CRISPR‐based genome editing reveals the roles of efflux pumps in Mycobacterium abscessus
Sishang Li, Aofei Duan, Lanyue Zhang, Chunliang Wang, Meiyi Yan, Gai‐Xian Ren, Li‐Ping Pan, Yi‐Cheng Sun
TL;DR
This study uses CRISPR to investigate efflux pumps in Mycobacterium abscessus, revealing their roles in drug resistance and virulence.
Contribution
Developed optimized CRISPR-Cas9 tools for M. abscessus and identified key efflux pumps critical for drug resistance and virulence.
Findings
Different efflux pumps have distinct roles in drug resistance and survival in Galleria mellonella larvae.
The MAB_2806-2807 operon is vital for drug resistance and virulence in M. abscessus.
CRISPR-based tools enable efficient genetic manipulation of M. abscessus for molecular research.
Abstract
Mycobacterium abscessus, one of the most antimicrobial‐resistant bacteria, is increasingly recognized as the cause of infections that are difficult to treat. Novel genetic manipulation tools are required to elucidate the biology, pathogenesis, and antibiotic resistance mechanisms of M. abscessus. In this study, we modified the method used to prepare M. abscessus electrocompetent cells to achieve efficient transformation, and then optimized the CRISPR‐Cas9‐assisted genome‐editing tools to allow efficient genetic manipulation. Using these tools, we constructed 66 efflux pump mutants of M. abscessus and investigated their roles in drug resistance and virulence. We found that different efflux pumps play distinct roles in drug resistance and survival in Galleria mellonella larvae. Finally, we confirmed that MAB_2806‐2807, involved in transportation of triacylglycerides, is vital for the drug…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7| Mutation | Zone of inhibition for mutants compared with the wild type | |||||||
|---|---|---|---|---|---|---|---|---|
| Target gene | length (bp) | Description | LZD | CLR | AMK | BDQ | CFZ | FOX |
|
| −182 | MFS transporter | − | − | − | − | − | − |
|
| −442 | Cation diffusion facilitator family transporter | ++ | ++ | + | − | − | − |
|
| −102 | Probable amino acid ABC transporter, ATP binding | − | − | − | − | − | − |
|
| −77 | Putative oligopeptide ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −452 | Putative iron compound ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −4 | Putative ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| +1 | Amino acid metabolite efflux pump | − | − | − | − | − | − |
|
| −627 | Putative oligopeptide ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −370 | Probable phosphate import ABC transporter, ATP‐binding protein PstB | − | − | − | − | − | − |
|
| −12 | MFS transporter | − | − | − | − | − | − |
|
| +99 | MFS transporter | − | − | − | − | − | − |
|
| −70 | Probable drug resistance transporter | − | − | − | − | − | − |
|
| −1 | Putative ABC transporter, ATP‐binding protein | ++ | ++ | − | − | − | − |
|
| −1 | Putative ABC transporter, ATP‐binding protein | − | ++++ | − | − | − | − |
|
| −15 | Putative ABC transporter, ATP‐binding protein | +++ | ++ | − | − | − | − |
|
| −1 | Putative multi‐drug efflux transporter | − | − | − | − | − | − |
|
| −346 | Putative ABC transporter, ATP‐binding protein | +++ | ++ | − | − | − | − |
|
| −11 | DHA2 family efflux MFS transporter permease subunit | ++ | ++ | − | − | − | − |
|
| −4 | Probable ABC transporter, ATP‐binding protein | ++ | − | + | − | − | − |
|
| −1126 | Probable ABC transporter, ATP‐binding protein | − | + | − | − | +++ | − |
|
| −18 | Probable ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −320 | Probable sulfate ABC transporter, ATP‐binding protein SysA | ++ | ++ | − | − | − | − |
|
| +1 | Probable daunorubicin resistance ABC transporter, ATP‐binding subunit | − | − | − | − | − | − |
|
| −453 | Hypothetical ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −1 | Putative ABC transporter, ATP‐binding protein | − | + | ++ | − | − | + |
|
| −12 | Probable ABC transporter antibiotic‐transport, ATP‐binding protein | − | − | − | − | − | − |
|
| −2 | MFS transporter | − | − | − | +++ | − | − |
|
| −8 | Sulfonate ABC transporter, ATP‐binding subunit SsuB | ++ | ++ | − | − | − | − |
|
| −184 | Hypothetical ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −2 | Hypothetical ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| +1 | Drug efflux membrane protein | − | − | − | − | − | − |
|
| +1 | Putative ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −557 | Molybdenum ABC transporter ModC, ATP‐binding protein | ++ | ++ | − | − | − | − |
|
| +1 | Fluoride efflux transporter CrcB | − | − | − | − | − | − |
|
| −448 | Putative sulfate ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −1 | ACR3 family arsenite efflux transporter | − | − | − | − | − | − |
|
| −45 | ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −2 | ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −716 | Probable ATP‐binding protein, ABC transporter CydD | − | − | − | − | − | − |
|
