Identification and Characterization of Non-protein Coding RNA Homologs in Serratia Marcescens by Comparative Transcriptomics

Abstract

The Serratia marcescens is a Gram-negative bacterium from the Enterobacteriaceae family. Recently, S. marcescens have evolved to become a versatile and opportunistic pathogen. Furthermore, this bacterium is also a multi-drug resistant pathogen exhibiting Extended-Spectrum Beta-Lactamases (ESBL) activity. This bacterium is highly associated with infections in healthcare settings and even leads to death. Hence, an advanced approach based on non-protein coding RNA (npcRNA) of S. marcescens was considered in this study to understand its regulatory roles in virulence, pathogenesis, and the differential expression of these transcripts in various growth phases of the bacterium. BLASTn search of known npcRNAs from Salmonella typhi, Escherichia coli, and Yersinia pestis against S. marcescens was performed to discover putative conserved homologous transcripts. The novelty of these putative homologous npcRNAs was verified by screening through the Rfam web tool. The target mRNA for the homologs was predicted via the TargetRNA2 webtool to understand the possible regulatory roles of these transcripts. The npcRNA homologs, which were predicted to regulate virulence target mRNA were assessed for their expression profile at different growth stages via reverse transcription PCR and the band intensity was quantitatively analysed using the Image J tool. The known npcRNA ssrS, from S. typhi showed expression in S. marcescens during three growth stages (lag, log, and stationary). Expression was observed to be high during the lag phase followed by a similarly low-level expression during the log and no expression during stationary phase. This ssrS homolog was predicted to regulate mRNA that encodes for protein FliR, which is associated with virulence. This is a preliminary study that lay the foundation for further elucidation of more virulence-associated npcRNAs that are yet to be discovered from S. marcescens, which can be useful for diagnostics and therapeutic applications.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12088-023-01160-y.

Introduction

Serratia marcescens is an opportunistic, Gram-negative and facultative anaerobic bacterium. S. marcescens belongs to the family, Enterobacteriaceae and is a member of the genus Serratia. S. marcescens can be found in various kinds of environments despite its preference for moist conditions [1]. This bacterium was identified in the year 1819 by an Italian pharmacist named Bizio [2]. Previously this bacterium was considered harmless and non-pathogenic, however this bacterium have evolved to become an opportunistic pathogen, which have been identified during the nosocomial outbreaks in intensive care units (ICUs), neonatal intensive care units (NICUs), and other hospital units over the last few decades [3]. This pathogenic bacterium affects immunocompromised individuals in addition to immunocompetent populations, and thus causing severe infections such as urinary tract infection, wound infection, endocarditis, pneumonia, and many more [4].

Since the year 1960, isolates of S. marcescens that are non-pigmented are prevailing over pigmented strains within the nosocomial settings and are increasingly found to be causing healthcare associated contamination, and later causing harm to compromised patients [5]. At the present time, Serratia is being classified as a genus that consists of 14 species, and more than half of them are involved in healthcare-associated infections [5]. S. marcescens is also the most prevalent clinical isolate and the most significant human pathogen among all Serratia species. From January 2016 to December 2017, a research study took place in Hospital Kuala Lumpur, Malaysia, with a total of 238 patients and 303 isolates obtained from patients that were used for the study. The study found out that the overall case fatality rate (CFR) was 27%, with S. marcescens having the highest CFR rate (53.3%). Furthermore, the number of deaths due to S. marcescens infections were much higher than those survived [6]

RNA can be categorized into two types, messenger RNAs (mRNAs) and non-protein coding RNAs (npcRNAs) [7]. The mRNAs encode for protein. NpcRNAs, which make up a large portion of the genome do not translate into protein but exert other post-transcriptional regulatory functions on its target genes [8]. Bacteria, as well as archaea, employ small non-protein coding RNAs regulators in order to modulate the expression of specific genes [9]. Several functions of npcRNAs have been identified in prokaryotes such as to regulate stress response, formation of biofilm, virulence, and procurement of nutrients [10]. Furthermore, controlling the gene expression of mRNAs in prokaryotes by npcRNAs were by the direct base pairing to the complement mRNA regions [10].

