Two classes of regulatory elements located at the 5′ UTR of mRNAs have been shown to play important roles in gene expression: RNA thermometers and RNA riboswitches. These RNA elements allow the bacteria to rapidly and efficiently react to environmental stimuli. Taking into consideration the mode of action and simplicity of these regulators, these RNA elements are very appealing for the development of new tools to regulate gene expression, and they can be used for many applications.
4.2. RNA Riboswitches
Riboswitches are
cis-acting RNA elements that recognize metabolites, thus modulating gene expression in response to specific small molecules. In bacteria, most of the riboswitches are located at the 5’ UTR of a particular transcript, and are composed by two functional domains: the ligand-sensing domain (or the aptamer domain) and the regulatory domain (or the expression platform). In certain conditions, a small molecule binds the aptamer domain, inducing a conformational change that stimulates the expression platform. The expression platform will act over the coding sequence, thus regulating its expression. For more information about the mechanism of action of the riboswitches, please read [
138].
Riboswitches were first discovered in 2002 [
139,
140,
141], and since then, they have been acknowledged as crucial contributors for the control of gene expression in many organisms. They can bind to a plethora of small molecules, from vitamins to sugars, amino acids or metals, and they can exert their function in different ways [
142]. Additionally, in some pathogens, important genes related with virulence are controlled by riboswitches. This was demonstrated in
L. monocytogenes, where the major regulator of virulence, PrfA, was shown to be controlled by two riboswitches, which also function as ncRNAs [
76], and in
Clostridium difficile, where it was shown that riboswitches are important for growth and infectivity [
143].
The existence of riboswitches in pathogenic bacteria presents novel targets for drug development. This has led researchers to start to manipulate how riboswitches bind to their ligands in order to design new molecules that could be used as antimicrobials [
144,
145,
146]. The high-resolution crystal structures of the riboswitches bound to their cognate ligands have helped to design potential inhibitors with improved drug-like properties [
147,
148,
149,
150,
151,
152,
153]. From the developed compounds, Ribocil (
Figure 2D), which is currently in preclinical development, was shown to inhibit the growth of different bacterial strains, including methicillin-resistant
S. aureus and
Enterococcus faecalis [
63,
154,
155].
Natural riboswitches combine both the sensory and regulatory functions. This principle of direct RNA-ligand interaction was exploited to synthetically design the aptamer-based conditional gene expression systems. Aptamers are single-stranded RNA or DNA molecules that can self-fold in a unique 3D-spatial conformation to specifically interact with their targets. The selection of aptamers with the capability to bind a plethora of different ligands can be performed in vitro through the so-called Systematic Evolution of Ligands by Exponential Enrichment (SELEX) Technology [
156,
157,
158]. Aptamers targeting pathogenic bacteria and viruses have attracted increasing attention [
159]. Such aptamers can be used for the specific recognition of infectious agents or to block their functions [
160,
161].
5. The CRISPR System
Upon a viral or plasmid invasion, bacteria (and archaea) integrate short fragments of foreign DNA into the host chromosome, namely, at a (variable) number of short repetitive
loci (approximately 20–50 base pairs) known as the CRISPR, in a stage called adaptation. These exogenous DNA fragments are inserted by the Cas proteins, Cas 1 and Cas 2, which are the only Cas proteins that are conserved amongst all the CRISPR–Cas systems. The repetitive
loci are subsequently transcribed and processed into a library of short CRISPR-derived RNAs (crRNAs) that are complementary to the previous invading nucleic acids. This is the stage of crRNA expression and biogenesis. Then, comes the interference stage, in which each crRNA can guide the effector nucleases to destroy the foreign genetic material through specific cleavage [
162,
163]. Thus, the integration of invasive DNA constitutes a genetic record of prior encounters with the transgressors, and reflects the surrounding environmental conditions, which change over time.
The CRISPR systems can be divided into two main classes, class 1 and class 2. The class 1 system is found in 90% of the CRISPR
loci in bacteria and archaea, whereas the class 2 systems only represent 10% of the CRISPR
loci that are found in bacteria. The specific types within each class are defined by the effector endonuclease— the Cas protein—which is responsible for cleavage [
164,
165].
The Cas effector proteins are, thus, non-specific nucleases that can be programmed by small guide RNAs, the crRNAs, to be directed to target DNAs or RNAs. Great emphasis has been given to these systems due to these RNA-guided programmable enzymes which exhibit remarkable flexibility in targeting. These have encouraged an ever-expanding array of applications. The most explored and used toolbox in genomic engineering is the class 2 (type II) system, which is better known as CRISPR–Cas9. Cas9 is the characteristic effector protein, and it is essential for immune mechanisms in bacteria [
166]. Furthermore, CRISPR–Cas9 are also abundant in pathogenic and commensal bacteria. Indeed, the
cas9 gene has been reported to play an important role in controlling virulence in various pathogens [
166,
167,
168]. As a virulence regulator, Cas9 is involved in specific steps of the pathogenesis of different bacterial species, as well as in common processes of virulence.
