When unfavorable environmental conditions affect cell growth, spore-forming bacteria activate and coordinate a temporal molecular program to assemble a dormant spore, which sequesters essential biomolecules for subsequent conversion back into growing cells. Bacillus subtilis sporulation has proved to be an extremely valuable model system for the study of the principal biomolecules that are stored and shielded during spore dormancy. Proteome analysis has revealed several categories of proteins that are stored in dormant B. subtilis spores, including proteins involved in ribosome biogenesis, carbon metabolism, RNA processing, and protein synthesis. These protein groups play important roles in spore germination and outgrowth ( Swarge et al., 2019). In addition, dormant spores store RNA, which suggests that these RNAs play specific roles during dormancy or germination and outgrowth or are used as ribonucleotide reservoirs for de novo RNA synthesis during spore germination. Some of the RNAs are transcripts involved in the spore formation process, such as those coding for membrane proteins, transporters, RNA processing enzymes, and proteins involved in protein synthesis, modification and degradation ( Bassi et al., 2016; Keijser et al., 2007; Nagler et al., 2016; Segev et al., 2012). Among these RNA categories, tRNA has been detected in significant amounts in dormant spores, and represents an essential reservoir ( Nagler et al., 2016). One aspect of RNA metabolism that has not been well studied during spore germination and outgrowth is tRNA processing, and knowledge of those mechanisms will be essential to our understanding of long term stress survival by the spores and of the possible role of stored tRNAs in the generation of vegetative cells ( Deutscher, 2006; Li & Deutscher, 1996; Li & Deutscher, 2002; Mohanty & Kushner, 2000; Ow & Kushner, 2002; Raynal et al., 1998; Reuven et al., 1997).

In B. subtilis, tRNA ^Cys is encoded by a single copy gene and its study has significant relevance to aspects of processing and to the understanding of the global factors that affect cysteine metabolism. Thus, our previous studies have shown that tRNA ^Cys processing during exponential growth in B. subtilis involves the ribonucleases RNase Z, PNPase, RNase R and the 3’ end modifying enzyme, CCAase. The activities of these enzymes converge in ways that have not yet been completely elucidated to facilitate CCA addition and tRNA ^Cys repair or degradation ( Campos-Guillén et al., 2010; Campos-Guillén et al., 2019). To date, however, similar studies have not been performed on tRNA ^Cys metabolism during spore germination and outgrowth in B. subtilis.

A dormant spore is highly organized within a dehydrated spore core, which is enveloped by a dense inner membrane, a germ cell wall, a cortex and a spore coat ( Setlow, 2003). Each spore component is formed by molecules that play major roles in protecting the spore from a broad range of detrimental environmental conditions such as radiation, heat, desiccation and chemicals ( Setlow, 2006). Under specific conditions, environmental nutrients are recognized by the germinant receptors (GRs) and promote biochemical changes such as the release of monovalent cations and Ca ²⁺-dipicolinate, and the cortex is hydrolyzed and rehydrated. These molecular changes initiate the germination process ( Setlow, 2006). After this first step, a phase called outgrowth follows, in which metabolic activity and molecular organization (“ripening”) convert the germinated spore into a growing cell. During the outgrowth phase, at least 30% of B. subtilis genes are activated ( Horsburgh et al., 2001; Keijser et al., 2007).

One interesting aspect of spore germination is the effect of salt stress on the process and thus changes in the gene expression profile of outgrowing B. subtilis spores in the presence of high concentrations of NaCl have been analyzed by RNA sequencing ( Nagler & Moeller, 2015; Nagler et al., 2014; Nagler et al., 2015; Nagler et al., 2016). In the last study, the presence of tRNAs in high percentage was detected in dormant spores but the nature of the tRNA population was not analyzed in depth. Taking advantage of new methodologies in high-throughput RNA sequencing, the availability of datasets in the Gene Expression Omnibus (GEO) database and computational analysis, we have investigated the profile of tRNA ^Cys in dormant and outgrowing B. subtilis spores in the presence of 1.2 M of NaCl. We were able to identify several categories of tRNA ^Cys species in dormant spores and during the first minutes of their conversion into growing cells. This study provides valuable insights into tRNA ^Cys processing during this state of transition.

All experimental conditions to obtain RNA from dormant and outgrowing B. subtilis spores in the presence and absence of NaCl were previously described ( Nagler et al., 2016). The raw RNA-Seq data indicated as SRR3488622 to SRR3488635 in Table 1, obtained from the Gene Expression Omnibus (GEO) database under the accession number GSE81238 were analyzed. The RNA-seq raw data correspond to: T0 (dormant spore), outgrowth of spores during 30 (T30), 60 (T60) and 90 (T90) minutes after the initiation of germination in the presence and absence of 1.2 M NaCl. Each sample time was sequenced twice.