Recent discovery in Serratia marcescens revealed that the npcRNA OxyS was present in cis-acting manner as part of a peroxidase mRNA regulating the translation of peroxidase enzyme controlling the abundance of hydrogen peroxide (H₂O₂) within the cell [11]. The S. marcescens is known to be responsible for a few healthcare-associated infections and even in the worst case, mortality is also an effect of the bacterium’s infection. Currently available therapeutics such as antibiotics and vaccines had been rendered non-effective due to increase in multi-drug resistance of S. marcescens. Hence, a new approach of identifying npcRNAs regulating virulence activity in S. marcescens was considered a platform for discovery of alternate therapeutics options [12]. The scope of this research is to screen known npcRNAs against S. marcescens to identify homologous transcripts and understand the differential expression of these homologs, which could correlate with the regulatory roles against the predicted target mRNAs. These undertakings can revolutionize current treatments against diseases by incorporating new npcRNA-based therapeutics and developing efficient early detection of infection using npcRNA as biomarkers [13].

Materials and Methods

Identification and Screening for S. marcescens npcRNA Homologs

The npcRNAs within the same Enterobacteriaceae family were selected to screen for homologs in S. marcescens. The known npcRNA sequences from Salmonella typhi (133 npcRNAs), Escherichia coli (26 npcRNAs), and Yersinia pestis (5 npcRNAs) were collected and were subjected to BLAST against S. marcescens via comparative analysis using BLASTn tool. The criteria for the selection of highly possible homologs are such as E-values lower than 0.001, more than 80% identities, and the corresponding sequences selected should be of best hits. The identified npcRNA homologs were analyzed via Rfam online database to screen for homologs conserved within S. marcescens and the source organism (S. typhi, E. coli and Y. pestis) only. The identified npcRNA homologs were screened through the TargetRNA2 online database to predict the possible mRNA targets and their binding region based on the lowest binding energy.

Bacterial Culture Conditions and Total RNA Extraction from S. marcescens

Hundred ml of Luria–Bertani (LB) medium was prepared in a conical flask and inoculated with S. marcescens ATCC 1388 glycerol stock. The inoculated LB medium was incubated at 37 °C with an agitation rate of 180 rpm. The bacterial cells were harvested at lag phase (OD₆₀₀ 0.2), exponential phase (OD₆₀₀ 0.6), and stationary phase (OD₆₀₀ 1.0) [12, 14]. The total RNA extraction from the harvested bacterial cells was performed using the TRIzol method based on the manufacturer’s instructions. The final concentrations of the total RNA were determined using Nanodrop.

Gradient PCR Amplification was Conducted to Confirm the Best Annealing Temperature

Gradient PCR amplification was conducted to standardize the annealing temperature of the custom-designed oligos (Table 1) with a thermal cycle procedure of initial single cycle denaturation of 95 °C for 5 min followed by 38 cycles of the second denaturation at 95 °C for 2 min, annealing temperature of 60.9 °C for 90 s, extension at 72 °C for 2 min and final single cycle extension at 72 °C for 5 min. Total DNA was extracted from S. marcescens via the boiling lysate method. The amplicons were resolved on 1% agarose gel.

Table 1.

List of primers designed for reverse transcription PCR

npcRNA	Forward primers (5′–3′)	Reverse primers (3′–5′)
RNaseP bact a	AACAGTTCGTGGCACGGT	TGACCGGTAAGCCGGGTT
ssrS	ATTTCTCTGAGATGTTCGCCA	GAATCTCCGAGATGCCGT

Reverse Transcription-PCR for Expression Analysis of Selected npcRNA Homolog

Total RNA extracted from S. marcescens was subjected to 1% agarose gel electrophoresis to check if any genomic DNA was still present in the total RNA after DNase treatment. The DNase-treated total RNA was subjected to Reverse Transcription-PCR using custom designed oligonucleotides for selected npcRNA and 5SrRNA (control) gene (Table 1) [15]. First strand cDNA was synthesized using two-step PCR protocol whereby, the first step is to anneal the oligos to the RNA template by incubating at 70 °C for 5 min, 25 °C for 5 min, 50 °C for 1 h and finally 70 °C for 5 min. This cDNA was subsequently amplified using the same oligo set with an annealing temperature of 60.9 °C. The amplicon was resolved on agarose gel and visualized using gel documentation system. Amplicon bands were analyzed for expression using ImageJ software by creating density histogram representing the intensity of each visible band (KishanRaj et al., 2021).