In
Streptococcus sp., Cas9 was reported to influence key regulators of virulence traits, such as adhesion and infection [
169,
170]. The same effect was verified in the knockout strains lacking Cas9 in
N. meningitidis [
171]. Curiously,
cas9 deletion in
Campylobacter jejuni highly affects its sensitivity to antibiotics, regulating several genes that promote antimicrobial resistance [
172]. This proves the connection of CRISPR with antibiotic resistance mechanisms. Interestingly, in the case of
L. pneumophila, Cas2 and not Cas9 is the CRISPR enzyme that is involved in the infection process of macrophages [
173]. Both the Cas9 and Cas2 proteins belong to the same CRISPR–Cas type II system. Although they maintain conserved functions regarding their role in the CRISPR bacterial immunity, they appear to have different functions in virulence, depending on the microorganism.
The interest in the relationship between CRISPR and virulence has grown, and it was later discovered that the CRISPR–Cas type I systems also have an important role in the evasion of bacteria from the host.
Streptococcus mutans contains a class 1-type I CRISPR, whose effector protein is Cas3. In the absence of the
cas3 gene, the strain formed less biofilm, became more sensitive to fluoride, and the expression of the virulence genes was significantly downregulated [
174]. Similar observations have been reported with the
S. enterica isolate 211 [
175]. Additionally, in
P. aeruginosa UCBPP-PA14, the
cas3 gene has been shown to be involved in the achievement of lower pro-inflammatory host responses in cell and mouse models [
176].
Biofilm development and antibiotic resistance are intimately connected since the biofilm matrix can delay the penetration of antimicrobial agents. Biofilm formation is a highly regulated process, and CRISPR has proven to be one of these regulators. Most pathogens involved in nosocomial infections have biofilm-forming abilities. Interestingly, an increased ability to form biofilms has been reported in CRISPR–Cas positive
Enterococcus faecalis and
P. aeruginosa strains [
177]. Additionally, in
Acinetobacter baumannii, specific genes that are involved in biofilm formation appear almost exclusively in strains that are enriched with CRISPR–Cas systems [
178]. It also appears that CRISPR contributes to a tight control depending on the surrounding environment. The lysogenic infection of
P. aeruginosa UCBPP-PA14 by the bacteriophage DMS3 inhibits biofilm formation and swarm motility in a manner that is dependent on the CRISPR regions and
cas genes [
179]. This strategy, by preventing the infected bacteria from forming biofilms and performing other group behaviors, can limit the effects of bacteriophage spread in bacterial communities.
The existence of group behaviors among the bacteria is indeed extremely important. During biofilm formation, bacteria have the ability to communicate with each other through the process of QS. In
Serratia marcescens, it appears that CRISPR–Cas immunity is integrated into the QS circuit, enabling greater defense at higher cell densities [
180]. Similarly,
P. aeruginosa UCBPP-PA14 also uses the QS process to activate
cas gene expression [
181]. Thus, bacteria seem to be able to use QS communication to control CRISPR–Cas expression according to the needs of the cell.
In 2011, Charpentier and co-workers [
182] reported the existence of a
trans-encoded small RNA (tracrRNA) that was transcribed upstream and in the opposite strand of the CRISPR
locus, with 24 nucleotides that were complementary to the repeat regions of the crRNA precursor transcripts (pre-crRNA). This tracrRNA is responsible for pre-crRNA maturation by promoting the cleavage of the tracrRNA-pre-crRNA duplex by the very well-known and widely conserved endoribonuclease RNase III [
182]. Soon after this, tracrRNA was reported to trigger Cas9 to cleave the target DNA [
41]. This discovery enabled the development of a breakthrough method of genome editing, which was later recognized by being awarded the Nobel Prize in Chemistry in 2020 to the scientists, Emmanuelle Charpentier and Jeniffer A. Doudna [
41] (
Figure 1). There is already evidence that ncRNAs related to the CRISPR systems play a role in bacterial virulence. In
Francisella novicida, Cas9 uses a small CRISPR/Cas-associated RNA (scaRNA) to repress an endogenous mRNA transcript encoding a bacterial lipoprotein, which elicits a proinflammatory innate immune response in the host [
168]. A CRISPR-associated ncRNA, RliB, has also been shown to play a role in
L. monocytogenes pathogenesis [
72]. Thus, it appears that ncRNAs constitute an extra layer in CRISPR regulation.