To avoid contributions to our results by tRNA isoacceptors whose genes do not encode CCA at the 3´end in B. subtilis, we examined products of the tRNA ^Cys gene, which is present in single copy in the genome of B. subtilis and is located at the distal end of the rrnD operon. The gene does not encode the CCA at the 3´end, which must be added post-transcriptionally. The CLC Genomics work bench 12.0.2 (CLC Bio, Aarhus, Denmark) was used to analyze reads from the tRNA ^Cys gene, and small RNA adapters and reads with ambiguous nucleotides were trimmed using default settings. Open source alternatives such as UniPro UGENE, UTAP or Galaxy could also have been used for this purpose. Trimmed reads were then were mapped to the B. subtilis 168 genome downloaded from the NCBI website (accession number NC_000964.3) with short read local alignment mapping using the default setting. We considered alignments with a length fraction of 0.8 and a similarity fraction of 0.8. Two mismatches and three insertions and deletions per read were allowed. The mapped reads results for all experiments ranged from 243 to 1,141.

Mapped reads were used in the sequence alignment program MUSCLE in MEGA X v10.1.7 with default settings ( Kumar et al., 2018) and high similarity aligned reads to the reference tRNA ^Cys sequence were manually counted and categorized as the following subpopulations: I) precursors (containing genomically encoded nucleotides at the 5´or 3´ ends), II) tRNA immature (these species were 71 nucleotides in length and lacked only the CCA-end), III) mature tRNA with 3´-CCA, IV) immature tRNA with 3´-CC. We also observed short, tRNA ^Cys-derived RNA fragments (tRFs) in this study. Reads representing such fragments were categorized as V) 3´-tRF fragments detected with CCA 3’ends, VI) 3´-tRF fragments detected with CC 3’ends, VII) fragments detected as 5´-tRF (71 bases – n, where n > 0, from the 3’ end) or internal fragments (between nucleotide positions 2 to 70), VIII) complete tRNA ^Cys with incorrect 3´ tails and IX) 3´-tRF fragments with incorrect 3´ tails. The distribution analysis of tRNA ^Cys subpopulations was done using the JMP 7 program (SAS Institute Inc., Cary, NC) with the frequency per group algorithm. Trend graphs were constructed in SigmaPlot software version 14.0 (Systat Software, San Jose, CA).

In order to provide in-depth insight into tRNA ^Cys processing in the dormant spore and during outgrowth in B. subtilis, RNA-Seq methodology and the GEO database ( GSE81238) were utilized. The total number of reads for all experiments ranged from 7 to 12 million, but an interesting finding was that there is a distribution of reads mapped for tRNA ^Cys in all experimental conditions tested ( Table 1).

In B. subtilis, tRNA genes are immersed in operons and tRNAs are therefore synthesized as precursors that undergo post-transcriptional modification to give functional molecules. These modifications include 5´and 3´ processing by ribonucleases, addition of CCA at the 3´end of those tRNAs that do not encode this sequence and post-transcriptional editing of ribonucleosides.

The tRNA ^Cys gene is present in single copy in the genome of B. subtilis and is located at the distal end of the rrnD operon, which codes for three rRNA molecules; 16S, 23S and 5S rRNA, and a cluster of tRNA genes found downstream of the 5S rRNA gene ( Vold, 1985; Wawrousek et al., 1984). Regulation of expression of the rrnD operon is mediated by two cognate σ ^A consensus promoter sequences (P1 and P2) upstream of the 16S rRNA gene ( Koga et al., 2006). A third σ ^A consensus promotor is localized in the 23S-5S intergenic space, and is associated with mechanisms of the stringent response ( Vold, 1985; Wawrousek et al., 1984). A number of reports have revealed that the regulation of expression of P1 and P2 is maintained during spore formation, and that these promoters remain active even during the late stages of sporulation ( de Hoon et al., 2010; Koga et al., 2006; Rosenberg et al., 2012). Other reports observed slight differences in the expression of the P1 and P2 promoters under nutritional stress in vegetative cells ( Natori et al., 2009; Samarrai et al., 2011).

Based on these reports, it is clear that tRNA ^Cys expression under the control of promoters P1 and P2 takes place during sporulation and that diverse sub-populations of tRNA ^Cys molecules might thus accumulate in dormant spores. To expand our knowledge of tRNA processing in dormant and germinating B. subtilis spores, RNA-Seq methodology was used to display the diversity of tRNA ^Cys species that were present in these conditions and in the spores germinated in the presence of high salt. Our results show for first time the diversity of the sub-populations of tRNA ^Cys stored in dormant spores in B. subtilis ( Figure 1) and contribute to the knowledge of tRNA ^Cys processing during the dormancy and outgrowth phases.

