1. Introduction
Among the seven members of the human coronavirus (HCoV) family, human coronavirus 229E (HCoV-229E), human coronavirus HKU1 (HCoV-HKU1), human coronavirus NL63 (HCoV-NL63), and human coronavirus OC43 (HCoV-OC43) usually cause less severe complications and are responsible for approximately one-third of common colds [
1,
2]. In contrast, severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1), Middle East respiratory syndrome (MERS-CoV), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can cause severe respiratory diseases [
3,
4,
5]. In particular, COVID-19, which was caused by SARS-CoV-2, represents the most devastating pandemic in recent human history.
HCoV-229E belongs to the alphacoronaviruses [
1]. This virus was first identified in humans in the 1960s and HCoV-229E-like coronaviruses have recently been identified in bats and camels [
6,
7]. HCoV-229E infections are usually mild and, in most cases, asymptomatic. However, more severe upper and lower respiratory tract infections can be common in infants, elderly individuals, and immunocompromised patients and clusters of HCoV-229E infections with pneumonia have been associated with otherwise healthy adults [
1,
8,
9]. Thus, efforts in developing novel strategies against HCoV-229E should not only help in the treatment of HCoV-229E-associated diseases but also human infections and the diseases associated with other coronaviruses such as SARS-CoV-2.
The nucleocapsid (N) protein is indispensable for coronaviral replication, which functions as a structural protein participating in viral genomic packing, replication, and cell signaling pathways [
10]. Being one of the most conservative functional proteins among coronavirus evolution, the N protein gene is more stable and conserved than the S protein, with more than 85% amino acid homology and less mutations over time [
11,
12]. The N protein is considered a promising target for the discovery of antiviral drugs because of its evolutionary conservation and pivotal role in viral replication.
Sequence-specific ribozymes with efficient endoribonuclease activity, including those derived from the catalytic RNAs of ribonuclease P (RNase P), represent a novel class of gene-targeting agents for therapeutic applications against human viruses, by knocking down viral gene expression and replication [
13,
14,
15]. The biological function of RNase P, which consists of a catalytic RNA subunit called M1 RNA and a protein subunit called a C5 protein in
Escherichia coli, is to hydrolyze a tRNA precursor (pre-tRNA) during tRNA 5′ end maturation (
Figure 1A) [
16,
17]. M1 RNA, which is 377 nucleotides in length, can cleave a pre-tRNA substrate in vitro in the absence of a C5 protein [
18]. Extensive studies have revealed the three-dimensional structure of M1 RNA, including its active site and substrate binding site [
17,
19]. Biochemical studies revealed that a stem structure resembling the acceptor and T stems of tRNA can be recognized as a substrate and cut by M1 RNA (
Figure 1A,B) [
20,
21]. Thus, M1 RNA can be engineered into a sequence-specific ribozyme, M1GS, by linking a guide sequence (GS) that base-pairs to a target RNA including a viral essential mRNA, such that the resulting GS-target RNA complex resembles a stem structure recognizable by the catalytic RNA (
Figure 1C) [
22]. Several studies have shown that engineered RNase P ribozymes can efficiently cut human and viral mRNAs, and knock down gene expression and the production of various human viruses such as herpes simplex virus 1 (HSV-1), human cytomegalovirus (HCMV), hepatitis B virus (HBV), and human immunodeficiency virus (HIV-1) in vitro and in cell culture models [
23,
24,
25,
26]. However, whether M1GS ribozymes can be used against human coronaviruses, such as HCoV-229E, has not been reported.
We reported here the engineering of an RNase P ribozyme to target the mRNA coding for the HCoV-229E N protein, which also overlaps with the full length viral genomic RNA and other small genomic RNAs (sgRNAs) due to their sharing a common 3′ transcription termination site [
1,
27]. The constructed functional ribozyme, M1-N-F, efficiently cut the N target sequence in vitro. HCoV-229E-infected cells expressing the M1-N-F ribozyme exhibited a reduction of about 85% in viral RNA levels and N protein expression, and a reduction of about 750-fold in infectious particle production, compared to cells expressing no ribozymes or a control, catalytically inactive ribozyme. Our study provides the first direct evidence of the therapeutic potential of RNase P ribozymes against human coronaviruses such as HCoV-229E.
