culture conditions (Figure 6B). BB0749 harbours an IDR containing stretches of basic and acidic residues, as well as polar residues. IDRs can directly engage in protein-protein interactions and RNA binding. In eukaryotic cells, IDRs are important for the formation of different types of RNPs, and posttranslational modifications of IDRs modulate their function [5,72–75]. IDRs have also been implicated in modulating liquid–liquid phase separation and formation of membrane-less organelles [74]. Further work is needed to determine the RNA-dependent function of BB0749 and the BB0749 targetome.

BB0749-3×FLAG binds RNA inB. burgdorferi. Crosslinking immunoprecipitation (CLIP) of BB0749-3×FLAG demonstrates RNA bindingin vivo. (A)N-terminally tagged BB0749 cultures were grown and split in half. One half was UV crosslinked (+CL), while the other was not (−CL). The cell lysates from the two samples were immunoprecipitated with anti-FLAG antibody. Cell lysates (input), supernatants after the immunoprecipitation (SN) and the immunoprecipitations (IP) were separated on a denaturing SDS polyacrylamide gel and transferred to a PVDF membrane and immunoblotted with anti-FLAG antibody. BB0749 (~50 kDa) is detected in all samples but is depleted in the SN and enriched in the IP. Preliminary mass spectrometry data suggest the smaller band detected and denoted with an * may be a small isoform or proteolytic fragment of BB0749 (~10 kDa). The heavy and light chain of the anti-FLAG antibody beads are also detected (~22 kDa and ~50 kDa). (B) Bioanalyzer gel image of non-crosslinked (−CL) and crosslinked (+CL) IP RNA samples. Green lines represent bioanalyzer lower marker. All experiments were performed in triplicate.

FlgV has RNA chaperone activity

FlgV (BB0268), previously termed HfqBb[49], anneals and strand displaces RNA in both the gel-based and in-solution assays. In addition, FlgV had antitermination activityin vivoand RNA unwindingin vitro(Table S5). Fifteen years ago, we identified BB0268 as an RBP that partially complementedrpoSposttranscriptional regulation in anhfqdeletion mutant inE. coliand we showedE. coli hfqpartially complemented thebb0268deletion mutant inB. burgdorferi. We termed BB0268 HfqBbbecause of the reciprocal complementation of the deletion phenotypes and demonstrated BB0268 bound RNA non-specificallyin vitro[49]. Recently, Zamba-Camperoet al. reported that BB0268 monomer is a flagellar protein and the Alphafold3 predicted structure of BB0268 is not similar to the structure of Hfq [48]. BB0268 was renamed FlgV and we agree that it was misannotated as HfqBb.However, Zamba-Camperoet al. also concluded that FlgV was not an RBP because RNA did not co-IP with FlgV after UV crosslinking [48]. Successful UV crosslinking relies on spatial arrangement and proximity of the RNA and protein, as well as the nucleotide and amino acid composition at the RNA protein interface [76]. Protein interactions with the phosphate backbone of RNA are suboptimal and are likely undetected by UV crosslinking. In addition, UV crosslinking efficiency is poor, approximately 1–5%; abundant and/or stable RNA-protein complexes are more likely to be captured than rare and/or transient complexes using UV crosslinking immunoprecipitation (CLIP) [76]. Despite these caveats, Zamba-Camperoet al. conclude that FlgV is not an RBP because levels of bulk RNA recovered after CLIP with FlgV do not match the levels of RNA recovered from two known RBPs,E. coliHfq andB. burgdorferiKhpB [48]. A direct comparison of the ability of these proteins to co-immunoprecipitate RNA is not feasible and, in fact, misleading, due to inherent structural differences between the proteins and the distinct mechanisms of RNA binding. Generally, instead of comparing RNA from co-IP’s of different proteins to determine RNA binding, a non-crosslinked sample is compared to the crosslinked sample to determine if the RNA observed in the co-IP is specific, as demonstrated for BB0749 (Figure 6). The study also included CLIP of endogenous FlgV with a polyclonal antibody produced against FlgV [48]. Again, a non-crosslinked control was not included, but a co-IP was performed on abb0268null mutant. The profiles of the RNA immunoprecipitated from the mutant is different than the wild-type strain, but the RNAs were not sequenced to determine if any specific RNA was pulled down with FlgV. In both cases, a transient interaction or an interaction with low abundance RNA would not be detected by comparing bulk RNA amounts and patterns.

FlgV forms a monomer and a dimerin vitroandin vivo[48,49]. While the monomer clearly plays a role in flagellar structure, the dimer was not observed in association with the flagella apparatus and its function was not discussed [48]. The data we presented previously and in this study demonstrate that FlgV has RNA chaperone activity and non-specific RNA bindingin vitro[49]. We hypothesize that FlgV dimer (and possibly monomer) moonlight as an RNA-binding protein. Moonlighting, enigma or unconventional RBPs are involved in biological processes with no apparent relation to RNA biology: thousands of unconventional RBPs have been identified [5,6,8,10]. Several dozen moonlighting RBPs have been validated by orthogonal approaches including co-IPs in eukaryotic cells. Notably, flagellar and flagellar associated proteins (FlhA, FliN, FliH, FlgN, FliL, FliS, and YgcR) were identified as RNA-binding proteins inE. coliusing total RNA-associated protein purification (TRAPP) [8]. TRAPP is based on UV crosslinking of RNA and RBPs in cells followed by purification of bulk RNA on silica beads under denaturing conditions. Proteins that are not crosslinked to RNA are washed away and proteins covalently crosslinked to RNA remain bound and purified with the RNA.In vivoidentification of flagellar and flagellar-associated proteins as RBPs using RNA-centric approaches in other bacteria suggest flagellar proteins can moonlight as RBPs [8].

Newly identified proteins with RNA chaperone activity

The primary objective of this study was to identify previously uncharacterized RNA binding proteins (RBPs) inB. burgdorferithat possess RNA chaperone activityin vitro. Among the proteins tested, uncharacterized proteins BB0713, BB0749 and BB0796 demonstrated robust RNA chaperone activity, yielding positive results in five of the six functional assays employed (Table S5). BB0713 and BB0749 both have domains that have been reported to bind RNA (Zn knuckle and IDR, respectively), while BB0796 does not. The Alphafold3 predicted structure of BB0796 has homology to the Skp (OmpH) family of proteins, which includes a coiled-coil domain. Skp has been reported as a molecular chaperone that interacts with unfolded proteins as they emerge in the periplasm inE. coli. In yeast and human cells, heat shock proteins, which also interact with unfolded and misfolded proteins, were reported to bind RNAin vivo[5,6,77,78]. In addition, CheY2 also demonstrated RNA chaperone activity in five of six functional assays (Table S5). Although CheY2 is annotated as a chemotaxis response regulator due to its sequence and structural homology to other chemotaxis response regulators, it is not necessary for chemotaxis or motility inB. burgdorferi. However, CheY2 is important for vertebrate infection and was postulated to be a transcriptional or posttranscriptional regulator of virulence gene expression [54]. Our data suggest CheY2 may regulate gene expression via its RNA chaperone activity. Notably, CheY and other chemotaxis associated proteins (CheW, CheZ, CheB, and Trg) were identified as RNA binding proteins via TRAPP inE. coli[8]. Chemotaxis proteins moonlighting as RNA binding proteins may be a conserved function of these proteins in bacteria.

