CRISPR Interference-Mediated Silencing of the mmpL3 Gene in Mycobacterium smegmatis and Its Impact on Antimicrobial Susceptibility

Background: The discovery of novel therapeutic agents, especially those targeting mycobacterial membrane protein large 3 (mmpL3), has shown promise. In this study, the CRISPR interference-Streptococcus thermophilus nuclease-deactivated Cas9 (CRISPRi-dCas9Sth1) system was utilized to suppress mmpL3 expression in Mycobacterium smegmatis, and its impacts on susceptibility to antimicrobial agents were evaluated. Methods: The repression of the mmpL3 gene was confirmed by RT-qPCR. The essentiality, growth curve, viability, and antimicrobial susceptibility of the mmpL3 knockdown strain were investigated. Results: mmpL3 silencing was achieved by utilizing 0.5 and 1 ng/mL anhydrotetracycline (ATc), resulting in reductions in the expression of 60.4% and 74.4%, respectively. mmpL3 silencing led to a significant decrease in bacterial viability when combined with one-half of the minimal inhibitory concentrations (MICs) of rifampicin, rifabutin, ceftriaxone, or isoniazid, along with 0.1 or 0.5 ng/mL ATc (p < 0.05). However, no significant difference was observed for clarithromycin or amikacin. Conclusions: The downregulation of the mmpL3 gene in mycobacteria was achieved through the use of CRISPRi-dCas9Sth1, resulting in growth deficiencies and resensitization to certain antimicrobial agents. The impact was dependent upon the level of gene expression.


Introduction
Mycobacterial infections have emerged as a significant global health problem, primarily due to the increase in the prevalence of drug-resistant Mycobacterium tuberculosis and drug-resistant nontuberculous mycobacteria (NTM) in various regions.These drugresistant strains present a substantial challenge, as they diminish the efficacy of commonly used antimicrobial agents in treatment regimens.To address this challenge, the discovery of new therapeutic agents and targets for effective pathogen elimination is important.An example of an essential gene of interest is mycobacterial membrane protein large 3 (mmpL3), responsible for transporting trehalose monomycolate, which serves as a precursor for producing trehalose dimycolate and mycolate-bound arabinogalactan within the cell [1,2].The efficacy of mmpL3 inhibitors against M. tuberculosis, such as SQ109 [3,4], NITD-304, NITD-349, AU1235, and AU36 [5], and NTM, such as PIPD1 [6], has been established, demonstrating their effectiveness as antimycobacterial agents.
Combination drug regimens are recommended for mycobacterial infections, including the use of rifampicin, isoniazid, ethambutol and pyrazinamide for tuberculosis [7], as well as macrolides, aminoglycosides, ethambutol, and rifampicin for NTM infections [8].Nevertheless, mycobacteria exhibit intrinsic resistance to diverse classes of antimicrobial drugs, such as rifampicin in M. smegmatis and β-lactams in both M. smegmatis and M. tuberculosis [9][10][11][12].Furthermore, acquired resistance to these antimicrobial drugs further limits treatment options.Therefore, enhancing the efficacy of existing antimicrobial drugs that are currently ineffective is imperative for managing and eradicating mycobacterial infections.Previously, the synergistic effects of MmpL3 inhibitors [4,5,13,14] or mmpL3 knockdown [15][16][17] and certain antimicrobial agents have been investigated.However, these studies demonstrated different susceptibility results depending on the experimental conditions.
One promising tool for studying mycobacterial genes is the clustered regularly interspaced short palindromic repeats interference (CRISPRi) system.This system has been developed to facilitate the manipulation and regulation of gene silencing in Mycobacterium.Specifically, the CRISPRi system utilizes Streptococcus thermophilus nuclease-deactivated Cas9 (dCas9 Sth1 ) to effectively suppress the expression of endogenous mycobacterial genes [18].The CRISPRi-dCas9 Sth1 system demonstrates both specificity and efficiency in the gene knockdown of essential and nonessential mycobacterial genes, including those associated with drug resistance [19,20].In this study, we investigated the impact of silencing the mycobacterial mmpL3 gene via the CRISPRi-dCas9 Sth1 system on susceptibility to different classes of non-MmpL3-dependent antimicrobial agents, including rifampicin, rifabutin, isoniazid, amikacin, clarithromycin, and β-lactam drug.A comprehensive understanding of how a reduction in mmpL3 expression influences antimycobacterial susceptibility is crucial for the advancement of therapeutic strategies.

