The B. burgdorferi isolate B31-5A4 used in this work has been characterized in previous studies ( (Casselli et al., 2012; Casselli et al., 2019); Table 1 ). B. burgdorferi clones were grown at 35°C under 1.5% CO₂ in modified Barbour-Stoenner-Kelly medium (BSK-II) supplemented with 6% rabbit serum (Cedarlane) (Barbour, 1984). Mutant strains were grown in BSK-II supplemented with kanamycin (200 µg/ml) and/or gentamicin (100 µg/ml) as indicated. Culture density was monitored by visualization under dark-field microscopy using a Petroff-Hausser counting chamber.

A list of mutant strains and their corresponding genotypes are provided in Table 1 . All transformations were performed using electrocompetent B. burgdorferi cells as previously described (Samuels, 1995; Bankhead and Chaconas, 2007). Transformations were recovered in drug-free media for 24 hours and plated by limiting dilution with antibiotic selection to isolate clonal transformants. Positive clones were identified by PCR for introduced sequence. In trans complement strains were further verified by visualization of the respective shuttle vectors on agarose gel. Endogenous plasmid content was analyzed in all strains by multiplex PCR using primers specific for regions unique to each plasmid as previously described (Bunikis et al., 2011).

The experiments on mice were carried out according to the protocols and guidelines approved by American Association for Accreditation of Laboratory Animal Care (AAALAC) and by the Office of Campus Veterinarian at Washington State University (Animal Welfare Assurance A3485-01 and USDA registration number 91-R-002). These guidelines are in compliance with the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals. The animals were housed and maintained in an AAALAC-accredited facility at Washington State University, Pullman, WA. The Washington State University Institutional Animal Care and Use Committee approved the experimental procedures carried out during the current studies.

Four-week-old male immunocompetent C3H/HeJ (C3H) and immunodeficient C3SnSmn.CB17-Prkdc^scid /J (SCID) (Jackson, Bar Harbor, ME) mice were infected by subcutaneous needle inoculation (near the base of the tail, to the right of the midline) with 100 µl BSK-II containing 5X10³ total B. burgdorferi cells unless indicated otherwise.

Blood (day 7 post infection), ear (day 14, 21, and 28 post infection), joint, heart, and bladder (day 28 post infection) tissues were collected and cultured in BSK-II supplemented with 20 μg/ml phosphomycin, 50 μg/ml rifampicin, and 2.5 μg/ml amphotericin-B. Dark-field microscopy was used to determine the presence or absence of viable spirochetes for each cultured tissue sample. A sample was deemed negative if no spirochetes could be seen in 10 fields of view after 3 weeks of culture.

For bacterial burden quantification, heart, bladder, joint, and ear tissues were harvested at day 28 post infection and immediately snap-frozen in liquid nitrogen prior to storage at -80°C. DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA) following the manufacturer’s instructions. DNA samples were then cleaned and concentrated using Genomic DNA Clean & Concentrator Kit (Zymo Research, Orange, CA). Quantitative PCR for the B. burgdorferi flaB and mouse β-actin genes was performed in duplex on each sample in triplicate using TaqMan probes and the droplet digital PCR system (ddPCR™, BioRad, Carlsbad, CA). All steps, including droplet generation, thermocycling, and droplet reading were performed following the manufacturer’s instructions. All reactions were performed with 4ng of template DNA. Primers and probes used for this assay are listed in Table 1 .

For RNA extraction from in vitro cultures, triplicate cultures of wild-type, Δ1-7, Comp7N_c, and Comp7N_t were grown to late log phase (1x10⁸ spirochetes ml^-1), pooled, and pelleted by centrifugation. RNA was extracted using a hot phenol method described previously (Jahn et al., 2008; Drecktrah et al., 2015; Popitsch et al., 2017; Drecktrah et al., 2018). For RNA isolation from mouse tissues, three 4- to 6-week-old C3H mice were infected with a total of 10⁵ wild type spirochetes. Infection was verified and monitored as described above. At day 21 post infection, heart, bladder, and joint tissue were harvested and snap frozen in liquid nitrogen. Tissues were homogenized by mortar and pestle under liquid nitrogen. Homogenates were suspended in Trizol reagent (Invitrogen), and RNA was extracted following manufacturer’s instructions using chloroform followed by alcohol precipitation.