Results and Discussion

Conservation of Homologous Known npcRNAs in S. marcescens

As observed from Table 2, total of 164 published npcRNAs from Salmonella typhi, Escherichia coli and Yersinia pestis were retrieved and screened against S. marcescens strain WW4 (accession NZ_CP027798). From a total of 164 npcRNAs candidates, 23 identified to be homologous (Table 4, supplementary data) to S. marcescens, of which 17 are from Salmonella typhi, 3 are from Escherichia coli and 3 are from Yersinia pestis. These 23 putative candidates were further screened in Rfam database to filter the annotated transcripts and found that none of these 23 transcripts present in the total 91 annotated sRNAs of S. marcescens in Rfam database (Table 5, supplementary data). Further BLASTn screening was conducted for the 23 putative candidates to detect for conservation in other organisms. Out of these 23 transcripts, 3 were identified to be conserved in S. marcescens and in the source organism only. The remaining 20 were conserved in a wide range of other bacterial species. In light of their narrow range conservation, these 3 candidates were subjected to target mRNA prediction using TargetRNA2 webtool and differential expression analysis via reverse transcription to understand their possible regulatory roles. From Table 3 it is observed that all three candidates are having percentage of identities above 80% and query coverage above 70%, which fulfill the selection criteria for highly conserved transcript [16].

Table 2.

Total number of homologous npcRNAs conserved in S. marcescens

Organism	Total number of npcRNA in source organism	Number of npcRNAs homologous in S. marcescens WW4	Conservation percentage (%) in S. marcescens WW4	Number of npcRNA conserved only in S. marcescens WW4 and corresponding S.typhi/E. coli/Y. pestis	Reference
S. typhi	133	17	12.78	3	[17–20]
E. coli	26	3	11.54	0
Y. pestis	5	3	60.00	0
Total	164	23	14.02	3

Table 3.

List of highly putativee npcRNA homolog candidates in S. marcescens identified by comparative analysis

Organism of known npcRNA	Name of known npcRNA	Region	Conservation in S. marcescens strain WW4
			Query coverage (%)	Identities (%)	E-value	Strand	5′ Coordinate	3′ Coordinate
E. coli	ssrS	Fully	100	89	3.00E-58	+	4,362,482	4,362,663
S. typhi	RNase_P	Antisense	75	97	7.00E-70	−	4,650,582	4,650,428
S. typhi	tmRNA	Fully	76	89	2.00E-45	+	4,097,661	4,097,805

Expression Profiling of Selected Homologous npcRNA of S. marcescens During Different Growth Stages

Searching for highly possible target mRNAs of the 3 promising npcRNA candidates using TargetRNA2 webtool predicted that ssrS could possibly regulate FliR, which is a flagella biosynthesis protein (Fig. 1), RNase_P could possibly regulate msgA coding for multidrug resistance associated virulence factors (Fig. 2) and tmRNA could possibly regulate mRNA encoding for the ABC transporter-like protein (Fig. 3).

Fig. 1.

Target mRNA (flagellar biosynthesis protein FliR) for ssrS homolog in S. marcescens predicted using TargetRNA2 webtool

Fig. 2.

Target mRNA (virulence protein msgA) for RNaseP homolog predicted using TargetRNA2 webtool in S. marcescens

Fig. 3.

Target mRNA (ABC transporter-like protein) for tmRNA homolog predicted using TargetRNA2 webtool in S. marcescens

A recent study demonstrated the role of ssrS in regulating oxidative stress response in E. coli [21]. The knock-out study of ssrS shows mutant strains fail to survive during oxidative stress condition due to the repression of mRNAs (soxS, ahpC, sodA and tpx) involved in oxidative stress adaptation. Hence, we could deduce that ssrS could regulate the same oxidative stress adaptation in S. marcescens. TargetRNA2 webtool finds the energetically most favourable binding sites between npcRNA and mRNAs targets, in which this study predicts that ssrS has a higher chance to bind with FliR mRNA (Fig. 1) [22].