CRISPRi Targeting of the mmpL3 Gene Impacts M. smegmatis Growth
The mmpL3 knockdown M. smegmatis (mmpL3_KD) strain was constructed to verify the essentiality of the mmpL3 gene using the CRISPRi-dCas9 Sth1 system which interferes the transcription elongation from binding of dCas9-single guide RNA (sgRNA) complex to mmpL3 gene [18].The specificity of sgRNA targeting mmpL3 gene [21] was verified using BLAST tool (NCBI) against the genome of M. smegmatis MC 2 155.No complementarity of the mmpL3 sgRNA sequence was identified in the other sites of the genome of M. smegmatis.The sgRNAs targeting the mmpL3 gene effectively suppressed the growth of the mmpL3_KD strain in a dose-dependent manner at anhydrotetracycline (ATc) concentrations ranging from 0.5 to 50 ng/mL, while the pLJR962 control strain exhibited normal growth (Figure 1 and Supplementary Figure S1).Mycobacterial growth of mmpL3_KD was completely inhibited at ATc concentrations of 5, 10, and 50 ng/mL.There was no apparent impairment of mycobacterial growth in the presence of 0.1 ng/mL ATc.Furthermore, a few ATc-resistant strains were observed on 7H10 agar plates containing high concentrations of ATc (5, 10, or 50 ng/mL).These ATc-resistant colonies were capable of growing upon subculture on agar containing ATc.Moreover, induction with ATc did not result in toxicity, as evidenced by the normal growth of the pLJR962 control strain (Supplementary Figure S1).

mmpL3 Gene Expression in the Presence and Absence of ATc
The impact of inducing the transcription response on mmpL3_KD expression in the M. smegmatis strain was assessed using RT-qPCR.mmpL3 expression in mmpL3_KD decreased upon induction with 0.1, 0.5, and 1 ng/mL ATc, resulting in reductions in the expression of 4.4% ± 7.1% (mean ± standard deviation (SD)), 60.4% ± 25.5%, and 74.4% ± 35.7%, respectively.Compared to that in the absence of ATc, mmpL3 expression was significantly lower in the presence of 0.5 and 1 ng/mL ATc, with p < 0.05 and p < 0.01, respectively (Figure 2).

Growth Curve and Viability of mmpL3_KD M. smegmatis at Different ATc Concentrations
The impact of mmpL3 transcriptional repression on the growth curve and viability (colony forming unit (CFU)/mL) of the mmpL3_KD strain was examined in liquid media supplemented with ATc at concentrations ranging from 0 to 50 ng/mL.A defect in the growth curve and viability of the mmpL3_KD strain was evident at ATc concentrations ≥0.5 ng/mL.Moreover, a significant reduction in the optical density of 600 nm (OD 600 ) of the culture was observed at 24, 36, and 48 h of induction with 1, 5, 10, and 50 ng/mL ATc (p < 0.05) compared to conditions without ATc (Figure 3A).In addition, the CFU/mL of mmpL3_KD rapidly decreased upon exposure to 1, 5, 10, and 50 ng/mL ATc (p < 0.05) within 4 h (Figure 3B).These significant reductions in CFU/mL corresponded to 3.65 ± 0.78-log10 CFU/mL.Furthermore, the OD 600 of the mmpL3_KD culture gradually increased after 24 h of ATc induction and was not significantly different from that in the absence of ATc at 72 and 96 h (Figure 3A).Similarly, the viability (CFU/mL) slowly increased and showed nonsignificant differences at 48, 72, and 96 h of ATc induction (Figure 3B).Additionally, the growth curve and viability of the pLJR962 control strain were comparable to those of the mmpL3_KD strain under conditions without ATc induction or with 0.1 ng/mL ATc (Supplementary Figure S2A,B).Moreover, the biofilm formation of mmpL3_KD was comparable to the pLJR962 control (p > 0.05).For pellicle formation, no pellicle was observed at high ATc concentrations (≥1 ng/mL), which could be due to defects in mycobacterial growth (Supplementary Figures S3 and S4).