2. Materials and Methods
Cells, viruses, and antibodies. Human foreskin fibroblasts (HFFs) and HCoV-229E were purchased from Lonza (Hayward, CA, USA) and the American Type Culture Collection (ATCC) (Manassas, VA, USA), respectively. They were maintained in a DMEM containing 10% fetal bovine serum. We followed the previously described protocols for infection and the propagation of HCMV in these cells [
23,
28]. Anti-CoV-229E N protein and anti-actin antibodies were purchased from Sino Biological Inc (Wayne, PA, USA) and Sigma (St Louis, MO, USA), respectively.
Synthesis of ribozyme and in vitro assays. We followed the previously described procedures [
23,
28] to determine the exposed regions of the N mRNA to be modified by dimethyl sulphate (DMS) [
22,
29,
30]. Cells were infected with HCoV-229E at 18 h and then treated with a DMEM in the absence of fetal bovine serum and in the presence of DMS for 10 min. Total RNAs were isolated and those DMS-modified regions of the N mRNA were mapped by a primer extension analysis as described previously [
23,
28].
Ribozyme M1-IE has been described previously [
28]. PCR was used to generate the DNA coding for the functional M1GS (i.e., M1-N-F) and the control M1GS (i.e., M1-N-C) from the M1 RNA sequence-containing construct pFL117 and the mutant C102 sequence-containing construct pC102, respectively [
31,
32], with 5′ primer Rb-N5-AF25 (5′-GGAATTCTAATACGACTCACTATAG-3′) and 3′ primer Rb-N3 (5′-CCCGCTCGAGAAAAAATGGTGATGCATCTGAACCACAATGTGGAATTGTG-3′). PCR was also used to generate the DNA coding for substrate n-39 with 5′ primer n5-AF25 (5′-GGAATTCTAATACGACTCACTATAG-3′) and 3′ primer n3 (5′-CGGGATCCGTTGTGGTTCAGATGCATCAGCCCATTTGTCTATAGTGAGTCGTATTA-3′). In vitro transcription procedures with T7 RNA polymerase were applied to synthesize the ribozymes and substrate n-39 for in vitro studies [
31,
32].
The in vitro cleavage of substrate n-39 by ribozymes was assayed in buffer A (50 mM Tris, pH 7.5; 100 mM NH
4Cl, 100 mM MgCl
2) and the overall cleavage rates (k
cat/K
m)
s were determined, following the procedures described previously [
28,
31]. Briefly, a trace amount of [
32P]-labeled substrate n-39 was mixed with an excess amount of ribozymes under single turnover conditions in buffer A at 37 °C. The concentrations of substrate n-39 were less than 0.1 nM and the ribozyme concentrations were from 0.8 to 100 nM. We showed that the cleavage reaction under these conditions followed pseudo-first-order kinetics and, at fixed excess ribozyme concentrations, the observed cleavage rate (k
obs) was not affected by variants in the amount of substrate [
28,
31]. At different time points, the reactions were stopped, and the mixture was loaded on 10% polyacrylamide gels containing urea. The substrates and cleavage products were separated, visualized, and analyzed with a Storm 840 Phosphorimager using ImageQuant software (Molecular Dynamics, Mountain View, CA, USA) [
28,
31].
The binding affinities of the ribozymes to substrate n-39 were studied in vitro in buffer E (50 mM Tris, pH 7.5; 100 mM NH
4Cl, 100 mM CaCl
2, 3% glycerol, 0.1% xylene cyanol, 0.1% bromophenol blue) and the equilibrium dissociation constants (K
d) were determined as described previously [
28,
31]. We preincubated various concentrations of the ribozyme in buffer E for 15 min before mixing with an equal volume of substrate n-39, preheated under identical conditions. The samples were incubated for 15–30 min to allow binding, then separated on a 5% polyacrylamide gel with running buffer F (100 mM Tris-Hepes, pH 7.5, and 10 mm MgCl
2). We obtained the K
d values by extrapolation from a graph plotting the percentage of the substrate bound versus the ribozyme concentration [
28,
31]. The (k
cat/K
m)
s and K
d values were the average of three separate experiments.