Concluding remarks

We identified approximately 10% of the annotatedB. burgdorferiproteome in our mass spectrometry analysis of the first ten fractions of the gradient-separated cell lysate. The coverage of the proteome was less than what has been reported in Grad-seq for other bacteria [11–17]. The lower coverage may be due to limited input material, technical differences in preparing fractions, or not partitioning of the fractions before mass spectrometry. In addition, gradient profiling is not efficient at capturing low abundant and transient RNA-protein interactions. The RNPs are not crosslinked, and transient or unstable complexes can disassociate after cell lysis and during gradient fractionation. There are likely RBPs that were not identified in our assay due to these caveats. Global RNA-centric approaches, such as 2C or TRAPP, or protein-centric approaches are needed to further identify RBPs inB. burgdorferi. Notably, gradient profiling detected most known RBPs and identified several putative RBPs, which were further characterized forin vitroRNA chaperone activity. Further investigation is necessary to elucidate the cellular functions and RNA interactomes of both previously characterized and newly identified RBPs. RBPs are integral to numerous cellular processes and serve as critical regulators of gene expression, with significant implications for physiology and virulence. The identification and functional characterization of novel RBPs inB. burgdorferiwill enhance our understanding of the molecular mechanisms that facilitate spirochete transmission, host adaptation, and disease pathogenesis.

Materials and methods

Bacteria and strains

B. burgdorferilow-passage B31 strain (clone B31-5A4), which retains plasmids required for infectivity, was cultivated in Barbour – Stoenner – Kelly II (BSK-II) complete medium at 37°C. Unless otherwise stated, cultures were passaged to 1 × 105cells ml−1and grown to 1–3 × 107cells ml−1. Cell density was determined using a Petroff Hausser counting chamber as previously described [79,80].E. coliRL211 strain constructed by Landicket al. [65] was cultivated in lysogeny broth (LB) at 37°C and kanamycin (50 ug/ml) when appropriate.

Gradient fractionation

The protocol was adapted from Smirnovet al., 2016 [11].B. burgdorferiB31-5A4 cultures were centrifuged at 5,000 × g for 30 min at 4°C and the cell pellets were snap frozen in liquid nitrogen (LN2). In total 5 × 108to 1.5 × 109cells were lysed in 100 µl of lysis buffer (20 mM Tris-HCl, pH 7.5, 150 mM KCl, 1 mM MgCl2, 1 mM DTT, 1 mM PMSF, 0.2% Triton X-100, 20 units/ml DNase I (Thermo-Fisher Scientific™), 200 units/ml SUPERase-IN (Life Technologies)) with a stainless-steel ball cryogrinding (Retsch) in the Retsch MM400 at 30 Hz for 10 min, refreezing the sample in LN2every 3 min. The lysate was cleared via centrifugation at 18,928 ×gfor 5 min at 4°C. Cleared lysate was applied to a 10–40% (wt/vol) glycerol gradient and ultra-centrifuged in a linear Beckman SW40Ti tube at 62,909 ×gfor 17 h at 4°C. The gradient was separated into 20 equal fractions of 600 µL and approximately 90 µL of each fraction was removed for protein analyses. The remaining sample was used for RNA isolation.

Protein gel electrophoresis

Gradient fractions (10 µL of each fraction in Laemmli buffer) were resolved on 12% SDS-PAGE at 150 V for 1 h at room temperature followed by Coomassie Brilliant Blue staining, overnight. The gel was destained with distilled water for 4–10 h, changing the water approximately every thirty min.

RNA isolation and gel electrophoresis of gradient fractions

RNA was isolated from approximately 500 µL of each fraction. Each fraction was treated with 1% SDS and one volume of saturated phenol (pH 4.3) and incubated at 55°C for 5 min, centrifuged 18,928 ×gfor 10 min at 4°C and the aqueous layer moved to a new tube. RNA was precipitated with 0.3 M NaOAc and three volumes of ice-cold absolute ethanol at −20°C overnight. RNA was pelleted for 45 min at 18,928 ×gat 4°C and washed with 75% ice-cold ethanol. RNA was centrifuged at 18,928 ×gat 4°C for 20 min, the supernatant was removed, and the pellet was air-dried and resuspended in water. RNA was resolved on a denaturing 6% polyacrylamide gel with 7 M Urea in 1× TBE buffer. The gel was post stained with ethidium bromide.

Mass-spectrometry sample preparation and analysis

Glycerol gradient fractions were diluted with lysis buffer (4% (w/v) SDS, 0.1 M Tris pH 7.4) and reduced with 5 mM TCEP at 90°C for 10 min. Samples were cleared by centrifugation and transferred to a 30kDa spin filter (Amicon). Samples were washed twice with UA buffer (8 M urea, 0.1 M Tris pH 7.4) then alkylated with 50 mM iodoacetamide at room temperature for 20 min in the dark. Samples were washed again three times with UA and then three times with 0.1 M ammonium bicarbonate (ABC), 0.02% (w/v) deoxycholic acid (DA) then suspended in 150 µL 0.1 M ABC, 0.02% DA with trypsin overnight at room temperature. Tryptic peptides were eluted and DA was removed by phase transformation using ethyl acetate. Tryptic peptides were then desalted using Pierce C18 spin columns according to the manufacturer’s instructions and dried with a vacuum centrifuge. Tryptic peptides were loaded directly onto a Waters nanoACQUITY UPLC BEH C18 column (130 Å, 1.7 µm, 75 µm ×250 mm) and separated with a nanoAcquity UPLC (Waters) at 0.3 µL/min using gradients of 3 to 8% acetonitrile in 5 min then 8 to 35% acetonitrile in 118 min. Detection of peptides was performed with a LTQ Orbitrap Velos mass spectrometer (Thermos Scientific), scanning from 300 to 1800 m/z at 60,000 resolution with an AGC target of 1E6 and selecting the 10 most intense ions for MS/MS. Maximal ion injection times were 500 ms for FT (one microscan) and 250 ms for LTQ with an AGC target of 1E4. The normalized collision energy was 35% with activation Q 0.25 for 10 ms. Raw data files were searched against the UniprotBorreliella burgdorferiproteome database (uniprot-proteome_Borrelia_burgdorferi_UP000001807), using the MaxQuant/Andromeda search engine (version 2.6.3.0). Trypsin cleavage was searched allowing for up to two missed cleavages, with cysteine carbamidomethylation as a fixed modification, and proteinN-terminus acetylation and methionine oxidation as variable modifications. Mass tolerances were set to 20 ppm (first search) and 4.5 ppm (main search) for precursor ions, and 0.5 Da for ITMS MS/MS ions. All peptides and proteins were thresholded at 1% FDR. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the Proteomics Identifications Database partner repository with the dataset identifier PXD068462 and 10. 6019/PXD068462