Resensitization of the mmpL3_KD M. smegmatis Strain to Antimicrobial Agents
To evaluate whether mmpL3 repression affects the susceptibility of the mmpL3_KD strain to non-MmpL3-dependent antimicrobial agents, the minimal inhibitory concentrations (MICs) of various antimicrobial agents against the mmpL3_KD strain were initially assessed.The MICs were determined as follows: 0.25 µg/mL for amikacin, 0.5 µg/mL for clarithromycin, 4 µg/mL for isoniazid, >512 µg/mL for ceftriaxone, >8 µg/mL for rifampicin, and 2 µg/mL for rifabutin.In the presence of 0.1 ng/mL ATc, the MICs of mmpL3_KD were not affected.However, the MICs decreased by one-to two-fold dilutions in the presence of 0.5 ng/mL ATc, which could be due to a growth defect resulting from mmpL3 repression.
Therefore, the impact on the viability (CFU/mL) of the mmpL3_KD strain was assessed in the presence of 0.1 ng/mL ATc, which suppressed mmpL3 expression but did not affect the growth of mycobacterial cells, and one-half of the MICs of the antimicrobial agents, which were 0.125 µg/mL for amikacin, 0.25 µg/mL for clarithromycin, 2 µg/mL for isoniazid, 256 µg/mL and 512 µg/mL for ceftriaxone, 4 µg/mL and 8 µg/mL for rifampicin, and 1 µg/mL for rifabutin.For rifampicin and ceftriaxone, two concentrations were tested because the MIC values exceeded the highest concentration tested in the experiment.One-half of the MIC of each antimicrobial agent reduced the CFU/mL of mmpL3_KD by approximately 0.17-log10 for isoniazid, 0.36-log10 for 256 µg/mL of ceftriaxone, 1.04-log10 for 512 µg/mL of ceftriaxone, 0.52-log10 for 4 µg/mL of rifampicin, 0.57-log10 for 8 µg/mL of rifampicin, and 1.20-log10 for rifabutin, compared to the no-drug control in the absence of ATc.No significant reduction in CFU/mL was observed with amikacin or clarithromycin.
Although the suppression of mmpL3 with 0.1 ng/mL ATc did not alter the MIC, the combination of this ATc level and one-half of the MIC of certain antimicrobial agents resulted in a significant decrease in mmpL3_KD viability.Specifically, the viability of mmpL3_KD with 0.1 ng/mL ATc with isoniazid, ceftriaxone, rifampicin (8 µg/mL), and rifabutin was significantly reduced for 0.77 ± 0.35-log10 CFU/mL) compared with antimicrobial agents alone (p < 0.05) (Figure 4).However, clarithromycin and amikacin exhibited reductions of only 0.07 and 0.08-log10 CFU/mL, respectively (Figure 4).To assess whether greater mmpL3 suppression can more effectively reduce mycobacterial viability, 0.5 ng/mL ATc was added to one-half of the MIC plates.As expected, the viability of the mmpL3_KD strain significantly decreased in the presence of 0.5 ng/mL ATc (p < 0.0001), showing a decrease of 1.47 ± 0.23-log10 CFU/mL compared to that in the absence of ATc, regardless of the antimicrobial agent concentration (0 µg/mL, one-half MIC) (Figure 5).Conversely, the addition of one-half MIC of isoniazid, rifabutin, ceftriaxone (512 µg/mL), or rifampicin (8 µg/mL) to mmpL3_KD with 0.1 or 0.5 ng/mL ATc significantly enhanced the decrease in bacterial viability (p < 0.05) compared to the condition with ATc alone (Figures 4 and 5).Additionally, induction of dCas9 Sth1 without a sgRNA in the pLJR962 control did not increase the defects of growth or viability of M. smegmatis compared to treatment with antimicrobial agents alone (p > 0.05) (Figures 4 and 5).