Our results show a progressive generation and degradation of tRNA ^Cys fragments ( Figure 3), which suggests that RNases are present and active at the various sampling times employed in these experiments. In fact, a recent report by Swarge et al. (2019) demonstrated via proteome analysis that several exo-ribonucleases are present in dormant spores, including PNPase, RNase R, RNase PH, YhaM, RNase Y and others. Transcripts for PNPase, YhaM and RNase Y were only observed at 15 minutes and for RNase R and RNAse PH at 30 minutes after the initiation of germination. Thus, these exo-ribonucleases remaining from sporulation are likely to play essential roles in the pathways of tRNA degradation and/or processing during the sporulation and ripening phases.

In the particular case of tRNA processing in B. subtilis, in vitro and in vivo studies of the endonucleolytic pathways of RNase P ( Evans et al., 2006; Kurz et al., 1998) or RNase Z ( Pellegrini et al., 2003) and the exonucleolytic pathways of RNase PH, PNPase, RNase R or YhaM have been performed ( Oussenko et al., 2002; Oussenko et al., 2005). From these studies it is clear that tRNA precursors with extensions at the 5´ end are eliminated by RNase P, while RNase Z processes the 3´end of almost all tRNAs lacking an encoded CCA ( Pellegrini et al., 2003) and the exonuclease RNase PH processes CCA-containing tRNA precursors ( Oussenko et al., 2002; Oussenko et al., 2005). The CCA motif at the 3´end is a key feature of all mature tRNAs and is essential for aminoacylation during protein synthesis ( Hou, 2010; Wellner et al., 2018). Our results show that an important subpopulation of mature tRNA with CCA at its 3’ end is stored in the dormant spore and its levels increase during outgrowth. A delay in this increase was observed in the presence of NaCl, due perhaps to an effect of salt on the activities of enzymes related to tRNA ^Cys processing. In fact, there are reports demonstrating the effects on CCA addition to tRNA ^Cys of RNase Z, PNPase, RNase R and CCAase ( Campos-Guillén et al., 2010; Campos Guillén et al., 2019). It remains to be seen how these enzymes are affected under salt stress conditions during spore outgrowth, but Nagler et al. (2016) reported the interesting observation that the cca gene was downregulated during 90 minutes of outgrowth in the presence of NaCl. A relevant feature is that the cca gene is immersed in the operon containing bacillithiol (a REDOX regulator and an α-anomeric glycoside of L-cysteinyl-D-glucosamine with L-malic acid) biosynthetic genes under σ ^A-dependent promoter control, which suggests that cca expression and the integrity of the 3’ terminus of tRNAs may be important during salt stress ( Campos Guillen et al., 2017; Gaballa et al., 2010; Gaballa et al., 2013).

According to previous reports, incorrect additions at the 3´end of tRNA ^Cys were detected in B. subtilis, which suggests that alternative pathways of repair or degradation may exist in that organism ( Campos-Guillén et al., 2010; Campos Guillén et al., 2019; Cruz Hernández et al., 2013). Although a high percentage of mature tRNA with the CCA 3´end was observed in dormant spores, the species that would be available to support initial protein synthesis during outgrowth, it remains to be determined whether this stored tRNA ^Cys is charged with cysteine. It should be noted in this regard that cysteinyl-tRNA synthetase was detected in dormant spores and its expression was also detected at 60 minutes after the initiation of germination ( Nagler et al., 2016; Swarge et al., 2019).

Detrimental responses to salt during spore germination in B. subtilis have been analyzed previously ( Nagler & Moeller, 2015; Nagler et al., 2014; Nagler et al., 2015; Nagler et al., 2016). From these studies, 402 genes related to the salt stress response were upregulated, some of them under σ ^A-dependent promoter control ( Nagler et al., 2016). Therefore, the observed accumulation of tRNA ^Cys precursors during the 90 minutes of outgrowth in our study suggests that upregulation of the rrnD operon, governed by σ ^A-dependent promoters is important for macromolecular synthesis during outgrowth. An unanswered question is which of the three rrnD promoters is responsible for this upregulation.

The ability to quantitatively monitor tRNA abundance changes from dormant spore to germination and outgrowth will be critical to understanding these processes and the roles of transcription, post-transcriptional modification and translation in them. The results reported here using RNA-Seq and data analysis provide new insights into the dynamic changes in the sub-populations of RNA species related to tRNA ^Cys molecules in dormant spores and during spore germination and outgrowth in the presence and absence of 1.2 M NaCl. An important unanswered question is what role these various subpopulations play in the metabolic economy of the spore and in the processes of germination and outgrowth.