Northern blotting

RNA was separated under denaturing conditions on pre-poured 6% Novex™ TBE-Urea (7 M) gels (Invitrogen™) in 1× TBE. RNA was denatured at 65°C for 15 min in 2× RNA load dye (Thermo-Fisher Scientific™) prior to loading on the gel. RNA was transferred to BrightStar™-Plus positively charged nylon membrane by electroblotting at 10 V for 1 h in 0.5× TBE. Membranes were UV crosslinked and probed with biotinylated RNA probes. A biotinylated 6S probe was generated fromin vitrotranscription of PCR-generated DNA template using the MAXIscript™ T7 Transcription Kit (Invitrogen™) with biotinylated-16-UTP (Roche) according to the manufacturer’s protocol. The T7 bacteriophage promoter was added to the 5ʹ end of the reverse primer that generated the 6S DNA template (Table S3). Probe length and integrity were examined by resolving thein vitrotranscription reaction on a pre-poured 6% Novex™ TBE-Urea (7 M) gel (Invitrogen™) in 1× TBE and post staining with ethidium bromide. The remainder of the reaction was ethanol precipitated and used for Northern blotting. Membranes were pre-hybridized in 4 mL of ULTRhyb™ Ultra-sensitive Hybridization Buffer (Invitrogen™) at 68°C for 30 min in 50-mL conical tubes in a hybridization oven with rotation. Subsequently, 500 µL of the pre-hybridization solution was transferred to 1.5-mL microfuge tube and the probe was added to the pre-hybridization solution to a final concentration of approximately 0.1 nM. The pre-hybridization buffer with probe was incubated at 95°C for 5 min, immediately returned to the prehybridizing membrane and incubated overnight at 68°C with rotation. Membranes were washed for 5 min in 2× SSC and 0.1% SDS at room temperature with rotation and twice for 15 min in 0.1× SSC, and 0.1% SDS at 68°C with rotation. Membranes were developed using the Chemiluminescent Nucleic Acid Detection Module Kit (Thermo-Fisher Scientific™) according to the manufacturer’s protocol and visualized on an Azure Sapphire Biomolecular Imager.

Immunoblotting

Each fraction was diluted in 2× Laemmli buffer (Sigma-Aldrich), heated to 95°C for 5 min and resolved on precast Criterion Any kD™ TGX Stain-Free™ Protein Gels (BioRad). Gels were semi-dry blotted onto pre-packaged Iblot™ 2 Transfer Stacks, PVDF membrane (Invitrogen™) using a custom Iblot2 program (20 V for 3 min, 23 V for 1 min, 25 V for 1 min). Membranes were blocked for 1 h in blocking buffer [1.5 g skim milk 1× TBST (1.5 M NaCl, 10 mM Tris – HCl, pH 7.5, 0.05% Tween 20)] (Thermo-Fisher Scientific™). Protein-specific antisera were used to detect RNA polymerase β subunit, BosR and BB0749 proteins in the gradient fractions. The mouse anti-RNA polymerase β subunit antibody was purchased (Biolegend), as was the anti-BosR antibody (General Biosciences), and rat anti-BB0749 was generated in this study. Primary antisera were diluted 1:20,000 in 1× TBST and incubated for 1 h at room temperature. Membranes were washed 3 × 5 min in 1× TBST. Secondary antibodies, goat anti-mouse HRP or goat anti-rat HRP (KPL) were diluted 1:20,000 in 1× TBST and incubated for 1 h at room temperature. Membranes were developed using chemiluminescence (ECL select, Amersham) and imaged on an Azure Sapphire biomolecular imager using the chemiluminescence setting with High Dynamic Range (HDR), auto exposure, and single capture.

BB0749 recombinant protein purification and antisera production

Recombinant protein was generated as previously described in Van Gundyet al. [50]. All animal studies were conducted using the Guide for the Care and Use of Laboratory Animals (Institute of Animal Research, National Research Council) with protocols approved by the Virginia Commonwealth University IACUC, permit number AD10000387. An eight-week-old female Sprague-DawleyRattus norvegicus, rat (Charles River) was immunized on a three-dose schedule, with each dose two weeks apart after an initial seven-day acclimatization period as previously described [81]. The rat was primed by IP vaccination with 50 µg antigen, BB0749 recombinant protein, in Freund’s complete adjuvant (Sigma-Aldrich). Two boosters of 25 µg antigen, BB0749 recombinant protein, each in Freund’s incomplete adjuvant were also administered IP (Sigma-Aldrich). One week following the final boost, the rat was euthanized by CO2asphyxiation and terminal bleeds conducted by cardiac puncture. Serum was harvested by centrifugation in VACUETTE Z serum sep-clot activator tubes (Greiner Bio-One) according to manufacturer instructions.

Recombinant protein expression and FPLC purification

Plasmids for recombinant protein production were constructed using GenScript Plasmid Synthesis service. Open reading frames were codon optimized forE. coli. The codon optimized open reading frames and final plasmid sequences used can be found in Table S3. Plasmids were purified using PureYield™ Midiprep plasmid system (Promega) per manufacturer’s protocol. Plasmids were transformed intoE. coliBL21(DE3) cells (Lucigen™) and transformants were screened by PCR using pETite T7 Forward Primer and pETite Reverse Primer (50 µM). Positive transformants were grown in LB containing kanamycin (50 μg/mL) overnight. The overnight culture was inoculated 1:100 into LB supplemented with kanamycin (50 μg/mL) and grown to mid-logarithmic phase (OD600 = 0.4–0.6). Protein expression was induced by the addition of isopropyl-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM and incubated with shaking for 16–18 hours at 20°C. Cultures were centrifuged at 5000 × g for 10 min. The cell pellets were resuspended in buffer A (20 mM NaH2PO4, 500 mM NaCl, 20 mM imidazole, pH 7.4) containing 1 mM MgCl2, Universal Nuclease for Cell Lysis (10 U/ml, Pierce™ Thermo-Fisher Scientific™), and 0.2 mg/mL lysozyme and incubated for 30 min and vortexed every 10 min. The lysate was centrifuged at 12,000 ×gfor 30 min at 4°C to separate the soluble and insoluble fractions. The soluble fractions were passed through a 5 μM filter (Millipore) and proteins were purified using nickel affinity chromatography by a Fast Protein Liquid Chromatography platform (ÄKTA Pure 25 M; Cytiva) with a 1-mL HisTrap FF column (Cytiva) at a flow rate of 1 mL/min. Columns were equilibrated using five column volumes of buffer A. Samples were loaded mechanically, followed by washing with 20 column volumes of buffer A. Proteins were eluted in 1 mL increments with five column volumes of elution buffer (20 mM NaH2PO4, 500 mM NaCl, 500 mM imidazole). Fractions were combined and dialyzed in G3 Slide-a-lyzer™ cassettes (Thermo-Fisher Scientific™, 3.5 MWCO, 3 ml cassette) in 1× PBS (phosphate-buffered saline, pH 7.4). Dilute proteins were concentrated with Pierce Concentrators, PES, 3K MWCO, 0.5 mL (Thermo-Fisher Scientific™). Proteins were quantified by bicinchoninic acid (BCA) assay (Thermo-Fisher Scientific™), resolved on precast Criterion Any kD™ TGX Stain-Free™ Protein Gels (BioRad) and visualized by silver staining to assess protein purity (Figure S7). Notably, StpA, GrpE, BosR, RRF, KhpA, SSB, and FlgV appear to be nearly homogeneous but do show minimal signal from additional protein species of differing molecular masses, which may represent oligomeric forms or degradation products of the proteins or minor contaminants.

RNAs

RNAs were ordered from Dharmacon or IDT, dissolved in water and stored at −20°C. Sequences are listed in Table S3.