The knowledge of CRISPR has opened avenues to the entire scientific community for the development of genetic engineering tools, namely in the creation of new and improved versions of CRISPR systems that are revolutionizing the world today. As the pieces of the CRISPR puzzle are being discovered, more and more applications are emerging. For instance, in 2014, the use of a type I CRISPR–Cas system in
E. coli enabled the successful removal of individual bacterial strains from mixed populations, which share a high homology. This highlights the extraordinary specificity of this tool, and has opened up the possibility of developing smart antibiotics that prevent multidrug resistance and differentiate between the pathogenic and beneficial bacteria [
183].
These novel antibacterial strategies can be based on CRISPR–Cas systems, primarily on CRISPR–Cas3 and CRISPR–Cas9, to target DNA, which can be designed to specifically eliminate the plasmids that carry antibiotic resistance genes and chromosomal virulence genes, among others, in order to attack the pathogens (
Figure 3A). The tool consists of integrating the CRISPR–Cas sequences into a plasmid vector, allowing the system to target and cut genes of interest. A system that was identified more recently by Feng Zhang’s lab, CRISPR–Cas13 (class 2), brought a new perspective to the CRISPR tool. The RNase Cas13 cleaves single-stranded RNA (ssRNA) molecules in a crRNA-guided manner [
184]. CRISPR–Cas13 also exhibits the promiscuous degradation of ssRNAs when it is performing targetted RNA cleavage, thus, limiting the host cell growth by inducing dormancy in the bacteria [
185]. Additionally, unlike Cas9-based antimicrobials, the CRISPR–Cas13 system exhibits strong bacterial killing activity, regardless of the target genes’ location (chromosome or plasmid) [
186] (
Figure 3A). This system has been successfully tested by constructing antibacterial nucleocapsids (CapsidCas13) that are capable of killing carbapenem-resistant
E. coli and methicillin-resistant
S. aureus through the recognition of the corresponding antimicrobial resistance genes [
186].
Figure 3. CRISPR technologies. (
A)
CRISPR-based antimicrobials. The system has been successfully tested through the directed degradation of the antibiotic resistance gene located in a plasmid (left side) leading to the recovery of the bacterial antibiotic sensitivity or the directed degradation of chromosomal genes, and consequently, cell death (bactericidal) [
186]. (
B) CRISPR-based diagnostics. When CRISPR effector proteins (Cas) recognize the specific target site, their collateral cleavage capability is triggered (this indiscriminate nucleic acid cleavage only happens when the crRNA finds its match). The addition of a reporter, that only releases the signal upon cleavage, enables the emission of a signal that can be easily detected [
187]. Figure created using BioRender.com (accessed on 11 November 2022).
Nevertheless, these CRISPR–Cas tools are still limited in terms of their clinical application due to their delivery systems. The use of conjugative plasmids [
188], phage vectors [
189,
190], membrane vesicles [
191] or their encapsulation into nanomaterials [
192] have been explored as delivery systems.
CRISPR has also received substantial attention as a diagnostic tool due to its potential to detect nucleic acids in a quick, sensitive and specific manner [
187] (example in
Figure 3B). Within the current pandemic context, CRISPR diagnostic technologies were quickly adapted and optimized [
193,
194,
195], being recently highlighted as one of the seven technologies to watch in 2022 [
196] (
Figure 1).
The role of CRISPR–Cas systems in modulating the genotypes, physiology and ecology of bacteria, plus the implication of CRISPR–Cas in limiting horizontal gene transfer, or in enabling the acquisition of advantageous genes are topics of great interest, as is the development of CRISPR for new applications in the area of treatment of infectious diseases. However, the application of CRISPR–Cas antimicrobials remains at a very preliminary stage and numerous obstacles await to be resolved.
6. RNA Technology
6.1. RNA Sequencing (RNA-Seq)
The development of Next Generation Sequencing (NGS), which is also referred to as deep-sequencing, or high-throughput sequencing, has provided a set of diverse modern technologies with applicability to the study of DNA, RNA and proteins [
197]. In particular, RNA-seq methodologies allow for the determination of the sequence of an overwhelming amount of different RNA molecules in a massively parallel way [
197,
198].
Nowadays, there is a panoply of distinct RNA-seq-based approaches that aim to uncover and characterize the RNA species being expressed at each moment in a cell culture or a single cell. Many fields of study have benefited from such methodologies [
199,
200,
201,
202]. In microbiology, RNA-seq derived technologies have been useful as tools for various purposes such as the optimization of bacterial chassis for industrial biotechnology [
203] and synthetic biology [
61,
204], and for the study of both human microbiota [
205,
206] and human pathogens. In this section, we will present some examples of the contribution of different RNA-seq protocols for the study of pathogenic microorganisms (reviewed in [
201,
202]).