Generation of ribozyme-expressing cells. We cloned the sequences coding for ribozymes M1-N-C, M1-N-F, and M1-IE into the expression vector pUC7SL to generate constructs pM1-N-C, pM1-N-F, and pM1-IE, respectively. Construct pUC7SL was derived from pUC19 and contained the expression cassette driven by the 7SL RNA promoter [
33]. HFFs (1 × 10
6 cells) were first transfected with the same amount (e.g., 50 µg) of plasmid DNAs containing the ribozyme coding sequences, then selected using neomycin (600 µg/mL) (Invitrogen, Carlsbad, CA, USA), and eventually cloned [
24,
28]. We applied Northern blot analyses to examine ribozyme expression [
31,
32]. Briefly, total RNAs were purified from the cells, separated in 1.5% agarose gels containing formaldehyde, electrically transferred to a membrane, and hybridized with [
32P]-radiolabeled probes containing M1 and H1 RNA sequences. The Northern blot results were analyzed with a STORM 840 Phosphorimager using ImageQuant software (Molecular Dynamics, Mountain View, CA, USA) to quantify the levels of ribozymes [
31,
32]. We conducted quantitation in the linear range of RNA detection.
We assessed the cytotoxicity associated with ribozyme expression using an MTT assay (Sigma, St Louis, MO, USA). We grew cells in 96-well plates, added 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT; Sigma) (5 mg/mL in PBS) to each well, and measured absorbance at 570 nm with a microplate reader to assay cell viability, following the manufacturer’s recommendations. We conducted experiments in duplicate and repeated them three times. We also examined cell morphology under a Nikon TE300 microscope at different times.
Assays of HCoV-229E and HCMV gene expression. Cells (5 × 10
5) were either infected with HCoV-229E and HCMV (MOI = 0.5–1) or not. RNA and protein samples were collected at time points postinfection as indicated in the Results Section [
23]. Western blot analyses were applied to quantify the N protein levels. We separated the protein samples from the cell lysates on 9% (
v/
v) SDS-polyacrylamide gels cross-linked with N,N′-methylenebisacrylamide, electrically transferred them to nitrocellulose membranes and reacted the membranes with anti-N and anti-Actin antibodies, followed by secondary anti-mouse antibodies conjugated with alkaline phosphatase. The membranes were then stained with a chemiluminescent substrate with the aid of a Western chemiluminescent substrate kit (Thermo Fisher, Waltham, MA, USA), and analyzed with a STORM 840 PhosphorImager using ImageQuant software (Molecular Dynamics, Mountain View, CA, USA) to quantify the protein levels [
23,
28]. We conducted the quantitation in the linear range of protein detection.
The HCoV-229E RNA and HCMV IE1 mRNA levels were quantified by qRT-PCR amplifying the HCoV-229E N mRNA region and the HCMV IE1 mRNA region, respectively. An iScriptTM cDNA synthesis kit (Bio-Rad, Hercules, CA, USA) was used to generate the cDNAs for the two-step reverse transcription quantitative PCR (RT-qPCR) assay following the manufacturer’s recommendations. Each 20 µL of the reverse reaction contained 4 µL of iScript Reaction Mix, 1 µL of iScript transcriptase, 5 µL of RNA template (~0.5 µg), and 10 µL of DEPC-treated water. The mixture was incubated in a thermal cycler using the following steps: at 25 °C for 5 min for priming, at 46 °C for 20 min for reverse transcription, at 95 °C for 1 min to inactivate the reverse transcriptase, and then held at 4 °C.
After the reverse transcription step, quantitative real-time PCR analyses were performed with the SsoAdvanced Universal SYBR Green Supermix (Bio-Rad, Hercules, CA, USA) and the PCR was run on a C-1000 Touch Thermal Cycler with the CFX96TM Real-time System (Bio-Rad, Hercules, CA, USA). Each 20 µL of a real-time PCR reaction contained 10 µL of 2 × SsoAdvanced Universal SYBR Green Supermix, 2.5 µL of 2.8 µM of a forward primer, 2.5 µL of 2.8 µM of a reverse primer, and 5 µL of a reverse transcription product (~10 ng cDNA). After one cycle at 98 °C for 30 s to activate the polymerase and denature the cDNA, 40 cycles of denaturation (98 °C for 10 s) and annealing/extension/plate reading were performed to quantify the HCoV-229E RNAs and HCMV IE1 mRNA.