In-solution RNA annealing and strand displacement assay

The combined RNA annealing and strand displacement assay was adapted from the Schroeder laboratory but utilizes fluorophores and quenchers instead of fluorescence resonance energy transfer (FRET). Two previously published 21-nucleotide complementary RNAs were used with 5ʹ and 3ʹ modifications [22,25,50,57–61]. M1 RNA included a 5ʹ 6-FAM fluorophore (M1_5FAM6), J1 RNA included a 3ʹ Dabcyl quencher (J1_3Dab), and the unlabeled competitor uJ1 contained no modifications (Table S3). Combined RNA annealing and strand displacement assays were carried out in 1× FRET buffer (50 mM Tris-HCl pH 7.5, 1 mM MnCl2, 10 mM NaCl and 1 mM DTT) at 37°C unless otherwise stated. Notably the 1× FRET buffer contains manganese instead of magnesium. Manganese was reported to be important forB. burgdorferisurvival and proper function of RNAPin vitro[82–84]. In-gel annealing and strand displacement assays in 1× FRET buffer with 1 mM MnCl2or 1 mM MgCl2were performed, and the rate of strand displacement was faster with Mn instead of Mg, while annealing kinetics were similar between both salts (Figure S8). Final concentrations of M1_5FAM6 and J1_3Dab RNA in each reaction were 25 nM and the unlabeled competitor M1 RNA (uM1) was added to a final concentration of 250 nM. Final concentrations of protein in the reaction were 1 µM. Proteins were stored in 1× PBS, therefore an equal volume of 1× PBS was added to all reactions, including the negative control RNA only reaction. 100 µl combined RNA annealing and strand displacement reactions were carried out in Nunc™ Microwell™ black flat-bottom 96-well plates (Thermo-Fisher Scientific™) and read on a Tecan Spark. The annealing reaction was started by injecting J1_3Dab into wells containing M1_5FAM6 in 1× FRET buffer with or without protein. The reaction was shaken for 3 s and the samples were excited at 485 nm (20 bandwidth), and emission intensity was measured at 535 nm (25 nm bandwidth) every 10 s for 180 s. The stand displacement assay was then initiated by injecting tenfold molar excess of unlabeled M1 (uM1). The reaction was shaken for 3 s and readings were taken every 10 s for 180 s. At least three reactions were performed for each protein.

The annealing reaction final time point was constrained to zero by subtracting the fluorescent units at the final time point from itself. Accordingly, all timepoints’ fluorescent units were normalized by subtracting the zero time point fluorescent units from them. In addition, the initiation of the annealing reaction was constrained to one by dividing the fluorescent units of the timepoint before injection of J1 by itself, and all timepoints’ fluorescence units were normalized by dividing them by the initiation time point fluorescent units. These normalizations constrain the data from 0 to 1 (Figure S3), which was used in the GraphPad Prism result analyses. Annealing rates were measured using one-phase decay and strand displacement rates were measured using one phase association.

Gel-based RNA annealing and strand displacement assays

The RNA gel annealing and strand displacement assay protocols were adapted from the Schroeder laboratory and performed as previously described [50]. The RNA annealing assay utilizes two short unstructured complementary 21-mer RNAs, termed J1 and M1, that are labeled with Cy5 and Cy3, respectively, and are found in Table S3. All RNAs were synthesized by IDT. RNA annealing reactions were performed in 1× MnCl2FRET buffer (50 mM Tris – HCl, pH 7.5, 1 mM MnCl2, 10 mM NaCl, 1 mM DTT) at 37°C. Two controls were included with each annealing assay. Control 1 (C1) is single-stranded M1 RNA (10 nM M1, 1× MnCl2FRET Buffer) and control 2 (C2) is preformed J1M1 dsRNA (10 nM M1, 10 nM J1, 1× MnCl2FRET Buffer). C2 dsRNA was formed by heating the two RNAs (J1 and M1) at 95°C for 2 min and then slowly cooling to room temperature. A master mix for the annealing reaction was assembled (10 nM J1 RNA in 1× MnCl2FRET buffer) on ice and moved to 37°C to equilibrate. A time point zero aliquot (10 μL) was taken from the master mix, added to M1 RNA and 2.5 μL of 5× stop/load buffer (150 mM EDTA, 2% SDS, 15% (w/v) sucrose, 12.5% (w/v) Ficoll 400, 0.2 mg/mL yeast tRNA and 3 μM of unlabeled M1 (uM1)) and immediately loaded into a running (100 V) 20% native polyacrylamide gel (20% TBE, 12 well, Novex precast gel). M1 RNA was added to the RNA annealing master mix to start the annealing reaction. Time points were withdrawn from the annealing reaction, stopped with 5× stop/load buffer, and loaded on the running gel. To track the progression of the running gel, a lane was loaded with 5× stop/load buffer with the addition of Gel loading buffer II (Invitrogen). Proteins were added to reactions for a final concentration of 1.2 μM. Gels were run at 100 V at 4°C for 3 h. RNA was visualized using Cy3 and Cy5 channels on an Azure Sapphire using the SmartScan feature, and overlays for each channel were used for gel images. Quantification of band intensity was measured using the rectangle and gel plot function in ImageJ. The rate of dsRNA production was quantified by plotting the amount of dsRNA (J1M1) divided by the total RNA (J1M1, J1uM1) over time, which was normalized to the zero time point and graphed using GraphPad Prism with linear regression analysis. The amount of RNA was quantified using the Cy5 channel. Differences between the assay endpoints (at 15 min) for each annealing reaction were determined using a one-way ANOVA and Tukey’s multiple comparisons (GraphPad Prism).

The RNA gel strand displacement assay utilizes M1, J1, and uM1 RNAs, as described above. RNA strand displacement reactions were performed in 1× MnCl2FRET buffer (50 mM Tris – HCl, pH 7.5, 1 mM MnCl2, 10 mM NaCl, 1 mM DTT) at 37°C. Two controls were included with each strand displacement assay. Control 1 (C1) is purified J1M1 dsRNA (30 nM purified M1J1, 1× MnCl2FRET Buffer) and control 2 (C2) is J1uM1 dsRNA (30 nM purified M1J1, 0.3 μM uM1, 1× FRET MnCl2Buffer). C2 dsRNA J1uM1 was formed by heating purified dsRNA J1M1 and uM1 at 95°C for 2 min and slowly cooling to room temperature. Strand displacement master mix was prepared with 30 nM purified dsRNA in 1× MnCl2FRET buffer. A zero time point was removed from the master mix, added to 0.3 μM uM1 and 2.5 μL of 5× stop/load buffer, and immediately loaded into a running (100 V) 20% native polyacrylamide gel (20% TBE, 12 well, Novex precast gel). Strand displacement reactions were started by adding uM1 RNA to the strand displacement master mix at 37°C. Time points were withdrawn from the reaction and stopped with 2.5 μL of 5× stop/load buffer. Gels were run at 100 V at 4°C for 3 h. Gels were visualized using Cy3 and Cy5 channels on an Azure Sapphire using the SmartScan feature and overlayed for gel images. Quantification of band intensity was measured using the Cy5 channel with the rectangle and gel plot function in ImageJ. The amount of strand displacement in each lane was quantified by plotting the amount of J1M1 (Cy5) dsRNA divided by the total RNA (J1M1, J1uM1, and M1; Cy5), which was normalized to the amount of dsRNA (J1M1; Cy5) at time point zero and graphed using GraphPad Prism with second-order polynomial regression analysis. Differences between the assay endpoints (at 15 min) for each strand displacement reaction were determined using one-way ANOVA and Tukey’s multiple comparisons (GraphPad Prism).

Anti-termination assay inE.coli

Transcription anti-termination assays were performed as described [50].E. coliRL211 strain containing thecatgene preceded by thetrpLterminator stem-loop were transformed with plasmids constructed for purifying recombinant proteins (described above and in Table S3). Cells were grown overnight in LB and, when appropriate, kanamycin (50 µg/mL) was added. Overnight cultures were spotted (5 µL) on LB and LB with chloramphenicol (34 µg/mL) plates and grown at 37°C. Cell growth was inspected daily. Plates were imaged using the Azure Sapphire biomolecular imager using the white light channel. Experiments were repeated in triplicate.