In a recent study, messenger RNA sequencing (mRNA-seq) was used to elucidate the function of a specific gene which was postulated to be involved in the virulence of the zoonotic bacterial pathogen
Streptococcus suis type 2 [
207]. In another study, Quant-seq, a variation of mRNA-seq, which is more focused on the 3′-end sequences of polyadenylated RNAs [
208], served to demonstrate that human neural progenitor cells infected by Coxsackievirus B3 change their expression patterns, upregulating antiviral innate immunity and inflammatory pathways during infection [
209].
As already addressed, ncRNAs are crucial regulators. NGS, and particularly small RNA sequencing (sRNA-seq), has largely contributed for the identification of new ncRNAs species in several pathogenic microorganisms [
72,
74,
82,
210].
Moreover, RNA modifications can influence the structure, stability, decoding, and recognition of RNA molecules. They often occur during transcription (e.g., the 5′ NAD cap) or post-transcriptionally (e.g., methylation resulting in N
6-methyladenosine, m
6A), and they may also play a prominent role both in the bacterial stress response and pathophysiology, and in host adaptation [
211,
212]. Combining mass spectrometry (MS) with RNA-seq procedures allows for the precise localization of the RNA modifications and the study of their dynamics [
213]. Remarkably, specific RNA-seq methodologies have been applied to bacterial pathogens to detect the RNA modifications that are crucial for cytotoxicity and virulence, such as NAD capture-seq which measures the NAD incorporation [
214,
215], and m
6A-seq which identifies the methylated residues in the transcripts [
216].
In biology, understanding the network of interactions in the cell is crucial. The RNA interaction by ligation and sequencing (RIL-seq) was designed to identify the RNA–RNA interactions, and this has been particularly useful to elucidate pairs of ncRNAs and their respective mRNA targets. In pathogenic
E. coli, this technique was sufficient to determine the global interactome of RNA molecules binding to Hfq, further detecting ncRNAs that had not been previously annotated [
217]. In turn, gradient profiling by sequencing (Grad-seq) was developed to analyze the native RNA–protein complexomes in the cellular environment. It combines two approaches (RNA-seq and liquid chromatography-tandem mass spectrometry (LC-MS/MS)) [
119], and it can: identify major RNA–protein complexes and RNA binding proteins, cluster ncRNAs according to their biochemical properties, and complement the information regarding the function of domains of uncharacterized proteins [
218]. In fact, thanks to this technology, the ProQ was discovered as an important RNA chaperone, which was a missing piece in the puzzle of ncRNA regulation [
119].
One of the major breakthroughs in this field was the establishment of single-cell RNA sequencing (scRNA-seq), which allows for the discrimination between RNA species being expressed in different cells belonging to the same population or different populations in the same sample (reviewed in [
219]). This technology gained special relevance in enlightening the mechanism of infectious diseases in several pathogens [
220,
221,
222]. Currently, a promising trend in the scRNA-seq approaches is the incorporation of droplet- and microwell-based microfluidics, improving sequencing throughput in an affordable, portable and scalable way [
223].
In the last decade, differential RNA-seq emerged with the advantage of distinguishing between the primary and processed transcripts. This way, it has provided an opportunity to map the transcriptional start sites (TSS), and exposed the existence of pervasive transcription and a generally high abundance of ncRNAs in the bacterial genomes [
224,
225,
226,
227].
Differential RNA-seq served as an inspiration for dual RNA-seq which has the capability of sequencing RNA molecules of two or more species simultaneously [
42] (
Figure 4A). The main goal is to get the best possible approximation to the in vivo conditions (reviewed in [
201,
202]). Although there are still many limitations to overcome, the dual RNA-seq advantages are undeniably evident: it brings the possibility of directly evaluating which genes are differentially expressed in each interacting species which can then be mapped against the known interaction networks or used to predict novel gene regulatory networks [
228]. This tool has been very important for unravelling the mechanisms of infection of several pathogens [
229,
230,
231,
232,
233] (
Figure 1). A surprising example of triple RNA-seq enclosed RNA isolation and sequencing starting from a sample containing human immune cells,
Aspergillus fumigatus (fungus) and Cytomegalovirus (CMV) [
234].