The primers for the HCoV-229E N mRNA analysis were N-5 (5′-AGGCGCAAGAATTCAGAACCAGAG-3′) and N-3 (5′-AGCAGGACTCTGATTACGAGAAAG-3′) [
34]. The primers for the HCMV IE1 mRNA analysis were IE1-5 (5′-TGACCGAGGATTGCAACGA-3′) and IE1-3 (5′-CCTTGATTCTATGCCGCACC-3′) [
35]. The actin mRNA level was used as the control and quantified by qRT-PCR with 5′ primer Actin5 (5′-TGACGGGGTCACCCACACTGTGCCCATCTA-3′) and 3′ primer Actin3 (5′-CTAGAAGCATTGCGGTGGCAGATGGAGGG-3′) [
36].
Assays of HCoV-229E and HCMV growth. Cells (n = 5 × 10
5) were infected with HCoV-229E at an MOI of 0.5–2. We collected the entire culture (i.e., the cells and medium) samples every day during a seven-day time-course study. Viral stocks were prepared from the collected samples and titered by infecting human foreskin fibroblasts with these samples and quantifying the plaques, following previously described protocols [
23,
28]. The growth of HCMV in the ribozyme-expressing cells was assessed as reported previously [
23,
28]. The values obtained were the average from three independent experiments.
Statistical Analysis. Assays were performed in duplicate and repeated independently three times. A data analysis was conducted with the analysis of variance (ANOVA) (GraphPad Prism software, version 10). A p value of <0.05 was viewed as statistically significant.
3. Results
Cleavage and binding of the HCoV-229E N mRNA sequence by RNase P ribozymes in vitro. The accessibility and secondary structure of an N mRNA sequence may substantially affect the cleavage efficiency of this mRNA sequence by RNase P ribozymes because these factors can influence ribozyme binding. To identify the N mRNA regions potentially exposed to ribozyme binding, dimethyl sulphate (DMS) was used to treat human foreskin fibroblasts infected with HCoV-229E. In addition, we also considered those N mRNA regions that are highly conserved among human coronavirus sequences [
1,
27]. The position of 22 nucleotides downstream from the translation initiation site of the N mRNA became our choice for the ribozyme cleavage site. Our experiments showed that this region was substantially modified by DMS. Moreover, this RNA region is highly conserved and is near the translation initiation site, which needs to be exposed to the ribosome for translation, and therefore is likely accessible for ribozyme binding [
1,
27].
We included three ribozymes in the current study. The first ribozyme is the functional ribozyme M1-N-F, in which M1 RNA was linked at its 3′ terminus to an 18nt long guide sequence targeting the HCoV-229E N mRNA sequence. The second ribozyme is the control ribozyme M1-N-C, in which the same guide sequence in M1-N-F was linked to an inactive M1 RNA mutant C102 [
32] that had mutations abolishing its catalytic activity. The inclusion of M1-N-C is to understand the effect of the catalytic activity of the functional ribozyme M1-N-F on targeting the HCoV-229E N mRNA region and to provide a control for the antisense effect of the ribozyme-mediated targeting since both M1-N-C and M1-N-F contained the same guide sequence. The third ribozyme, M1-IE, was previously constructed to target the overlapping mRNAs coding for the immediate-early proteins IE1 and IE2 of the human cytomegalovirus (HCMV), which are required for HCMV gene expression and replication [
37]. M1-IE efficiently cleaved the IE1 and IE2 mRNA sequences in vitro and inhibited IE1 and IE2 expressions in the cultured cells [
28]. The inclusion of M1-IE is to determine the targeting effect of the M1GS ribozyme on the N mRNA region and HCoV-229E infection when the ribozyme had an incorrect targeting sequence that targeted another target RNA but not the N mRNA region of HCoV-229E.
The in vitro cleavage experiments showed that M1-N-F efficiently cut substrate n-39, a 39 nt long RNA covering the target N mRNA sequence. The overall cutting rate (k
cat/K
m)
s was measured as 0.32 ± 0.07 µM
−1·min
−1 (
Table 1), indicating that M1-N-F exhibited excellent catalytic activity in hydrolyzing substrate n-39 in vitro. The apparent binding of substrate n-39 by M1-N-F was detected in our gel-shift assays. The value of K
d, which represents binding affinity, was measured as 0.25 ± 0.06 nM (
Table 1). In contrast, a lack of cutting of substrate n-39 by M1-N-C was found, because this control ribozyme was not catalytically active due to the mutations in the M1 RNA sequence. However, the binding affinity of substrate n-39 by control ribozyme M1-N-C (K
d = 0.22 ± 0.06 nM) was comparable to that of functional ribozyme M1-N-F (
Table 1). Thus, M1-N-C served as the antisense effect control because this M1GS RNA did not possess catalytic activity but bound to n-39 as strongly as M1-N-F.