In vitroRNA unwinding assay

The truncated Rho-independenttrpLterminator fluorescent reporter RNA contained a 5ʹ 6-FAM fluorophore and a 3ʹ Dabcyl quencher (trpLterm). A 100 nM working stock oftrpLterm in 1× FRET buffer was heated to 65°C for 10 min, slow cooled to 37°C and stored on ice. Reactions contained 50 nMtrpLterm and 1 µM protein (final concentrations). Proteins were stored in 1× PBS, therefore an equal volume of 1× PBS was added to all reactions, including the negative control RNA only reaction. 100 µL unwinding reactions were carried out in Nunc™ Microwell™ black flat-bottom 96 well plates (Thermo-Fisher Scientific™) buffer at 24°C using the Tecan Spark. The unwinding reaction was started by injecting thetrpLterm RNA into wells containing 1× FRET buffer (50 mM Tris-HCl, pH 7.5, 1 mM MnCl2, 10 mM NaCl and 1 mM DTT) with or without protein and incubated for 10 min. The samples were excited at 485 nm (20 bandwidth), and emission intensity was measured at 535 nm (25 nm bandwidth). At least three reactions were performed for each protein. Data were background normalized by removing the background fluorescence signal (RNA without protein) from all samples. Dunnett’s Multiple Comparison Two Way ANOVA was used to compare each protein against RNA only (ns,p< .050,,p< .005,**,p< .0001,***).

Bb0749:3×flagstrain construction

Low-passage virulentB. burgdorferistrain B31-5A4 genomic DNA was used as template for amplifying thebb0749upstream region andbb0749ORF using Phusion™ High Fidelity DNA polymerase and region-specific primers (Table S3). The PCR products were cloned into pGEM®-T Easy (Promega) per manufacturer’s instructions and transformed into DH5αE. colicells (Thermo-Fisher Scientific™). A gBlock™ was synthesized by IDT that contained the streptomycin/spectinomycin resistance cassette (flgBp-aadA), thebb0749predicted promoter and 5’ UTR, and theB. burgdorferioptimized3×flagsequence (Table S3). The gBlock DNA was used as template to amplify the sequence using Phusion™ High Fidelity DNA polymerase and gBlock specific primers (Table S3). The PCR product was cloned into pGEM®-T Easy (Promega) per manufacturer’s instructions. Positive transformants were validated by sequencing. Thebb0749ORF region 5’ end was engineered with an XmaI and HindIII site and the 3’ end was engineered with an AscI site. The gBlock 5’ end was engineered with an XmaI site and the 3’ end was engineered with a HindIII site. The NdeI site found between theflgBpromoter sequence and theaadAORF of the streptomycin/spectinomycin antibiotic resistance cassette was mutated and an NdeI site was added between the 3’ end ofbb07495’ UTR and the3×flagsequence. The gBlock was inserted into thebb0749ORF construct with XmaI and HindIII yielding the gBlock:ORF construct. Positive transformants were sequence validated. Thebb0749upstream region 3’ end was engineered with an XmaI and AscI site, which was used to insert the gBlock:bb0749ORF into the upstream region, yieldingbb0749up:gBlock:ORF. Positive transformants were validated by sequencing. The final plasmid, pGEM®-T Easybb0749up:gBlock:ORF, was linearized with AhdI. Transformation ofB. burgdorferiwith the linearized plasmid was performed as previously described [79]. Positive transformants were screened with primers flanking the insertion site on the chromosome (Table S3). Thebb0749:3×flagstrain was validated by sequencing and verified by Western blot for expression of BB0749-3×FLAG.

Crosslinking immunoprecipitation ofbb0749:3×flag

B. burgdorferi bb0749:3×flagstrain was cultivated in BSK-II media at 37°C. Cells were grown to 5 × 107cells mL−1, split into two 500-mL cultures and collected by centrifugation. One pellet was treated as the non-crosslinked control and was washed in 2 ml of dPBS, spun at 18,000 ×gfor 2 min, the supernatant was discarded, and the cell pellets were placed at −80°C. The crosslinked sample pellet was resuspended in 100 mL of dPBS and 20 mL of the resuspended cells were crosslinked at a time. 20 mL of resuspended cells were placed into a glass baking dish and UV crosslinked twice with 80,000 μJ/cm2in a Spectrolinker™XL-1000 UV Crosslinker (Spectroline™). The crosslinked samples were combined and centrifuged at 5000 ×gfor 10 min, the supernatant was discarded, the cell pellet was resuspended in 2 mL of dPBS and centrifuged at 18,000 ×gfor 2 min, the supernatant was discarded. The crosslinked pellet was stored at −80°C. Non-crosslinked and crosslinked pellets were thawed on ice and resuspended in 1 mL of ice-cold lysis buffer (50 mM Tris-HCl, pH 7.5, 5 mM MgCl2, 250 mM NaCl, complete protease inhibitor mini tablet, 10 U/µL SupeRNasin.). Cells were placed in 5-mL steel cryo-vials (Retsch), submerged in LN2and subsequently lysed at 30 Hz for 10 min, submerging samples in LN2every 3 min, in the Retsch MM400. The cell lysate was centrifuged at 18,000 ×gfor 30 min at 4C° and the cleared lysate transferred to a new tube. Input samples were taken for Western blot analysis. Cleared lysate was added to 15 µL of anti-FLAG M2 magnetic beads (Sigma-Aldrich) that were equilibrated with lysis buffer. Samples rotated overnight at 4°C and placed on a magnetic stand. The supernatant from each sample was collected for Western blot analysis. Samples were washed three times with 500 µL TBS buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl). The beads were resuspended in 220 µL and 20 µL were collected for Western blot analyses. The sample was placed on the magnetic stand and the supernatant was discarded. The beads were resuspended in Zymo Research Quick-RNA Miniprep Lysis Buffer (Cat #: R1060-1–50) and incubated at RT for 10 min. The samples were placed on the magnetic stand and the supernatant was collected. The RNA was purified using the Zymo Research Quick-RNA™MiniPrep Plus (Zymo Research #R1057) per manufacturer’s instructions. The protocol for the RNA miniprep kit was repeated twice and ethanol was added before the Spin-Away™ Filter to increase the binding of DNA and large RNAs. Isolated RNA samples were run on a Bioanalyzer 2100 Pico chip and protein was visualized via Western blot.

Statistical analyses

In vitroRNA unwinding assay relative fluorescent units of unnormalized data were analyzed by a two-way ANOVA with multiple comparisons (p-value ≥ 0.050, ns,p-value, ≤0.050, *,p-value ≤ 0.0001, ****). Gel annealing endpoint signals were quantified by ImageJ and analyzed by a two-way ANOVA with multiple comparisons (p-value ≤ 0.050, *,p-value ≤ 0.010, **,p-value ≤ 0.0001, ****). Gel strand displacement endpoint signals were quantified by Image J and analyzed by a two-way ANOVA with multiple comparisons (p-value ≤ 0.001, ***,p-value ≤ 0.0001, ****). StpA RNA annealing and strand displacement endpoint signals in the presence of different salts were quantified by ImageJ and analyzed by a Welch’s t-Test (p-value > 0.050, ns).