Genomic DNA was removed from RNA samples using the Turbo DNase free kit (Invitrogen). cDNA was synthesized from 1μg of RNA using the iScript cDNA synthesis kit (BioRad). Non-quantitative control PCR reactions were performed using RNA and genomic DNA templates corresponding to each sample to verify successful removal of contaminant DNA, primer specificity, and the absence of bbd07/SR0726 amplification from Δ1-7 derived DNA. A total of 1uL of cDNA from each sample was diluted 1:10 and added to quadruplicate PCR reactions. PCR was performed using Evagreen supermix and the QX200 ddPCR system according to the manufacturer’s instructions (BioRad). Replicate reactions for each sample were performed at least in quadruplicate.

Cultures of indicated strains were grown in triplicate to late log-phase (1x10⁸ cells mL^-1) under standard culture conditions as described above. Replicate cultures were pooled and pelleted by centrifugation, washed, and resuspended in 90μL PBS. Lysis was performed by adding 30μL of 4X Laemmli loading dye containing 10% β-mercaptoethanol, and then heating the samples to 80°C for 10 minutes following at least one freeze-thaw cycle. Lysates (10⁹ cells) were electrophoresed on a precast 4%-15% polyacrylamide gradient gel (Bio-Rad). Western blotting was performed as described previously (Casselli et al., 2012), using sera harvested from C3H mice infected with 5x10³ spirochetes of the indicated strain for 28 days. Purified polyclonal antibodies (Rockland) were used in and anti-FlaB Western blots.

Cultures were grown in triplicate under standard culture conditions to mid-log phase (5x10⁷ spirochetes mL^-1), then temperature shifted by 1:100 dilution into room temperature BSK-II. Cultures were allowed to grow back to mid log phase, at which point they were subcultured again 1:100 into warm BSK-II. These final cultures were then allowed to grow under standard conditions to late log-phase (1x10⁸ cells mL^-1). Cultures were pooled, and RNA was extracted using a hot phenol method described previously (Jahn et al., 2008; Drecktrah et al., 2015; Popitsch et al., 2017; Drecktrah et al., 2018). After DNase treatment, RNA was assessed for integrity using an Agilent 2100 Bioanalyzer. Clean, high quality RNA samples were shipped to the Novogene Corporation for library preparation, sequencing, and data analysis. 3 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext^® Ultra™ Directional RNA Library Prep Kit for Illumina^® (NEB, USA) following manufacturer’s recommendations and index codes were added to attribute sequences to each sample. Following rRNA removal, fragmentation was carried out using divalent cations under elevated temperature in NEBNext First Strand Synthesis Reaction Buffer (5X). First strand cDNA was synthesized using random hexamer primer and M-MuLV Reverse Transcriptase (RNaseH-). Second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H. In the reaction buffer, dNTPs with dTTP were replaced by dUTP. After adenylation of 3’ ends of cDNA fragments, NEBNext Adaptor with hairpin loop structure were ligated to prepare for hybridization. In order to select cDNA fragments of preferentially 150~200 bp in length, the library fragments were purified with AMPure XP system (Beckman Coulter, Beverly, USA). Then 3 μl USER Enzyme (NEB, USA) was used with size-selected, adaptor-ligated cDNA at 37°C for 15 min followed by 5 min at 95°C before PCR. Then PCR was performed with Phusion High-Fidelity DNA polymerase, Universal PCR primers and Index (X) Primer. Lastly, products were purified (AMPure XP system) and library quality was assessed on the Agilent Bioanalyzer 2100 system. The clustering of the index-coded samples was performed on a cBot Cluster Generation System using HiSeq PE Cluster Kit cBot-HS (Illumina). After cluster generation, the library preparations were sequenced on an Illumina platform and paired-end reads were generated. Raw data of fastq format was firstly processed through in-house perl scripts. In this step, clean reads were obtained by removing reads containing adapter, reads containing ploy-N, and low-quality reads. Alignment of clean reads to reference genome was performed with used Bowtie2-2.2.3. (Langmead and Salzberg, 2012). HTSeq v0.6.1 was used to count the read numbers mapped to each gene. Prior to differential gene expression analysis, for each sequenced library, the read counts were adjusted by edgeR program package through one scaling normalized factor. Differential expression analysis between the strains was performed using the DEGSeq R package (1.20.0). P values were adjusted using the Benjamini & Hochberg method (Green and Diggle, 2007).