Discussion
MmpL3 functions as a transporter of trehalose monomycolate, a precursor of trehalose dimycolate [1,2].mmpL3 in M. smegmatis consists of 3042 base pairs encoding 1013 amino acids, serving as an ortholog of rv0206c in M. tuberculosis.Extensive studies have underscored the importance of mmpL3 as an essential gene in M. tuberculosis [15,22], M. smegmatis [23], and M. abscessus [24].Therefore, mmpL3 has been identified as a promising target for developing new therapeutic agents against M. tuberculosis and NTM [3][4][5][6]25,26].Previously, the synergistic effects of MmpL3 inhibitors [4,5,13,14] or mmpL3 knockdown using various genetic tools [15][16][17] and certain antimicrobial agents have been investigated.However, these studies have demonstrated varying susceptibilities to non-MmpL3-dependent antimicrobial agents, which may be explained by differences in the level of mmpL3 inhibition or repression achieved under different experimental conditions and in different organisms which can differ substantially in many aspects, limiting the direct application of findings from one to another.
The CRISPRi system is easily constructed and requires only a single transformation, enabling the rapid generation of transcriptional knockdown strains of mycobacteria.In mycobacteria, CRISPRi has demonstrated effectiveness in suppressing drug resistance genes, particularly those associated with β-lactams, by targeting essential peptidoglycan synthesis genes such as pbpB and cwlM [20].Furthermore, it has been utilized to target the rifampicin resistance gene ADP-ribosyltransferase (arr) [19].This versatility makes CRISPRi a valuable tool for investigating resistance mechanisms and devising strategies to combat drug-resistant mycobacteria.In our study, the CRISPR-dCas9 Sth1 system was used to repress mmpL3 expression, and its impacts on M. smegmatis viability and susceptibility to non-MmpL3-dependent antimicrobial agents were evaluated.The induction of dCas9 Sth1 with a high concentration of ATc did not adversely affect the growth or viability of M. smegmatis wild-type or vector control strains, as similarly demonstrated in a previous study [18].Similar to a previous study [17], the depletion of MmpL3, which is below the levels required for the in vitro growth of mycobacteria, was rapidly achieved with low concentrations of ATc (0.5 ng/mL) in an ATc dose-dependent manner.In our study, this is attributed to the permissive PAM sequences utilized (NNAGAAA), which achieved a gene repression of up to 158.1-fold [18], and the sgRNA targeting the coding region of the mmpL3 gene.Additionally, the gradual increase in growth was observed after 24 h of ATc induction.This phenomenon could be attributed to the degradation of ATc or the growth of the mmpL3_KD M. smegmatis escape mutants.Consequently, their potential influence on the results and subsequent statistical analysis cannot be disregarded.The emergence of strains unresponsive to the ATc inducer is consistent with findings from a previous study [17] and with other tetracycline-regulated promoters utilized in mycobacteria [27].
Importantly, this study investigated the effects of mmpL3 repression on susceptibility to antimicrobial agents.Although a reduction in MIC and CFU/mL was not observed with an approximately 4% to 10% decrease in mmpL3 expression in response to 0.1 ng/mL ATc, a significant reduction (p < 0.05) in CFU/mL was noted when this decreased expression was combined with half of the MICs of several classes of antibiotics, including rifamycin, βlactam, and isoniazid, demonstrating a similar pattern.However, this reduction in viability did not meet the criteria for synergy, defined as a ≥2 log10 reduction in CFU compared to the antimicrobial agent alone [20].Several hypotheses could explain these findings.For instance, the ATc concentration might be too low, resulting in insufficient mmpL3 repression.Additionally, the emergence of nonresponsive escape mutants resistant to ATc could have contributed to these results.The significant decrease in CFU/mL observed in the rifamycin group, regardless of rifampicin or rifabutin, can be attributed to disruption of the mycolic acid in the cell wall, resulting in alterations in hydrophobicity [16].This explanation could also apply to isoniazid, where mmpL3 knockdown may enhance the inhibition of mycolic acid synthesis.In contrast, repression of mmpL3 in M. tuberculosis using the CRISPRi system did not change the MICs of rifampin, isoniazid, or linezolid [17].Additionally, the synergistic effect of SQ109, an MmpL3 inhibitor, against M. tuberculosis has been demonstrated with isoniazid and rifampicin but not with ethambutol and pyrazinamide in vitro [13].With respect to ceftriaxone, mmpL3_KD was observed to result in improved susceptibility due to enhanced penetration of β-lactams facilitated by alterations in outer membrane assembly.However, no study has explored the effect of clarithromycin on mmpL3 knockdown.One study demonstrated that specific single nucleotide polymorphisms (SNPs) within the mmpL3 gene of M. smegmatis resulted in growth defects and susceptibility to ampicillin, rifampicin, and erythromycin, but not to chloramphenicol or kanamycin.This susceptibility arose because the depletion of mmpL3 disrupts cell wall formation in mycobacteria [16].These findings suggest that the synergistic mechanism of mmpL3 repression or inhibition with antimycobacterial agents relies on optimal conditions, such as the level of repression and classes or concentration of antimicrobial agents used.Therefore, further studies are needed to elucidate the true mechanism involved.
In conclusion, these findings emphasize the utility of the CRISPRi-dCas9 Sth1 system for selectively targeting and suppressing genes associated with different classes of antimicrobial resistance.Moreover, this approach has the potential to enhance studies focused on understanding of gene functions and resistance to antimicrobial agents.Additionally, the repression of the mmpL3 gene in mycobacteria represents a promising strategy for treating mycobacterial infections, particularly those caused by drug-resistant strains.
These sgRNA oligos were annealed, ligated into a BsmBI-digested CRISPRi backbone (pLJR962), and transformed into E. coli DH5α, which was selected on LB agar supplemented with kanamycin.Purified sgRNA-pLJR962 plasmids from E. coli were verified using Sanger sequencing with the pLJR962-965-SS primer: 5 ′ TTCCTGTGAAGAGCCATTGATAATG 3 ′ .Next, sequence-verified plasmids (100 ng) and pLJR962 plasmids without sgRNA (empty vector control) were electroporated at 25 kV and 25 µF with 1000 W resistance into electrocompetent M. smegmatis MC 2 155 strains, which were prepared as previously described [21].The transformants were selected on 7H10 agar supplemented with 25 µg/mL kanamycin in the presence or absence of 50 ng/mL ATc to determine the bacterial viability of the mmpL3_KD strains.Subsequently, the presence of the CRISPRi-dCas9 Sth1 construct in M. smegmatis was confirmed by PCR.