While NGS technologies usually produce short reads, the Third-Generation Sequencing (TGS) has emerged, enabling the sequencing of longer fragments (long reads). As the raw reads can be disclosed in real time, TGS permits data interpretation to occur prior to the samples being fully sequenced [
235]. There are two main TGS categories: single-molecule real-time (SMRT) sequencing, and nanopore single-molecule sequencing (
Figure 4B). Distinctively, nanopore sequencing relies on registering the changes in the electrical current during the translocation of the template molecule along a protein nanopore, rather than recording the optical or chemical signals that are emitted during the polymerization of a complementary strand, as it commonly happens in other RNA-seq techniques [
236]. In the cases where this technology directly uses an RNA molecule as template it may then be called direct RNA-seq. These TGS methods have been of particular relevance for studying pathogenic microorganisms to further disclose the link between post-transcriptional RNA modifications and microorganisms’ mutability and virulence [
237,
238], as well as to characterize transcript isoforms [
239,
240,
241].
Overall, when they are compared with first-generation sequencing (Sanger sequencing), the NGS and TGS methods are faster, more sensitive and produce a greater amount of data encompassing a wide repertoire of RNA molecules [
236]. The employment of NGS and TGS in the meta-transcriptomics through whole-genome or full-length 16S rRNA sequencing has already been shown to accelerate the diagnosis of infectious diseases, namely, by reducing the waiting time, improving the pathogen taxonomic classification and the effectiveness in the detection of RNA viruses, and by extending the spectrum of antibiotic resistance genes that are detected in clinical samples [
242,
243].
Finally, the above-mentioned RNA-seq strategies might help in the identification of diagnostic biomarkers, the choice of the appropriate treatment for different severity stages of a certain disease, of drug target candidates and potential drugs which can also be repurposed and used for the efficient treatment of specific infectious diseases [
244,
245].
In fact, independently of the specific RNA-seq method that is employed, it will always require bioinformatic pipelines to process the enormous volume of data. In the past, programming skills were a prerequisite, but many tools with graphical user-friendly interfaces have been progressively developed and made accessible for everyone, as it is the case of the Galaxy platform [
246]. Many online resources are also available, namely, several specific transcriptome browsers, or simply, brief explanations of the different techniques, protocols and data.
Figure 4. (
A)
Simplified workflow of a dual RNA-seq protocol. Host cells are infected in vitro with pathogen cells, lysed and total RNA is extracted. The sequencing library is prepared, and sequencing is performed in a NGS platform, obtaining simultaneously the results for both species. During bioinformatic data analysis, after quality control and data cleaning, the reads from the host and the pathogen are separated in silico in the mapping step. Annotation and quantification are carried out independently for each species, allowing to analyze host and pathogen differential gene expression in parallel, as well as to predict functional correlations between species [
230]. (
B)
Main categories of third-generation sequencing (TGS). (Left panel) Single-molecule real-time (SMRT) sequencing—Sequence is determined through emission of fluorescence due to the incorporation of a fluorescently labelled deoxyribonucleotide (dNTP) by the DNA polymerase in the nascent complementary strand of the cDNA template molecule. The DNA polymerase is anchored to the bottom of a nanowell. (Right panel) Nanopore sequencing—Sequence is obtained without imaging. The template nucleic acid is bound to a motor protein which takes the molecule to a protein nanopore. When the template molecule is translocated through the pore, each nucleotide with its own modifications produces a characteristic current shift that is recorded. Unlike the other methods, direct RNA-seq uses an RNA molecule as template [
236]. (
C)
Antisense oligonucleotides (ASOs) mechanism. (Left panel) General mechanism of ASOs activity. The oligonucleotide binds to the complementary RNA, impairing ribosome progression and/or causing transcript cleavage of a target duplex of mRNA/ASO by RNase H. (Right panel) Targeting of ncRNA–mRNA interaction. In this case, the ASO can be designed to mimic the ncRNA and block its binding to the mRNA (anti-mRNA ASO) or mimic the mRNA sequence to sequester the ncRNA (anti-ncRNA ASO) [
247]. (
D)
mRNA vaccines mechanism. The nucleoside-modified mRNA containing the coding sequence of the protein of interest (SARS-CoV-2 Spike protein) is encapsulated in a lipid nanoparticle (LNP). Upon human vaccination, the LNP is internalized, and the mRNA coding sequence is recognized by the host translation machinery, leading to the production of Spike proteins. This will induce the production of specific antibodies by the host immune system, inducing an immune response cascade [
248]. Figure created using BioRender.com (accessed on 11 November 2022).
6.2. ASOS—The Use of Antisense Antimicrobial Therapeutics
An alternative strategy to fight the growing antibiotic resistance phenomena is to design gene-specific oligomers that can specifically target any single pathogen. Antisense antimicrobial therapeutics are a biotechnological form of antibiotic therapy using short, single-stranded oligomers that mimic the structure of DNA or RNA and bind to specific, complementary RNA in a target organism [
249,
250]. In microorganisms, ASOs (antisense oligonucleotides) bind to their complementary mRNA and inhibit its translation into proteins through the steric blockage of the ribosome progression and/or by promoting the degradation of the targeted mRNA through the RNase degradation of the ASO/mRNA duplex [
250].