RNA was extracted from triplicate cultures of indicated strains exactly as described above for RNAseq. Fifteen µg (SR0726) or 1.5 µg (5S) of RNA from each strain was separated using the SequaGel UreaGel 19:1 Denaturing Gel System (National Diagnostics) and transferred to nylon membranes (Rosche) using a Trans-Blot^® Turbo^™ transfer system (BioRad) run at 25V for 45 minutes. After transfer and UV crosslinking, dried membranes were prehybridized with probe-free UltraHyb^® oligo hybridization buffer (Invitrogen) for 2 hours at 42°C, then hybridized overnight at the same temperature with biotinylated probe P1339 or P1349 ( Table S1 ) at a final concentration of 200 pM. After washing (2X SSC, 0.1% SDS) and blocking (1X Intercept^® PBS blocking buffer, 0.5% SDS), membranes were treated with IRDye^® 800CW Streptavidin (LI-COR Biosciences) in blocking buffer, then washed with PBST. Detection was performed on the Odyssey^® CLx infared imaging system (LI-COR Biosciences).

Fisher’s exact test was used to determine significant differences in the ability to recover a given recombinant strain by culturing of tissues compared with wild type B. burgdorferi. Student’s t-test was performed to determine significant differences in spirochete burden in heart tissue samples from qPCR analyses, where the average burden in heart tissues from mice infected with a given recombinant strain was compared to that of mice infected with the wild type. Student’s t-test was also used to determine significant differences in the average in vitro SR0726 transcription levels between a given recombinant strain and the wild type by qRT-PCR. One‐way ANOVA followed by all pairwise multiple comparison (Holm‐Sidak) was used to determine significantly different levels of average in vivo SR0726 transcription by the wild type between each tissue tested, as well as significantly different bb0360 and bb0733 transcription levels between strains by qRT-PCR. All statistical tests were performed using SigmaPlot.

The recent detection of a putative sRNA encoded within the intergenic space of lp17 containing the discontinued bbd07 ORF annotation (NC_001849.1 [discontinued]), coupled with numerous failed attempts by our laboratory to detect a BBD07 protein product, provided strong indication that the functional product encoded by the bbd07 locus is the SR0726 sRNA transcript (Popitsch et al., 2017).

To further rule out the possibility of functional BBD07 protein production, two in cis complement strains were generated in our previously described lp17 mutant strain lacking a region containing bbd01-bbd07 that is incapable of murine heart tissue colonization (Δ1-7; (Casselli et al., 2019)). These two genetically complemented strains harbored a bbd07 copy containing either a premature stop codon or a disrupted start codon, and were denoted Comp7Nstop _c and Comp7NΔstart _c , respectively ( Table 1 ). Both constructs included 200 bp of native sequence upstream of the previously annotated bbd07 that are also present in the previously described native complement Comp7N _c (Casselli et al., 2019). The rationale was that if the genetic region encodes a protein that contributes to the mutant phenotype, then these strains would not be capable of translating the functional protein, leading to a defective tissue colonization phenotype comparable to the Δ1-7 mutant strain. Alternatively, if the functional product encoded by the region is sRNA SR0726, then these mutations would be expected to have no effect on functionality, and thus the strains would exhibit tissue colonization levels comparable to wild-type control and native cis complement (Comp7N _c ) strains. Single nucleotide mutations were made in frame as defined by the previously predicted TTG start codon at bp 4373 of B. burgdorferi B31A lp17(NC_001849). For Comp7Nstop _c , the stop codon was generated via a cytosine to adenine mutation corresponding to bp 4306 of the annotated lp17 DNA sequence. For Comp7NΔstart _c , a thymine to cytosine mutation was introduced in the middle nucleotide of the previously predicted start codon that corresponds to bp 4372 of the annotated lp17 DNA sequence ( Figure 1A ). Neither mutation was predicted to significantly perturb the putative RNA secondary structure that might be important for the function of the SR0726 sRNA product. To generate the clones, suicide vector-derived plasmids harboring the 317 bp sequences were used to transform Δ1-7 B. burgdorferi cells, and transformants were screened for insertion of the altered sequences in lp17. PCR verification of these clones demonstrated successful complementation as shown in Figure 1B . Sequencing was performed to verify the single base pair mutations in each construct, and isogenicity to the parent clone was confirmed by multiplex PCR for native plasmid content (data not shown) (Bunikis et al., 2011).