Acknowledgments

We thank Bob Gilmore, Kevin Brandt, Shelby Lennon, Joseph Cardiello, and Marisa Foster for thoughtful and critical reading of the manuscript. We thank Jeremy Bono and James Kovaks for useful discussions. We thank Robert Landick for the RL211 strain. The findings and conclusions in this report are those of the authors and do not represent the official position of the Centers for Disease Control and Prevention.

Funding Statement

This work was funded by Public Health Service Grants [AI123672] (to ML, JH and JTS), [U01 AI169840-01A1] (to RTM, DSS and C Davies), and [AI170940] (to JH and JTS).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the Proteomics Identifications Database partner repository with the dataset identifier PXD068462 and 10.6019/PXD068462. Any further underlying data will be made available upon request.

Ethical approval

This article does not contain any studies with human participants. All animal studies were conducted using the Guide for the Care and Use of Laboratory Animals (Institute of Animal Research, National Research Council) with protocols approved by the Virginia Commonwealth University IACUC, permit number AD10000387.

Supplementary material

Supplemental data for this article can be accessed online athttps://doi.org/10.1080/15476286.2026.2684395

Associated Data

Data Availability Statement

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the Proteomics Identifications Database partner repository with the dataset identifier PXD068462 and 10.6019/PXD068462. Any further underlying data will be made available upon request.