A key advantage of this antisense approach is that ASOs can be rationally designed to target any microbe through sequence complementation, thus, significantly enlarging the available selection of potential therapeutic targets [
249]. A main goal in ASO design is the achievement of high specificity with minimal off-target effects. The sequence specificity and the short length of the antisense antimicrobials pose a minimal risk to human gene expression. Moreover, the specificity of antisense antimicrobials avoids the non-selective killing of the beneficial commensal bacteria by broad-spectrum antibiotics. This overcomes the unintended side-effects that are caused by the dysbiosis of the microbiome, and the consequent medical complications.
The use of antisense therapeutics has been progressively advancing towards clinical use, but in recent years the field has been accelerating. The identification of essential genes and the number of sequenced genomes has largely contributed to this. However, despite the fast advances in the eukaryotic fields [
251], the progress in the use of ASOs as antibacterials has been delayed due to the poor uptake efficiency of the antisense molecules by bacteria [
249]. This is mainly due to the electrostatic charge or the size barrier that is imposed by the cellular envelope (plasma membrane and cell wall). Other challenges regarding ASO efficiency are its intracellular concentration, oligomer length, nuclease resistance and binding kinetics.
ASOs are typically 10–30 nucleotides in length. The cellular nucleases rapidly attack the unmodified ASOs. Therefore, numerous chemical modifications have been described (e.g., phosphorothioates, locked nucleic acids, peptide nucleic acids, and phosphorodiamidate morpholino oligomers) to confer resistance against nucleases, to improve the stability of the ASO/mRNA hybrid formation and/or to preserve the target specificity.
In the sense of overcoming the challenge of bacterial cellular uptake, the most common strategy for facilitating antisense oligonucleotide delivery is the conjugation of a cell-penetrating peptide (CPP) to the antisense oligonucleotide. The attachment of a compound that can penetrate the bacterial cell wall facilitates the delivery of synthetic antisense oligomers into the bacterial cytoplasm. CPPs are short cationic or amphipathic peptides, which are usually composed of less than 30 amino acids. CPPs have been used with success to deliver modified ASOs in different bacteria ([
252] for a review).
Phosphorodiamidate morpholino oligomers (PMOs) are synthetic single-stranded oligomers with a modified backbone which makes them resistant to nucleases [
250]. The use of CPP-PMOs has been effective against infections caused by antibiotic resistant bacteria of the genus
Acinetobacter (
A. lwoffii and
A. baumannii) and
Klebsiella pneumoniae [
253]. Wesolowski et al. described a CPP-PMO conjugate that targeted
E. coli gyrA, a highly conserved gene that is found across multiple bacterial species [
254]. The authors show that
gyrA CPP-PMO reduced the viability of both the Gram-positive and Gram-negative bacterial strains (
Enterococcus faecalis,
Staphylococcus aureus).
GyrA mRNA was also targeted in
S. pyogenes, but it used a CPP-PNA. Peptide nucleic acids (PNAs) are constructed by attaching bases to a modified polyamide backbone. The PNAs are uncharged, which in part accounts for their high affinity for RNA [
255]. Successful examples of PNA targeting in different bacteria have been described [
256,
257]. In the foodborne pathogen
C. jejuni, the
cmeABC operon encodes a multidrug efflux pump that confers resistance to a broad range of antibiotics [
258]. The use of PNAs targeted to different regions of the
cmeABC operon restored the antibiotic susceptibility [
259].
Locked nucleic acids (LNAs) are oxyphosphorothioate analogues with a 2′-O,4′-C-methylene bridge that locks the ribose ring in the C3′-endo conformation [
260]. Both the CPP-PNAs and CPP-LNAs have been used in
S. aureus to target the
ftsZ mRNA, a gene that is required for cell division [
261,
262].
As it is mentioned in the previous sections, RNase E is an essential enzyme that is highly conserved in Gram-negative bacteria, and it has no known human orthologue [
2]. Thus, the
rne gene is a good target for antisense antibiotic development. Using
E. coli as a model, Goddard and colleagues have used LNA gapmers, oligonucleotides consisting of a central region of DNA that is flanked by regions of chemically modified LNA nucleotides, to target RNase E [
263]. Using this antisense antibiotic strategy, the authors were able to block the translation activity and trigger the RNase H-mediated cleavage of the
rne mRNA in vitro, introducing the way to the use of this novel anti-bacterial target in different pathogens (
Figure 4C, left panel).