To determine if these altered sequences were able to rescue the mutant phenotype, the capacity for heart tissue colonization of immunocompetent mice was selected for use as a readout. We reasoned that this was a reliable indicator of gene product functionality due to the fact that the Δ1-7 mutant has been previously shown to be unable to colonize heart tissue in immunocompetent mice (Casselli et al., 2019). Groups of five C3H mice each were needle inoculated (5x10³ total spirochetes) with either wild type or one of the various complement clones. Successful establishment of infection was monitored weekly by culturing blood and ear tissues to visually verify the presence of spirochetes by dark field microscopy ( Table 2 ). All mice in the experiment displayed spirochetemia at 7 days post infection and remained infected for the duration of the 28-day experiment. Recovery of Δ1-7 spirochetes from ear tissue samples on day 14 post infection was variable in a previous experiment, while here all ear tissue samples taken at this time point from Comp7Nstop _c - and Comp7NΔstart _c -infected mice yielded spirochete growth in culture. At day 28 post infection, heart tissue was collected from each mouse and divided in half longitudinally. One half of each heart was deposited into culture media, while the other half was snap-frozen for DNA extraction and qPCR to determine bacterial burden.

Comp7Nstop _c and Comp7NΔstart _c spirochetes were recovered from heart tissue to the same extent as the wild type and Comp7N _c strains ( Table 2 ). Moreover, qPCR analysis of total DNA from the tissues showed no statistically significant differences in spirochete burden between any of the tested strains (P<0.05, Figure 1C ). These combined findings demonstrate that neither introduction of a premature stop codon nor the destruction of the potential start codon affect the functionality of the gene product in the context of heart tissue colonization, further suggesting that the functional gene product encoded by the bbd07 intergenic region is not a protein, but rather the sRNA SR0726.

Our laboratory recently reported that absence of the region encoding SR0726 affects antigen expression in vivo, which could potentially explain the observed attenuation in tissue colonization and pathogenesis by the mutant (Casselli et al., 2019). Thus, it was hypothesized that SR0726 overexpression may lead to altered murine infectivity due to abnormal antigen production. B. burgdorferi has been shown to maintain the pBSV2G shuttle vector at a higher copy number than its native plasmids, and expression levels of genes complemented on this vector can be elevated compared to wild type (Elias et al., 2003; Tilly et al., 2006). Thus, in trans complemented strains that harbored bbd07 gene copies on pBSV2G were generated in the Δ1-7 mutant background to assess the effects of SR0726 overexpression. Complementation was performed by transforming Δ1-7 cells with pBSV2G carrying either a wild-type copy of bbd07 (Comp7N _t ) or a copy containing an early stop codon (Comp7Nstop _t ). An empty vector control strain harboring pBSV2G without a bbd07 copy was also generated (CompE _t ). PCR verification of the presence or absence of bbd07 DNA in these strains is illustrated in Figure 2A , and isogenecity to the parent strain was confirmed via multiplex PCR (data not shown) (Bunikis et al., 2011).

To determine the relative in vitro expression levels of SR0726 in the wild type and Comp7N _t , qRT-PCR analysis was performed using cDNA derived from pooled triplicate cultures of wild type, Δ1-7, and Comp7N _t grown to late log-phase (1x10⁸ cells mL^-1). As shown in Figure 2B , a ~24-fold increase in SR0726 transcription by Comp7N _t spirochetes compared to the wild type was observed. Wild-type transcription of SR0726 was found to be low, producing less than two SR0726 copies per 10⁴ flaB copies. These results indicate that transcriptional induction of SR0726 may be tightly regulated, and that residence of the encoding sequence on a high-copy vector results in elevated in vitro transcription.