Expression of ribozyme in human cells. The DNA sequences coding for ribozymes M1-N-F, M1-N-C, and M1-IE were cloned into the expression plasmid pUC7SL and under the control of the 7SL RNA promoter to generate constructs p7SL-M1-N-F, p7SL-M1-N-C, and p7SL-M1-IE [
33], which were used for the expression of ribozymes M1-N-F, M1-N-C, and M1-IE, respectively, in human cells. Human foreskin fibroblasts were transfected with expression vector pUC7SL containing the M1GS DNAs (i.e., p7SL-M1-N-F, p7SL-M1-N-C, and p7SL-M1-IE) and an empty pU7SL vector without any ribozyme sequences, and the cell lines containing the empty vector or the ribozyme sequences were generated and cloned [
24,
28].
As shown in the Northern blot analyses, ribozymes M1-N-F, M1-N-C, and M1-IE of ~400 nucleotides were present in the constructed cell lines, with human H1 RNA expression (~350 nucleotides) as the loading control (
Figure 2). An examination of these constructed cell lines and the parental cells revealed that all these cells exhibited similar viability and growth in a culture for two months and therefore suggested little cytotoxicity associated with the presence of ribozymes.
Suppression of HCoV-229E RNA level and N protein expression by M1-N-F. Two series of experiments were performed to investigate the effects of ribozymes on HCoV-229E gene expression. In the first series of experiments, cells were infected with HCoV-229E and total RNAs were isolated at 24 h postinfection. Because all the sgRNAs and full-length genomic RNA completely overlap with the N mRNA, a qRT-PCR assay amplifying a specific region of the N mRNA sequence should quantify the combined total levels of all these RNAs [
1,
27]. Compared to the cells with the empty vector without any ribozyme sequence, the total levels of all these viral RNAs were similar in the cells with control ribozymes M1-N-C and M1-IE but reduced by about 85% in the cells with functional ribozyme M1-N-F (
Figure 3). These results revealed that M1-N-F decreases the levels of viral RNAs, which contained the ribozyme-targeting sequence. Moreover, these data suggested that the decrease of the viral RNA levels is due to the catalytic activity of M1-N-F, since no reduction was found in the cells with control ribozyme M1-N-C, which had the identical guide sequence to target the N RNA region of HCoV-229E as functional ribozyme M1-N-F but was not active catalytically.
In the second series of experiments, viral protein samples were collected from cells at 24 postinfection and analyzed for N protein expression with human actin expression as the loading control (
Figure 4, lanes 5–8). Compared to the cells with the empty vector without any ribozyme sequence, the N protein levels were similar in the cells with control ribozymes M1-N-C and M1-IE but reduced by about 85% in the cells with functional ribozyme M1-N-F (
Figure 4, lanes 1–4). Thus, M1-N-F appeared to decrease the expression of N protein, which is encoded by the HCoV-229E RNA targeted by the ribozyme, and the decrease of N protein levels was due to the catalytic activity of M1-N-F, since no reduction was found in the cells with control inactive ribozyme M1-N-C.
Suppression of HCoV-229E growth and replication by M1-N-F. The reduction of viral genome RNA levels and the expression of N protein, a protein essential for viral replication, is expected to decrease the production and growth of HCoV-229E [
1,
27]. To confirm if this is the case, we performed a time-course study to monitor virus production and replication in the constructed cell lines for an infection period of 7 days. Virus samples were prepared from the cells infected with HCoV-229E at different time points postinfection and the levels of viruses in these samples were quantified using a plaque infection assay. At each of the time points, virus titers in the cells with functional ribozyme M1-N-F were consistently lower than those in the cells with the empty vector or with control ribozymes M1-N-C or M1-IE. For example, compared to the cells with the empty vector without any ribozyme sequence, virus titers were similar in the cells with ribozymes M1-N-C and M1-IE but reduced by 750-fold in the cells with ribozyme M1-N-F at 6 days postinfection (
Figure 5). Thus, virus replication and production appeared to be blocked by functional ribozyme M1-N-F.