Beyond the targeting of essential genes to reduce the viability of the pathogens, an alternative strategy for using antisense antibiotics is to target non-essential genes, which are required for virulence. Some examples of these are the genes required for invasiveness, biofilm formation [
264], and antibiotic resistance genes. In this latter case, the co-administration of the PMO with the antibiotic would restore the susceptibility of the bacteria to its administration [
264].
There are also other levels through which ASOs can reprogram the gene expression. For instance, ASOs can target the regulation by ncRNAs over their mRNA targets. In this case, the ASO can be designed to mimic the ncRNA and block its binding to the mRNA target (anti-mRNA ASO) or mimic the mRNA sequence to sequester the ncRNA (anti-ncRNA ASO). In both cases, the ncRNA–mRNA interaction is impaired (Figure 4C, right panel).
Henderson and co-workers [
247] designed PNAs to target the ncRNA–mRNA interactions related to a QS system in
V. cholerae. The Qrr ncRNAs are composed of four redundant regulators that target, among other genes, the
hapR mRNA. At a low cell density, the expression of Qrr ncRNAs represses the master regulator HapR to promote the host colonization and virulence factor production in this human pathogen. At a high cell density, attained at later stages of the infection, the Qrr ncRNAs are no longer expressed, thus reactivating HapR expression and causing the release of the bacterium from the host. The use of two CPP-PNAs designed to sequester the Qrr ncRNAs (anti-Qrr ncRNA ASOs) prevented the Qrr-
hapR mRNA interaction. This impaired the HapR downregulation, locking it in the HapR expression state (high cell density profile), with antibacterial implications.
The specific inhibition of a riboswitch by an ASO lead to the inhibition of the growth of
S. aureus, L. monocytogenes and
E. coli, which widen the lists of possible targets of this antimicrobial alternative system [
265]. The potential of the applications of the different types of chemically modified ASOs and the creation of new and improved carrier compounds will expand their uses in multiple pathogenic bacteria.
6.3. RNA Interference (RNAi)
RNA interference (RNAi) is a biological process in which small ncRNAs recognize a specific mRNA, thereby promoting their degradation by Argonaute proteins, thus leading to gene silencing. This eukaryotic mechanism works as an innate defense mechanism against invading viruses [
266]. The RNAi system was first described in 1998, and its important role in gene regulation rendered a Nobel Prize to Andrew Fire and Craig Mello [
43] (
Figure 1). Soon after their discovery, small interfering RNAs (siRNAs) were explored as a tool to treat several diseases, including viral infections [
267,
268]. The use of siRNA as a therapeutic agent implies the delivery of these molecules into the target cells, thereby activating the RNAi mechanisms in order to silence a specific gene. siRNAs have a high degree of specificity, targeting a unique mRNA, they have reduced toxicity and can reach inaccessible targets. The use of siRNAs to inhibit the replication of SARS-CoV [
269], SARS-CoV-2 [
270], respiratory syncytial virus (RSV) [
271], and hepatitis C virus [
272] has already been demonstrated, and this validates the potential of these molecules for the treatment of viral infections. The siRNA molecules target regions of the viral genome that are important for replication, such as mRNA that codes for the spike protein from SARS-CoV-2 [
270], or the mRNA that codes for the nucleocapsid protein from RSV [
271]. There are still some limitations for the use of siRNA-based therapies, such as siRNA stability, effective carriers, delivery routes and off-target effects. Regardless, clinical trials have already been performed with siRNA-based drugs to treat Ebola and RSV infections ([
273], and reviewed in [
274]).
6.4. mRNA Vaccines
Vaccination continues to be the most successful and cost-effective public health intervention to control and prevent infectious disease outbreaks. In fact, the conventional application of inactivated, live-attenuated or subunit vaccines had enormous success in the eradication of several infectious diseases, with a classic example being the complete eradication of the smallpox virus; however, many others were not as efficient in treating human immunodeficiency virus (HIV),
M. tuberculosis and
Plasmodium spp. [
275] and other common vaccine-preventable diseases such as influenza [
276].
Despite the promising results in the mRNA therapeutics field [
36,
277], mRNA was seen as too unstable and expensive to be used as a drug or a vaccine for several years [
278]. A landmark experience was performed by Robert Malone in 1989 when he discovered the possibility of transfecting mRNA into eukaryotic cells which would induce their intracellular translation, thus recognizing the potential of exploring the RNA molecule for therapeutic purposes [
279]. A year after this, the same principle was successfully applied in vivo [
36]. In the 1990s, mRNA was tested as a therapeutic agent for the first time using lipid nanoparticles (LPNs) as the delivery method [
277,
280,
281] (
Figure 1). At that point, the challenges were to overcome RNA instability, to control the excessive host inflammatory responses and also to improve in vivo delivery systems. Katalin Karikó and Drew Weissman were central players in this context. They unraveled that the incorporation of modified, naturally occurring nucleosides in the mRNA molecules, particularly pseudouridine, prevents the activation of the immune response, reducing the synthetic mRNA immunogenicity in vivo [
282] and provides a higher translation capacity [
38,
283] (
Figure 1). More recently, it was demonstrated that N
1-methylpseudouridine could provide even better results [
284]. In addition, LNPs have become one of the most appealing and commonly used mRNA delivery tools [
285].