To assess the effects of high copy SR0726 expression on murine infection, groups of five C3H mice each were needle inoculated (5x10³ total spirochetes) with wild type, Comp7N _t , Comp7Nstop _t , or CompE _t spirochetes. Infection was monitored weekly by culturing of tissues as before. Interestingly, while infection was established in all mice inoculated with wild type or the CompE _t control clone, infection was not established in any of the mice inoculated with Comp7N _t or Comp7Nstop _t spirochetes ( Table 3 ). To ensure reproducibility, a second independent infection assay with the Comp7N _t and Comp7Nstop _t clones was conducted with the same results. Next, groups of three C3H mice each were inoculated with 10⁴, 10⁵, or 10⁶ Comp7N _t spirochetes alongside three mice infected with 10⁴ wild type spirochetes in order to test if the infectivity defect associated with SR0726 overexpression could be overcome by increased inoculum dosage. While all wild type-inoculated mice became infected, none of the mice inoculated with Comp7N _t became infected, indicating that this phenotype is not dose dependent (data not shown). To determine whether the lack of infectivity exhibited by these strains in C3H mice was due to an inability to evade the adaptive immune response, Comp7N _t and Comp7Nstop _t spirochetes were used to inoculate groups of five immunodeficient SCID mice. Neither strain was able to establish infection in any of the mice tested during the 28-day experiment ( Table 4 ). Together, these findings suggest that high-copy expression of SR0726 renders spirochetes unable to establish murine infection, even in the absence of an adaptive immune response.

Recently, we showed that the murine tissue colonization defect and attenuated pathogenesis exhibited by the Δ1-7 mutant during murine infection was associated with dysregulated antigen expression (Casselli et al., 2019). Considering this, along with the finding that the in trans bbd07 complement strains were non-infectious in mice, it seemed possible that overexpression of SR0726 might also result in dysregulated antigen expression that in turn renders those strains non-infectious. Because SR0726 transcription by Comp7N _t spirochetes is highly elevated during in vitro cultivation ( Figure 2B ), we predicted that any potential resultant protein expression changes would be observable under the same conditions.

To assess the effects of SR0726 overexpression on overall antigen production, cultures of wild type, Δ1-7, Comp7N _c , Comp7N _t , and CompE _t spirochetes were grown in triplicate under the same conditions used to quantify SR0726 transcript levels in the Comp7N _t strain. Pooled protein lysates (10⁹ total cells) of each strain were subjected to Western blot analysis using murine immune sera harvested from mice that had been infected with wild-type B. burgdorferi for 28 days. Unexpectedly, no notable differences were observed between the in vitro antigenic profiles of the tested strains ( Figure 2C ). This result suggests that the non-infectious phenotype exhibited by Comp7N _t is unlikely to be due to dysregulated antigen expression during inoculum preparation of in vitro-grown spirochetes. Western blotting of in vitro-grown wild-type whole cell lysates using Comp7N _t antisera was also attempted, but it was found that insufficient antibody titers are generated following inoculation with this non-infectious clone.

Transcription of the SR0726 locus was previously shown to increase during in vitro growth at 37°C compared to 23°C, suggesting that it may be upregulated during transmission and/or infection of the mammalian host (Popitsch et al., 2017). To determine whether SR0726 transcription is in fact induced in the mammalian host, and to assess any potential expression differences in spirochetes colonizing various tissue sites, SR0726 transcripts were quantified from infected murine tissues by qRT-PCR. Groups of three mice each were infected with 1x10⁵ total wild-type spirochetes, and heart, bladder, and joint tissues were harvested at 28 days post infection for total RNA extraction and subsequent analysis. Relative abundance of SR0726 expression was calculated for each tissue. Expression of SR0726 was detected in each tissue examined, with an average expression level of approximately 20, 65, and 95 copies of SR0726 per 10³ copies of flab in spirochetes colonizing either the heart, bladder, or joints, respectively ( Figure 3 ). These values range from a 100- to 400-fold increase in SR0726 transcript levels relative to that observed for in vitro-grown wild type spirochetes (~2 copies of SR0726 per 10⁴ copies of flab, see Figure 2B ), supporting the hypothesis that SR0726 transcription is upregulated during mammalian infection. Despite SR0726 transcription being ~100-fold higher in heart-resident spirochetes than those grown in vitro, this in vivo SR0726 expression level was significantly less than that observed for spirochetes colonizing joint tissue (P<0.05). Differences in SR0726 transcription between heart and bladder, and bladder and joint did not reach significance.