Specific antiviral effects on HCoV-229E but not human cytomegalovirus by RNase P ribozyme M1-N-F. To examine if the antiviral effects of M1-N-F are specific for HCoV-229E but not for other viruses, the constructed cell lines were infected with the human cytomegalovirus (HCMV) and we examined the effects of the ribozymes on HCMV gene expression and growth. By amplifying the IE1 mRNA sequence, our qRT-PCR assays did not uncover any difference in HCMV IE1 mRNA expression levels among the cells without ribozymes or with M1-N-F or M1-N-C (
Figure 6A). Our plaque assays of the virus samples collected at different time points postinfection did not identify any difference in virus titers from the cells without ribozymes or with M1-N-F or M1-N-C (
Figure 6B). However, an 85% reduction in IE1 mRNA expression and a decrease of 200-fold in the virus titers were found in the cells expressing M1-IE at 2 and 7 days postinfection (
Figure 6), respectively, consistent with previous observations that this ribozyme effectively inhibited IE1 and IE2 mRNA expression and blocked HCMV replication and growth in cultured cells [
28]. These results indicated that the targeting effects of M1-N-F, such as decreasing the expression of its target and reducing virus replication and production, are specific for HCoV-229E but not for HCMV.
4. Discussion
Nucleic acid-based gene targeting approaches, such as ribozymes, siRNAs/shRNAs, antisense molecules, and CRISPR/guide RNAs, pose great potential to be used in clinical applications [
13,
14,
15,
38]. M1GS RNA, a ribozyme engineered from RNase P catalytic RNA, has been shown to efficiently cleave numerous mRNA targets and effectively inhibit gene expression and the infection of several human viruses [
17]. However, it has not been reported if M1GS RNAs can be used for the inhibition of a human coronavirus infection. In the current study, we provide the first direct evidence that the M1GS ribozyme slices the genomic RNA sequence of HCoV-229E in vitro and decreases the gene expression and production of HCoV-229E in cultured cells.
We developed a functional ribozyme, M1-N-F, from M1 RNA by adding a guide sequence that targeted the HCoV-229E N mRNA sequence overlapping with the viral RNA genome. In human foreskin fibroblasts, M1-N-F expression decreased viral N expression and genomic RNA level by at least 85% and virus growth by at least 750-fold (
Figure 3,
Figure 4 and
Figure 5). On the contrary, no decrease in viral N protein expression, viral RNA levels, and virus growth was found in the cells expressing no ribozymes or the control ribozyme M1-N-C (
Figure 3,
Figure 4 and
Figure 5). M1-N-C, derived from an inactive M1 RNA mutant, had an identical guide sequence and comparable binding affinity (measured as K
d) to the n-39 target as M1-N-F but was not catalytically active due to its mutations at the M1 RNA active domain (
Figure 1 and
Table 1). Thus, the antiviral effects associated with M1-N-F expression appeared to be due to the M1GS RNA-mediated catalytic cleavage of the target viral RNA but not the antisense effect of the guide sequence or the other nonspecific effects associated with ribozyme expression.
The potential off-target effect of RNase P ribozyme targeting represents an important issue to be addressed in developing this technology for clinical applications in vivo. The salient features of our results suggest that the effects mediated by M1-N-F ribozyme targeting are specific for the intended HCoV-229E target and infection. Firstly, M1-N-F expression displayed no substantial cytotoxic effects because of the comparable viability and growth observed among the cells with no ribozymes or with the constructed ribozymes including M1-N-F. Moreover, ribozyme expressions did not alter the expression of human H1 RNA and actin mRNA and proteins, suggesting no impact of M1-N-F expression on the gene expression of host cells.
Secondly, the antiviral effects associated with M1-N-F expression appeared to be due to the ability of this ribozyme to cut the N mRNA region of HCoV-229E. The N mRNA overlaps and is completely within the viral RNA genome with the same 3′ terminal sequence [
1,
27]. Thus, M1-N-F targeting would cut the N mRNA as well as all other sgRNAs and the full length genomic RNA, leading to the inhibition of the expression of N and other viral genes and overall viral RNA genome synthesis, as evidenced by a reduction of 85% in N protein expression and viral RNA levels and a decrease of about 750-fold in viral titers at 6 days postinfection in the M1-N-F-expressing cells (
Figure 3,
Figure 4 and
Figure 5). In contrast, no changes in HCoV-229E gene expression and growth were found in the cells with no ribozymes and with control ribozyme M-N-C.