In face of a sudden new coronavirus pandemic, previous advances in mRNA technology have enabled the rapid release of two highly efficacious mRNA vaccines in the market, BNT162B2 by Pfizer-BioNTech [
248] and mRNA-1273 by Moderna [
286]. Both of them are LNP-formulated nucleoside-modified RNA vaccines that encode the spike protein of SARS-CoV-2 as the target antigen (
Figure 4D). These were the first mRNA-based vaccines to gather an emergency FDA approval, and their success in providing a robust immune response against SARS-CoV-2 was a game changing in the world of immunology and vaccine development (
Figure 1). An important point to make is that the speed at which the COVID-19 vaccines were developed was influenced by a global emergency that resulted in an unseen alliance of the scientific community and in a massive funding.
For HIV, since the virus was reported in 1981, many unsuccessful attempts to produce a vaccine were announced [
275,
287,
288,
289], but Moderna has currently two mRNA vaccine candidates which are in Phase 1 clinical studies [
290]; for tuberculosis and Malaria, the BioNTech company has announced that it is planning to move forward with the clinical trials of mRNA formulations for both of the diseases; for the influenza virus, tremendous effort has been invested in improving the current vaccines, and it is believed that the mRNA platform is well positioned to address the significant unmet need in the season flu [
291,
292,
293]. Finally, the application of mRNA-based therapeutics is also being evaluated for other priority diseases by the CureVac and Moderna companies, such as Rabis, Respiratory Syncytial Virus, Human Cytomegalovirus, Human metapneumovirus and parainfluenza virus, Zika virus, Epstein–Barr Virus, Nipah virus and Chikungunya Virus. For the Ebola Virus and for
Streptococcus sp. infections, preliminary mRNA vaccine studies in animal models are already being developed [
294,
295].
With increased scientific interest in this area, the next-generation mRNA technology will continue to mature both for vaccine development and therapeutics. The field of nucleotide-based vaccines came to the spotlight as a novel, faster and cheaper way to achieve vaccine development when compared with the conventional technologies [
296]. Nevertheless, improvements in the storage and stability, production costs, geographic distribution capacity and research alliances are essential to ensure a more effective and prompt response to fight current and future endemic and/or pandemic infectious diseases.
7. Conclusions
RNA is back in the spotlight. The diverse role of RNA in all biological processes, together with the recognition of its important functional properties, have led to its exploitation in a wide range of biotechnological and medical applications. A great contributor to this change of perspective was the validation of the mRNA vaccines at an unpredictable scale and speed at which they fought against the COVID-19 pandemic. As they are natural molecules, RNAs present, in general, low toxicity and immunogenicity. The use of RNA elements presents advantages such as its independent control, tunability, composability and portability which empower their use as genetic tools. However, the advances in the application of these RNA tools have been limited by the rhythm of the progression of the technological advances which have enabled the characterization of new molecules and biological mechanisms. Built upon decades of scientific research, robust and prompt RNA technologies have now emerged, highlighting the importance of fundamental and applied research. For instance, we have testified in recent years, a fast discovery for the new classes of RNA molecules and molecular mechanisms that have transformed our comprehension of RNA metabolism. This review puts together the major discoveries regarding the connection between RNA metabolism and pathogenesis, and how this knowledge has been used to create new strategies to fight microbial pathogenicity. Antibiotic resistance is a serious problem that requires the creation of alternative therapeutics. As such, several RNA tools have surfaced as alternatives to control the virulence of pathogenic bacteria, namely, using synthetic non-coding RNAs, antisense antimicrobial therapeutics with antisense oligonucleotides (ASOs) or CRISPR–Cas antimicrobials. The application of these tools in prokaryotic organisms has been limited by different obstacles. In the case of ASOs, their use has been mostly limited by the development of delivery systems to improve their uptake by the bacterial cells. The same has happened with the CRISPR–Cas tools, and despite the new delivery systems which have been used with success, more research is needed to assure their safety and effectiveness. In the case of the mRNA vaccines, their implementation was possible thanks to the curiosity-driven studies of lipids and experiments with synthetic mRNA. The establishment of mRNA vaccines seems promising because of the speed with which they can be developed and produced, and their flexibility and adaptability to variants.