Bacterial Growth Curve and Viability Count
The mid-log phase cultures of the mmpL3_KD and pLJR962 control M. smegmatis strains were prepared at an OD 600 of 0.1 and induced with various concentrations of ATc: 0, 0.1, 0.5, 1, 5, 10, and 50 ng/mL.The cultures were then incubated, and the OD 600 was measured at different time points from 0 to 96 h.For viability assessment, the CFU/mL of cultures were determined by collecting cultures both with and without ATc at various time points ranging from 0 to 96 h.Each culture condition was serially diluted (10 −2 -10 −8 ), and 20 µL of the diluted samples was plated and incubated at 37 • C for 3 days or until visible colonies were observed to determine the CFU/mL.All experiments were conducted in biological and technical triplicates.
The impact of the ATc inducer and antimicrobial agents on the viability of the mmpL3_KD and pLJR962 control M. smegmatis strains was evaluated.Five microliters of 10-fold serial dilutions (10 −1 -10 −6 ) of mid-log phase cultures were plated on 7H10 agar containing one-half of the MIC of the respective antimicrobial agents in the presence or absence of 0.1 and 0.5 ng/mL ATc.The plates were then incubated at 37 • C until visible colonies were observed to determine the CFU/mL.All experiments were conducted with biological and technical replicates.

RNA Extraction, cDNA Synthesis, and RT-qPCR
To investigate the strength of the transcriptional repression, mid-log phase cultures of the mmpL3_KD and pLJR962 control M. smegmatis strains were initiated at an OD 600 of 0.1 in 7H9 broth supplemented with ADC, 0.05% Tween 80, and 25 µg/mL kanamycin.These

Figure 1 .
Figure 1.Essentiality of the mmpL3 gene in the mmpL3_KD M. smegmatis strain.Serial dilutions (10 −1 to 10 −5 ) of log-phase cultures of the mmpL3_KD M. smegmatis strain were plated on 7H10 agar in both the absence and presence of varying concentrations of ATc (0.1, 0.5, 1, or 5 ng/mL).

Figure 2 .
Figure 2. mmpL3 gene expression in the mmpL3_KD M. smegmatis strain in the presence and absence of ATc (ng/mL).The experiments were conducted in biological and technical triplicates.Statistical significance is indicated as * p < 0.05 and ** p < 0.01, and "ns" indicates not statistically significant.The error bars represent the SDs of the means.

Figure 3 .
Figure 3. Growth curve and viability of the mmpL3_KD M. smegmatis strain at different time points.(A) OD 600 of the mmpL3_KD M. smegmatis strain at different ATc concentrations (ng/mL).(B) The viability (log10 CFU/mL) of the mmpL3_KD M. smegmatis strain at different ATc concentrations (ng/mL).The experiments were conducted in biological and technical triplicates.The error bars represent the SDs of the means.

Figure 4 .
Figure 4. Repression of mmpL3 with 0.1 ng/mL ATc resulted in resensitization to isoniazid, ceftriaxone, rifampicin, and rifabutin.The viability (log10 CFU/mL) of the mmpL3_KD and pLJR962 control M. smegmatis strains treated with antimicrobial agents in the presence or absence of 0.1 ng/mL ATc was determined in biological and technical triplicates.Statistical significance is indicated as * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001, and "ns" indicates not statistically significant.The error bars represent the SDs of the means.

Figure 5 .
Figure 5. Repression of mmpL3 with 0.5 ng/mL ATc significantly impacted M. smegmatis viability and antimicrobial susceptibility.The viability (log10 CFU/mL) of the mmpL3_KD and pLJR962 control M. smegmatis strains treated with antimicrobial agents in the presence or absence of 0.5 ng/mL ATc was determined in biological and technical triplicates.Statistical significance is indicated as * p < 0.05, **** p < 0.0001, and "ns" indicates not statistically significant.The error bars represent the SDs of the means.