To further examine a potential regulatory role for SR0726, we sought to determine if deletion of SR0726 affects the transcriptome of B. burgdorferi. To eliminate any possible secondary effects brought about by the additional loss of bbd01-bbd06 in the Δ1-7 mutant clone, it was necessary to generate a knockout clone that lacked only the bbd07 locus. Deletion of the 317 bp region of lp17 encoding SR0726 was performed through allelic exchange using a suicide vector-derived plasmid construct containing a gentamycin resistance cassette flanked by upstream and downstream regions of DNA homologous to lp17 ( Figure 4A ). A positive clone (Δd07) was selected and recovered via PCR screening as shown in Figure 4B . The endogenous plasmid profile of this clone was identical to the wild type 5A4 parent strain with the exception of cp32-6, which is not required for infectivity (Brisson et al., 2013). Deletion was further verified by RT-PCR to ensure the absence of SR0726 transcription ( Figure 4C ). This clone did not display any apparent defects in in vitro growth rate, morphology or motility, consistent with the absence of these defects in the Δ1-7 mutant tested in a previous work (Casselli et al., 2019). Numerous attempts to complement the SR0726 region via allelic exchange were unsuccessful. As an alternative strategy, a complement clone was generated by transforming Δd07 cells with gDNA from a previously described strain harboring an lp17 plasmid in which the bbd01-bbd03 region was replaced with a kanamycin resistance gene (Δd07Comp _c ). bbd01-bbd03 are known to be dispensable in terms of murine tissue colonization (Casselli et al., 2019). This plasmid replacement approach has been utilized in several past studies where traditional complementation was not achievable in other B. burgdorferi mutant clones (Imai et al., 2013; Raffel et al., 2014). This clone was PCR negative for a gentamycin resistance cassette and did not grow in culture in the presence of gentamycin, which confirmed complete plasmid replacement (data not shown).

A mouse infection assay was carried out to ensure the newly generated Δd07 mutant strain would exhibit an in vivo phenotype similar to the Δ1-7 mutant clone. Groups of five C3H mice each were needle inoculated with the wild type, Δd07, or Δd07Comp _c at a dose of 5x10³ total spirochetes. All mice became infected as determined by culture positivity of blood and ear tissue collected weekly following infection. At day 28 post infection, heart, bladder, joint, and ear tissue were harvested and cultured for detection of spirochete growth by dark field microscopy as described before. As was previously observed for Δ1-7, none of the heart tissue cultures from Δd07-infected mice were positive for spirochete growth, further demonstrating that SR0726 transcription is required for heart tissue colonization in mice ( Figure 5A ). Unexpectedly, joint tissue colonization was attenuated for Δd07 compared to wild type, with spirochetes being recovered from that tissue in only 2/5 Δd07-infected mice. This contrasts with Δ1-7, which was recovered from joints of all 13 mice at the 28-day time point, despite its recovery from joints being attenuated at day 14 in prior experiments (Casselli et al., 2019). Δd07Comp _c spirochetes were recovered from all tissues in all mice, supporting the validity of the observed phenotype of targeted bbd07 deletion.

As a final confirmation that the Δd07 mutant exhibits similar defects to those of the Δ1-7 strain, the in vivo antigenic profile of Δd07 was compared to those of the wild type and Δ1-7 strains. This was done by subjecting in vitro-grown wild-type spirochetes to Western blot analysis using pooled antisera from groups of 3 mice infected for 28 days with either wild-type, Δ1-7, or Δd07 spirochetes. Sera from Δd07- and Δ1-7-infected mice displayed detectable reactivity to at least two additional antigens compared to sera from wild type-infected mice, indicating altered expression of these antigens during murine infection by strains lacking SR0726 transcription ( Figure 5B ).