Differential expression assay of ssrS (Fig. 4) observes high expression during lag phase followed by low expression during log phase, while there was no expression during stationary phase. Correlating this expression pattern could draw an assumption that mRNAs involved in oxidative stress adaptation in S. marcescens could be highly regulated during the lag phase by ssrS as this growth stage is crucial for the adaptation of microbial cells to the surrounding factors. Previous studies has described the roles of ssrS in the regulation of ribosomal RNA transcription in E. coli whereby ssrS directly binds to the RNA polymerase, interfering with the transcription initiation [23]. E. coli mutant deficient in ssrS has downregulated ribosomal RNA transcription during log and lag phase. Hence, in this study the lower expression of ssrS observed during the log and no expression during the stationary phase could lead to the translational inhibition of ribosomal RNA in S. marcescens.

Fig. 4.

Expression profile of ssrS homolog in S. marcescens during different growth stages. Right image shows quantitative analysis of the expression level using ImageJ tool

Recently, the FliR mRNA was found to participate in major virulence activity of S. marcescens inclusive of adhesion, biofilm formation, motility, and flagella synthesis [24]. Biofilm formation is also crucial for bacterial resistance to antimicrobials [25]. Bacteria actively form biofilm to survive and confer high virulence activity during lag and log phase [26]. The bacterium survive by moving to more conducive environments via their flagellated structure. Hence, by referring to Fig. 1 and 4, it can be deduced that ssrS could participate in the antimicrobial resistance activity of S. marcescens by regulating the expression of its target FliR during the lag and log phase of the bacteria [27].

Figure 5 shows the secondary structure of ssrS illustrating the binding region of Flir.

Fig. 5.

Secondary structure of ssrS homolog predicted using RNAfold webtool. Blue-colored sequence is the binding site of target mRNA (FliR) in S. marcescens

The virulence protein MsgA was recently identified to participate in multidrug resistance in carbapenem-resistant Serratia marcescens [28]. The MsgA protein’s role during multidrug resistance could be either through binding to the antimicrobial surface and modifying the surface structure rendering the molecule inactive or preventing the antimicrobials to bind its target molecule or receptor by acting as competitors to the antimicrobials. TargetRNA2 search for RNase_P homolog predicted MsgA  mRNA with the lowest hybridization energy in S. marcescens. Hence, RNase_P is assumed to regulate MsgA controlling the virulence mechanism associated with multidrug resistance. The secondary structure prediction for RNase_P (Fig. 6) also shows less folded binding region for MsgA.

Fig. 6.

Secondary structure of RNaseP homolog predicted using RNAfold webtool. Blue-colored sequence is the binding site of target mRNA in S. marcescens

The tmRNA posses dual roles as tRNA and mRNA helping in repairing broken translation machineries, restoring synthesis of full-length proteins within the bacterial cell. The tmRNA was identified to be involved in the various pathogenesis pathways in S. typhi [29]. TargetRNA2 search for tmRNA predicts ABC transporter-like protein (Fig. 3). This protein is involved in the export or import of a wide variety of substrates ranging from small ions to macromolecule, deducing the functional role of tmRNA in regulating the transport of molecules between cell membrane of S. marcescens [30]. The secondary structure prediction for tmRNA (Fig. 7) also shows less folded binding region for the target mRNA.

Fig. 7.

Secondary structure of tmRNA homolog predicted using RNAfold webtool. Blue-colored sequence is the binding site of target mRNA in S. marcescens

Conclusion

In this study, we have adopted the comparative genomic method for the first time to screen and identify conserved npcRNA homologs in S. marcescens. The known npcRNAs from E. coli, S. typhi and Y. pestis were screened against the genome of S. marcescens using BLASTn webtool. The output was a total of 23 transcripts initially found to be potential homologs in S. marcescens. After further screening of these 23 transcripts via Rfam and NCBI databases, 3 transcripts (ssrS, RNase_p and tmRNA) were found to be more promising homologs as they are only present in S. marcescens and the corresponding source bacteria (E. coli and S. typhi). These 3 homologs were further subjected to target mRNA prediction during the post-transcriptional stage. Interestingly ssrS and RNase_P were showing possibilities to bind to virulence associated mRNAs (FliR and MsgA). All 3 transcripts were subjected to differential expression analysis and only ssrS was observed to be expressed during the lag and log growth stage of the bacteria. Hence, we predicted the possible regulatory roles of ssrS against its target mRNA FliR during these two growth stages. These findings, could be a stepping stone for further functional elucidation of these npcRNA homologs using experimental strategy.

Supplementary Information

Below is the link to the electronic supplementary material.