References

1. Babitzke P, Lai Y-J, Renda AJ, et al. Posttranscription initiation control of gene expression mediated by bacterial RNA-binding proteins. Annu Rev Microbiol. 2019;73(1):43–67. doi.org/10.1146/annurev-micro-020518-115907
2. Ng Kwan Lim E, Sasseville C, Carrier M-C, et al. Keeping up with RNA-based regulation in bacteria: new roles for RNA binding proteins. Trends Genet. 2021;37(1):86–97. doi.org/10.1016/j.tig.2020.09.014
3. Holmqvist E, Vogel J.. RNA-binding proteins in bacteria. Nat Rev Microbiol. 2018;16(10):601–615. doi.org/10.1038/s41579-018-0049-5
4. Corley M, Burns MC, Yeo GW. How RNA-binding proteins interact with RNA: molecules and mechanisms. Mol Cell. 2020;78(1):9–29. doi.org/10.1016/j.molcel.2020.03.011
5. Hentze MW, Castello A, Schwarzl T, et al. A brave new world of RNA-binding proteins. Nat Rev Mol Cell Biol. 2018;19(5):327–341. doi.org/10.1038/nrm.2017.130
6. Castello A, Fischer B, Eichelbaum K, et al. Insights into RNA biology from an atlas of mammalian mRNA-binding proteins. Cell. 2012;149(6):1393–1406. doi.org/10.1016/j.cell.2012.04.031
7. Smith JM, Sandow JJ, Webb AI. The search for RNA-binding proteins: a technical and interdisciplinary challenge. Biochem Soc Trans. 2021;49(1):393–403. doi.org/10.1042/BST20200688
8. Shchepachev V, Bresson S, Spanos C, et al. Defining the RNA interactome by total RNA-associated protein purification. Mol Syst Biol. 2019;15(4):e8689. doi.org/10.15252/msb.20188689
9. Asencio C, Chatterjee A, Hentze MW. Silica-based solid-phase extraction of cross-linked nucleic acid-bound proteins. Life Sci Alliance. 2018;1(3):e201800088. doi.org/10.26508/lsa.201800088
10. Chu LC, Arede P, Li W, et al. The RNA-bound proteome of MRSA reveals post-transcriptional roles for helix-turn-helix DNA-binding and Rossmann-fold proteins. Nat Commun. 2022;13(1):2883. doi.org/10.1038/s41467-022-30553-8
11. Smirnov A, Förstner KU, Holmqvist E, et al. Grad-seq guides the discovery of ProQ as a major small RNA-binding protein. Proc Natl Acad Sci USA. 2016;113(41):11591–11596. doi.org/10.1073/pnas.1609981113
12. Hor J, Garriss G, Di Giorgio S, et al. Grad-seq in a Gram-positive bacterium reveals exonucleolytic sRNA activation in competence control. Embo J. 2020;39(9):e103852. doi.org/10.15252/embj.2019103852
13. Hor J, Di Giorgio S, Gerovac M, et al. Grad-seq shines light on unrecognized RNA and protein complexes in the model bacteriumEscherichia coli. Nucleic Acids Res. 2020;48(16):9301–9319. doi.org/10.1093/nar/gkaa676
14. Gerovac M. A Grad-seq view of RNA and protein complexes inPseudomonas aeruginosaunder standard and bacteriophage predation conditions. MBio. 2021:12(1):10–1128. doi.org/10.1128/mBio.03454-20
15. Gerovac M, El Mouali Y, Kuper J, et al. Global discovery of bacterial RNA-binding proteins by RNase-sensitive gradient profiles reports a new FinO domain protein. RNA. 2020;26(10):1448–1463. doi.org/10.1261/rna.076992.120
16. Chihara K, Gerovac M, Hörb J, et al. Global profiling of the RNA and protein complexes ofEscherichia coliby size exclusion chromatography followed by RNA sequencing and mass spectrometry (SEC-seq). RNA. 2022;29(1):123–139. doi.org/10.1261/rna.079439.122
17. Gerovac M, Vogel J, Smirnov A. The world of stable ribonucleoproteins and its mapping with Grad-seq and related approaches. Front Mol Biosci. 2021;8:661448. doi.org/10.3389/fmolb.2021.661448
18. Katsuya-Gaviria K, Paris G, Dendooven T, et al. Bacterial RNA chaperones and chaperone-like riboregulators: behind the scenes of RNA-mediated regulation of cellular metabolism. RNA Biol. 2022;19(1):419–436. doi.org/10.1080/15476286.2022.2048565
19. Woodson SA, Panja S, Santiago-Frangos A. Proteins that chaperone RNA regulation. Microbiol Spectr. 2018;6(4). doi.org/10.1128/microbiolspec.RWR-0026-2018/
20. Rajkowitsch L, Chen D, Stampfl S, et al. RNA chaperones, RNA annealers and RNA helicases. RNA Biol. 2007;4(3):118–130. doi.org/10.4161/rna.4.3.5445
21. Semrad K. Proteins with RNA chaperone activity: a world of diverse proteins with a common task-impediment of RNA misfolding. Biochem Res Int. 2011;2011:532908. doi.org/10.1155/2011/532908
22. Rajkowitsch L, Semrad K, Mayer O, et al. Assays for the RNA chaperone activity of proteins. Biochem Soc Trans. 2005;33(Pt 3):450–456. doi.org/10.1042/BST0330450
23. Woodson SA, Panja S, Santiago-Frangos A. Proteins that chaperone RNA regulation. Microbiol Spectr. 2018;6(4):RWR-0026–2018. doi.org/10.1128/microbiolspec.RWR-0026-2018
24. Woodson SA. Taming free energy landscapes with RNA chaperones. RNA Biol. 2010;7(6):677–686. doi.org/10.4161/rna.7.6.13615
25. Rajkowitsch L, Schroeder R. Dissecting RNA chaperone activity. RNA. 2007;13(12):2053–2060. doi.org/10.1261/rna.671807
26. Doetsch M, Schroeder R, Furtig B. Transient RNA-protein interactions in RNA folding. FEBS J. 2011;278(10):1634–1642. doi.org/10.1111/j.1742-4658.2011.08094.x
27. Duss O, Stepanyuk GA, Puglisi JD, et al. Transient protein-RNA interactions guide nascent ribosomal RNA folding. Cell. 2019;179(6):1357–1369.e16. doi.org/10.1016/j.cell.2019.10.035
28. Kugeler KJ, Schwartz AM, Delorey MJ, et al. Estimating the frequency of Lyme disease diagnoses, United States, 2010–2018. Emerg Infect Dis. 2021;27(2):616–619. doi.org/10.3201/eid2702.202731
29. Samuels DS, Lybecker MC, Yang XF, et al. Gene regulation and transcriptomics. Curr Issues Mol Biol. 2021;42(1):223–266. doi.org/10.21775/cimb.042.223
30. Sapiro AL, Hayes BM, Volk RF, et al. Longitudinal map of transcriptome changes in the Lyme pathogenBorrelia burgdorferiduring tick-borne transmission. eLife. 2023;12:RP86636. doi.org/10.7554/eLife.86636
31. Grassmann AA, Tokarz R, Golino C, et al. BosR and PlzA reciprocally regulate RpoS function to sustainBorrelia burgdorferiin ticks and mammals. J Clin Invest. 2023;133(5). doi.org/10.1172/JCI166710
32. Lybecker M, Bilusic I, Raghavan R. Pervasive transcription: detecting functional RNAs in bacteria. Transcription. 2014;5(4):e944039. doi.org/10.4161/21541272.2014.944039
33. Lybecker MC, Samuels DS. Small RNAs ofBorrelia burgdorferi: characterizing functional regulators in a sea of sRNAs. Yale J Biol Med. 2017;90(2):317–323.https://pmc.ncbi.nlm.nih.gov/articles/pmid/28656017/.
34. Drecktrah D, Hall LS, Rescheneder P, et al. The stringent response-regulated sRNA transcriptome ofBorrelia burgdorferi. Front Cell Infect Microbiol. 2018;8:231. doi.org/10.3389/fcimb.2018.00231
35. Popitsch N, Bilusic I, Rescheneder P, et al. Temperature-dependent sRNA transcriptome of the Lyme disease spirochete. BMC Genomics. 2017;18(1):28. doi.org/10.1186/s12864-016-3398-3
36. Arnold WK, Savage CR, Brissette CA, et al. RNA-seq ofBorrelia burgdorferiin multiple phases of growth reveals insights into the dynamics of gene expression, transcriptome architecture, and noncoding RNAs. PLOS ONE. 2016;11(10):e0164165. doi.org/10.1371/journal.pone.0164165
37. Petroni E, Esnault C, Tetreault D, et al. Extensive diversity in RNA termination and regulation revealed by transcriptome mapping for the Lyme pathogenBorrelia burgdorferi. Nat Commun. 2023;14(1):3931. doi.org/10.1038/s41467-023-39576-1
38. Medina-Perez DN, Wager B, Troy E, et al. The intergenic small non-coding RNAittAis required for optimal infectivity and tissue tropism inBorrelia burgdorferi. PLoS Pathog. 2020;16(5):e1008423. doi.org/10.1371/journal.ppat.1008423
39. Lybecker MC, Samuels DS. Temperature-induced regulation of RpoS by a small RNA inBorrelia burgdorferi. Mol Microbiol. 2007;64(4):1075–1089. doi.org/10.1111/j.1365-2958.2007.05716.x
40. Shapiro BL, Jaiswal P, Funkhouser-Jones LJ, et al. TheBorreliella burgdorferiBosR-associated small non-coding RNA BasA regulates virulence. Infect Immun. 2026;94(4):e0002626. doi.org/10.1128/iai.00026-26
41. Savage CR, Jutras BL, Bestor A, et al.Borrelia burgdorferiSpoVG DNA- and RNA-binding protein modulates the physiology of the Lyme disease spirochete. J Bacteriol. 2018;200(12). doi.org/10.1128/JB.00033-18
42. Jutras BL, Jones GS, Verma A, et al. Posttranscriptional self-regulation by the Lyme disease bacterium’s BpuR DNA/RNA-binding protein. J Bacteriol. 2013;195(21):4915–4923. doi.org/10.1128/JB.00819-13
43. Saylor TC, Savage CR, Krusenstjerna AC, et al. Quantitative analyses of interactions between SpoVG and RNA/DNA. Biochem Biophys Res Commun. 2023;654:40–46. doi.org/10.1016/j.bbrc.2023.02.044