The Δd07 mutant was then analyzed by RNA-seq to determine if the absence of SR0726 transcription affects the B. burgdorferi transcriptome. To accomplish this, triplicate cultures of wild-type or Δd07 spirochetes were grown to a density of 1x10⁸ spirochetes mL^-1 at 37°C following a temperature shift involving a 1:100 subculture from room temperature cultures. This temperature shift was performed in an attempt to induce SR0726 expression by the wild type (Popitsch et al., 2017). Pooled RNA from each strain was DNase treated and sent to the Novogene Corporation for library preparation, sequencing, and data analysis. Table 5 shows a list of 17 transcripts that were differentially expressed between wild type and Δd07 transcriptomes with a log₂-fold change > ± 2. In total, 91 transcripts were differentially detected with a log₂-fold change > ± 1, omitting those transcripts encoded by cp32-6, which is absent in the Δd07 clone. Full DEG analysis results can be found in Supplementary Tables 2, 3 .

The most significant transcriptional difference observed was in bb0360, which was detected approximately 34-fold more from Δd07 compared to wild type (P=3.51x10^-12) and encodes a hypothetical protein. Sultan et al. demonstrated that this gene is the first in a co-transcribed operon including the subsequent four genes (bb0361-bb0364) (Sultan et al., 2010). Despite this, none of the other transcripts in the putative polycistron were detected differentially between the strains tested. bb0733 was detected 5.5-fold less from Δd07 compared to wild type and encodes PlzA, a well characterized c-di-GMP binding protein. A volcano plot illustrating a compilation of the differential expression analysis is shown in Figure 6A . In order to corroborate the findings of this experiment, qRT-PCR analysis was performed to compare the relative transcription levels of bb0360 and bb0733 between wild type, Δd07, and Δd07Comp _c . Consistent with the trends observed by RNAseq, bb0360 transcription was significantly elevated (P<0.05, Figure 6B ) and bb0733 transcription significantly diminished (P<0.05, Figure 6C ), in Δd07 compared to wild type and Δd07Comp _c . The fold change in bb0733 transcription between wild type and Δd07 was nearly identical to what was observed by RNAseq, while the degree of difference in bb0360 transcription was not as stark. Combined, this data demonstrates that SR0726 has effects on the in vitro transcriptome of B. burgdorferi, and that the overall effects of its activity may be diverse.

Sequencing of the temperature-dependent sRNA transcriptome by Popitsch and colleagues has previously revealed that SR0726 is likely expressed as a longer proto-transcript which is processed into a 98 nucleotide (nt) sRNA corresponding to bps 4269-4366 of lp17 that is highly favored during growth in vitro at 37°C compared to at room temperature (Popitsch et al., 2017). To test this potentiality and to confirm the existence of a stably transcribed SR0726 sRNA, Northern blot analysis was performed using a probe complementary to a run of nucleotides contained in the sequence reported in the study mentioned above. RNA from wild type, Δd07, Δd07Comp _c , and Comp7N _t was prepared identically to that used for the RNAseq analysis reported above. Comp7N _t -derived RNA was used in this experiment to ensure detectability of the transcript, as this clone had been shown to overexpress SR0726 ( Figure 2B ). As illustrated by the most intense band in Figure 7 , this analysis confirmed the presence of a stably transcribed sRNA of approximately 98 nucleotides being produced in each strain except the Δd07 mutant. Additionally, a band corresponding to a larger transcript is visible in the wild type and Comp7N _t lanes which may represent the hypothetical unprocessed transcript. The band intensity from Δd07Comp _c RNA was remarkably low and may have been due to variability of SR0726 expression levels in vitro.

Using this dominant 98 nt sequence as an input, a folding prediction with the mfold webtool was conducted, and a binding interaction prediction with the bb0360 coding sequence with intaRNA (Freiburg RNA Tools) was then performed (Zuker, 2003; Busch et al., 2008). These in silico analyses revealed a potentially highly structured SR0726 RNA ( Figure 8A ) that may bind through complementarity to the extreme 3’ end of the bb0360 mRNA ( Figure 8B ).