Thirdly, the effects of M1-N-F-mediated targeting appeared to be specifically against its intended HCoV-229E target. This is because M1-N-F expression had no effect on HCMV gene expression and growth (
Figure 6). On the contrary, we observed a substantial inhibition of HCMV IE1 expression and growth in the cells expressing another control ribozyme, M1-IE, which targeted HCMV IE1 mRNA. M1-IE has been shown to cut efficiently the target IE1 mRNA sequence in vitro and inhibit viral IE1 expression and growth in cultured cells [
28]. Thus, M1GS ribozymes appear to operate selectively on their RNA target.
Additional experiments should be carried out to address the potential off-target effect of RNase P ribozyme targeting. In our current study, we examine the anti-HCoV-229E effects of ribozyme M1-IE, which targets HCMV mRNAs but not any HCoV-229E RNA sequences. We included M1-IE to determine the targeting effect of the M1GS ribozyme on the N mRNA region and HCoV-229E infection when the ribozyme had an incorrect targeting sequence that targeted another target RNA but not the N mRNA region of HCoV-229E. Additionally, RNase P ribozymes with scramble guide sequences can be constructed and assayed for their anti-HCoV-229E activities and compared with M1-N-F. These results will provide insight into the potential off-target effect of the RNase P ribozyme-based technology.
HCoV-229E is a member of the human coronavirus family that also includes highly pathogenic viruses such as SARS-CoV, MERS-CoV, and SARS-CoV-2, which is the causative agent of COVID-19 [
1,
2]. HCoV-229E belongs to a group of four “low pathogenic” human coronaviruses that usually cause mild complications such as the common cold but may lead to severe upper and lower respiratory infections in infants, elderly individuals, and immunocompromised patients [
1,
8,
9]. Thus, efforts in developing novel strategies against HCoV-229E should not only help with the treatment of HCoV-229E-associated diseases but also human infections and the diseases associated with other coronaviruses such as SARS-CoV-2.
The N protein is indispensable for the infection and replication of all human coronaviruses [
10]. Moreover, the N protein gene is one of the most conservative functional proteins in coronavirus evolution and is more conserved and stable than the S protein [
11,
12]. Our study provides the first direct evidence that RNase P ribozymes are effective in knocking down HCoV-229E gene expression and replication and further demonstrates the therapeutic potential of RNase P ribozymes against all human coronaviruses by targeting the N gene.
More studies are needed before employing RNase P ribozymes in practical clinical applications against HCoV-229E. One of the key issues is the delivery of the therapeutic agents to the tissues and cells that HCoV-229E infects in vivo. Like all coronaviruses, HCoV-229E replicates within cytoplasm and specific expression cassettes, such as the 7SL RNA promoter used in our study [
33], may need to be used for the cytoplasmic expression of the M1GS ribozyme for anti-coronavirus applications. Different methods, such as viral vector-based agents or liposome-based nanoparticles, are currently being explored for the efficient and specific delivery of different gene-targeting molecules, including ribozymes and siRNAs, to numerous tissues and cells for therapeutic applications [
15,
39]. These methods can potentially be developed to deliver RNase P ribozyme expression cassettes or chemically synthesized M1GS ribozyme molecules to the respiratory tract and lung tissues, where the infection of HCoV-229E occurs in vivo.
Two other important issues include the use of an appropriate expression system to allow the highest levels of ribozyme expression in vivo and the development of RNase P ribozyme variants with increased catalytic activities. An examination of the expression of a known well-expressed ribozyme driven by the 7SL RNA promoter in the pUC7SL expression system would confirm whether the RNase P ribozymes have been optimally and efficiently expressed in our study. Additional expression systems, other than the system with the 7SL RNA promoter used in our study, should be explored and used for ribozyme expression. Moreover, further engineering of RNase P ribozymes to increase their catalytic activity should be conducted using different methods such as in vitro evolution and selection procedures [
40,
41,
42]. Novel RNase P ribozyme variants with increased catalytic activities have been generated by in vitro evolution and selection procedures and exhibited better efficacy in shutting down the HSV-1 gene expression and HCMV replication in cultured cells [
14,
31]. We hope that the RNase P ribozyme variants generated with these procedures will lead to better efficacy in reducing HCoV-229E gene expression and replication. These studies will further facilitate the development of RNase P ribozymes for therapeutic applications against the infections of human coronaviruses including HCoV-229E.