44. Raghunandanan S, Priya R, Alanazi F, et al. A fur family protein BosR is a novel RNA-binding protein that controlsrpoSRNA stability in the Lyme disease pathogen. Nucleic Acids Res. 2024;52(9):5320–5335. doi.org/10.1093/nar/gkae114
45. Jutras BL, Chenail AM, Carroll DW, et al. Bpur, the Lyme disease spirochete’s PUR domain protein: identification as a transcriptional modulator and characterization of nucleic acid interactions. J Biol Chem. 2013;288(36):26220–26234. doi.org/10.1074/jbc.M113.491357
46. Krusenstjerna AC, Saylor TC, Arnold WK, et al.Borrelia burgdorferiDnaA and the nucleoid-associated protein EbfC coordinate expression of thednaX-ebfCoperon. J Bacteriol. 2023;205(1):e0039622. doi.org/10.1128/jb.00396-22
47. Sze CW, Morado DR, Liu J, et al. Carbon storage regulator A (CsrABb) is a repressor ofBorrelia burgdorferiflagellin protein FlaB. Mol Microbiol. 2011;82(4):851–864. doi.org/10.1111/j.1365-2958.2011.07853.x
48. Zamba-Campero M, Soliman D, Yu H, et al. Broadly conserved FlgV controls flagellar assembly andBorrelia burgdorferidissemination in mice. Nat Commun. 2024;15(1):10417. doi.org/10.1038/s41467-024-54806-w
49. Lybecker MC, Abel CA, Feig AL, et al. Identification and function of the RNA chaperone Hfq in the Lyme disease spirocheteBorrelia burgdorferi. Mol Microbiol. 2010;78(3):622–635. doi.org/10.1111/j.1365-2958.2010.07374.x
50. Van Gundy T, Patel D, Bowler BE, et al. c-di-GMP regulates activity of the PlzA RNA chaperone from the Lyme disease spirochete. Mol Microbiol. 2023;119(6):711–727. doi.org/10.1111/mmi.15066
51. Samuels DS, Lybecker MC. The BosR is back! Mol Microbiol. 2026;125(2):81–84. doi.org/10.1111/mmi.70046
52. UniProt C, Martin M-J, Orchard S. Uniprot: the universal protein knowledgebase in 2025. Nucleic Acids Res. 2025;53(D1):D609–D617. doi.org/10.1093/nar/gkae1010
53. Caly DL, O’Toole PW, Moore SA. The 2.2-Å structure of the HP0958 protein fromHelicobacter pylorireveals a kinked anti-parallel coiled-coil hairpin domain and a highly conserved Zn-ribbon domain. J Mol Biol. 2010;403(3):405–419. doi.org/10.1016/j.jmb.2010.08.051
54. Xu H, Sultan S, Yerke A, et al.Borrelia burgdorferiCheY2 is dispensable for chemotaxis or motility but crucial for the infectious life cycle of the spirochete. Infect Immun. 2017;85(1). doi.org/10.1128/IAI.00264-16
55. Motaleb MA, Sultan SZ, Miller MR, et al. CheY3 ofBorrelia burgdorferiis the key response regulator essential for chemotaxis and forms a long-lived phosphorylated intermediate. J Bacteriol. 2011;193(13):3332–3341. doi.org/10.1128/JB.00362-11
56. Sahoo BR, Bardwell JC.Protein and RNA chaperones. Mol Aspects Med. Mol Aspects Med. 2025;104:101384. doi.org/10.1016/j.mam.2025.101384
57. Stampfl S, Doetsch M, Beich-Frandsen M, et al. Characterization of the kinetics of RNA annealing and strand displacement activities of theE.coliDEAD-box helicase CsdA. RNA Biol. 2013;10(1):149–156. doi.org/10.4161/rna.23475
58. Rajkowitsch L, Schroeder R. Coupling RNA annealing and strand displacement: a FRET-based microplate reader assay for RNA chaperone activity. Biotechniques. 2007;43(3):304, 306, 308. doi.org/10.2144/000112530
59. Doetsch M, Gstrein T, Schroeder R, et al. Mechanisms of StpA-mediated RNA remodeling. RNA Biol. 2010;7(6):735–743. doi.org/10.4161/rna.7.6.13882
60. Doetsch M, Fürtig B, Gstrein T, et al. The RNA annealing mechanism of the HIV-1 Tat peptide: conversion of the RNA into an annealing-competent conformation. Nucleic Acids Res. 2011;39(10):4405–4418. doi.org/10.1093/nar/gkq1339
61. Doetsch M, Stampfl S, Fürtig B, et al. Study ofE.coliHfq’s RNA annealing acceleration and duplex destabilization activities using substrates with different GC-contents. Nucleic Acids Res. 2013;41(1):487–497. doi.org/10.1093/nar/gks942
62. Hohmann KF, Blümler A, Heckel A, et al. The RNA chaperone StpA enables fast RNA refolding by destabilization of mutually exclusive base pairs within competing secondary structure elements. Nucleic Acids Res. 2021;49(19):11337–11349. doi.org/10.1093/nar/gkab876
63. Moreira BG, You Y, Owczarzy R. Cy3 and Cy5 dyes attached to oligonucleotide terminus stabilize DNA duplexes: predictive thermodynamic model. Biophys Chem. 2015;198:36–44. doi.org/10.1016/j.bpc.2015.01.001
64. Moreira BG, You Y, Behlke MA, et al. Effects of fluorescent dyes, quenchers, and dangling ends on DNA duplex stability. Biochem Biophys Res Commun. 2005;327(2):473–484. doi.org/10.1016/j.bbrc.2004.12.035
65. Landick R, Stewart J, Lee DN. Amino acid changes in conserved regions of the β-subunit ofEscherichia coliRNA polymerase alter transcription pausing and termination. Genes Dev. 1990;4(9):1623–1636. doi.org/10.1101/gad.4.9.1623
66. Phadtare S, Severinov K. RNA remodeling and gene regulation by cold shock proteins. RNA Biol. 2010;7(6):788–795. doi.org/10.4161/rna.7.6.13482
67. Li J, Zhang B, Zhou L, et al. The archaeal RNA chaperone TRAM0076 shapes the transcriptome and optimizes the growth of Methanococcus maripaludis. PLoS Genet. 2019;15(8):e1008328. doi.org/10.1371/journal.pgen.1008328
68. Nakaminami K, Karlson DT, Imai R. Functional conservation of cold shock domains in bacteria and higher plants. Proc Natl Acad Sci U S A. 2006;103(26):10122–10127. doi.org/10.1073/pnas.0603168103
69. Ouyang Z, Deka RK, Norgard MV. BosR (BB0647) controls the RpoN-RpoS regulatory pathway and virulence expression inBorrelia burgdorferiby a novel DNA-binding mechanism. PLoS Pathog. 2011;7(2):e1001272. doi.org/10.1371/journal.ppat.1001272
70. Ouyang Z, Ouyang Z, Norgard MV, et al. Identification of a core sequence for the binding of BosR to therpoSpromoter region inBorrelia burgdorferi. Microbiology. 2015;161(Pt 4):931. doi.org/10.1099/mic.0.000060
71. Mason C, Liu X, Prabhudeva S, et al. The CXXC motifs are essential for the function of BosR inBorrelia burgdorferi. Front Cell Infect Microbiol. 2019;9:109. doi.org/10.3389/fcimb.2019.00109
72. Modic M, Adamek M, Ule J. The impact of IDR phosphorylation on the RNA binding profiles of proteins. Trends Genet. 2024;40(7):580–586. doi.org/10.1016/j.tig.2024.04.004
73. Lovci MT, Bengtson MH, Massirer KB. Post-translational modifications and RNA-binding proteins. Adv Exp Med Biol. 2016;907:297–317. doi.org/10.1007/978-3-319-29073-7_12
74. Priyanka G, Raj EJ, Prabhu NP. Liquid-liquid phase separation of intrinsically disordered proteins: effect of osmolytes and crowders. Prog Mol Biol Transl Sci. 2025;211:249–269. doi.org/10.1016/bs.pmbts.2024.11.005
75. Ottoz DSM, Berchowitz LE. The role of disorder in RNA binding affinity and specificity. Open Biol. 2020;10(12):200328. doi.org/10.1098/rsob.200328
76. Wheeler EC, Van Nostrand EL, Yeo GW. Advances and challenges in the detection of transcriptome-wide protein-RNA interactions. Wiley Interdiscip Rev RNA. 2018;9(1):e1436. doi.org/10.1002/wrna.1436
77. Beckmann BM, Horos R, Fischer B, et al. The RNA-binding proteomes from yeast to man harbour conserved enigmRBPs. Nat Commun. 2015;6(1):10127. doi.org/10.1038/ncomms10127
78. Castello A, Fischer B, Frese C, et al. Comprehensive identification of RNA-binding domains in human cells. Mol Cell. 2016;63(4):696–710. doi.org/10.1016/j.molcel.2016.06.029
79. Samuels DS. Electrotransformation of the spirocheteBorrelia burgdorferi. Methods Mol Biol. 1995;47:253–259. doi.org/10.1385/0-89603-310-4:253
80. Samuels DS, Drecktrah D, Hall LS. Genetic transformation and complementation. Methods Mol Biol. 2018;1690:183–200. doi.org/10.1007/978-1-4939-7383-5_15
81. Izac JR, O’Bier NS, Oliver LD, et al. Development and optimization of OspC chimeritope vaccinogens for Lyme disease. Vaccine. 2020;38(8):1915–1924. doi.org/10.1016/j.vaccine.2020.01.027
82. Wagh D, Pothineni VR, Inayathullah M, et al. Borreliacidal activity ofBorreliametal transporter A (BmtA) binding small molecules by manganese transport inhibition. Drug Des Devel Ther. 2015;9:805–816. doi.org/10.2147/DDDT.S7706
83. Boyle WK, Sorensen HN, Bourret TJ. Essential components ofBorreliella(Borrelia)burgdorferi invitrotranscription assays. J Vis Exp. 2022(185):e64134. doi.org/10.3791/64134
84. Boyle WK, Hall LS, Armstrong AA, et al. Establishment of aninvitroRNA polymerase transcription system: a new tool to study transcriptional activation inBorrelia burgdorferi. Sci Rep. 2020;10(1):8246. doi.org/10.1038/s41598-020-65104-y

---

Republished from the open web under CC0. Authors: Van Gundy T, Schneider C, Ebmeier CC, Chambers GZ, Cramer NA, Chambers KM, Drecktrah D, Hyde JA, Skare JT, Marconi RT, Samuels DS, Lybecker MC. Read the original.