| −279 | Probable ATP‐binding protein, ABC transporter CydC | +++ | ++ | − | − | ++++ | − |
|
| −1 | Quaternary ammonium compound efflux SMR transporter | − | − | − | − | − | − |
|
| −8 | MFS transporter | − | − | − | − | − | − |
|
| −13 | Probable macrolide ABC transporter, ATP‐binding protein | ++++ | ++ | − | + | − | +++ |
|
| −431 | Methionine import ATP‐binding protein metN | − | − | − | − | − | − |
|
| −14 | Major facilitator superfamily MFS_1 | +++++ | +++++ | +++++ | +++++ | +++++ | +++++ |
|
| −488 | Putative transmembrane‐transport protein | − | − | − | − | − | − |
|
| −4 | Putative glutamate ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −1 | MATE family efflux transporter | − | − | − | − | − | − |
|
| −17 | Putative transmembrane efflux protein | +++ | ++ | ++ | + | − | +++++ |
|
| −20 | Putative ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −22 | Putative ABC transporter, ATP‐binding protein | +++ | ++ | − | − | − | ++ |
|
| −629 | EamA family transporter | − | − | − | − | − | − |
|
| +1 | Probable ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −376 | Putative ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −2 | Possible ribonucleotide ABC transporter, ATP binding protein | − | − | − | − | − | − |
|
| −1 | LysE family translocator | − | − | − | − | − | − |
|
| −54 | MFS transporter | +++ | − | ++ | − | − | − |
|
| −357 | Putative ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −14 | Putative amino acid ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −164 | MFS transporter | − | − | − | − | − | − |
|
| −1 | HAMP domain‐containing protein | − | − | − | − | − | − |
|
| +1 | Probable ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −6 | Putative ABC transporter, ATP‐binding protein | − | − | − | − | − | − |
|
| −11 | Phosphate ABC transporter, ATP‐binding protein | ++ | − | + | − | − | + |
|
| −38 | ACR3 family arsenite efflux transporter | − | − | − | − | − | − |
|
| −534 | MFS transporter | − | − | − | − | − | − |
- —the National Key R&D Program of China
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsCRISPR and Genetic Engineering · Mycobacterium research and diagnosis · Tuberculosis Research and Epidemiology
INTRODUCTION
Mycobacterium abscessus is an emerging opportunistic pathogen that can cause skin infections and lung diseases in patients with cystic fibrosis, resulting in either acute or chronic disease and a decline in pulmonary function1, 2, 3. Recent epidemiological studies have documented the spread of M. abscessus clones between hospitals worldwide, posing a major threat to global health care4, 5. M. abscessus is widely recognized as one of the most drug‐resistant organisms due to its intrinsic resistance, making infections extremely difficult to treat6. Current treatment regimens for M. abscessus infections often involve prolonged intravenous therapy with multiple drugs, and yet, the cure rate remains low.7 Consequently, treatment options for M. abscessus infections are limited, and despite extended treatment courses, the therapeutic outcomes remain poor8, 9.
The mechanism underlying the drug resistance of M. abscessus involves several factors, with efflux pumps being the most crucial6, 10. Bioinformatics analysis has identified at least 386 transport proteins within the M. abscessus genome11, 12, of which approximately 18% are putative drug efflux pumps13. In addition, efflux pump inhibitors like verapamil and carbonyl cyanide m‐chlorophenylhydrazine enhance the antimicrobial efficacy of antibiotics against M. abscessus clinical isolates14, 15, 16; however, the roles and mechanisms of action of efflux pumps in drug‐resistant M. abscessus remain to be elucidated.
Genetic approaches, such as recombineering, transposon mutation, and specialized transduction, have been developed for genome editing of M. abscessus 17, 18, 19, 20; however, these approaches require multiple steps to generate mutants with low efficiency, making it inconvenient to edit the genome of M. abscessus. Therefore, there is an urgent need to develop efficient genome‐editing technology for M. abscessus. Recently, we developed effective CRISPR‐mediated genome‐editing methods for Mycobacterium smegmatis, Mycobacterium marinum, and Mycobacterium tuberculosis based on the CRISPR‐Cas system21, 22, 23, 24. These technologies laid an important foundation for the recent development of CRISPR‐assisted genome‐editing tools for M. abscessus 21, 22, 23, 24; however, these methods have some disadvantages: (1) the transformation efficiency in M. abscessus is relatively low; (2) the CRISPR‐assisted genome‐editing tool has a high false‐positive rate and low efficiency; and (3) the CRISPR‐nonhomologous end joining (NHEJ)‐assisted method uses an integrated plasmid, making it challenging to generate clean M. abscessus mutants.
In this study, optimized CRISPR‐Cas9‐assisted tools were developed and used to create 66 efflux pump knockout mutants. The MAB_2806–2807 operon was identified as the primary determinant of both antimicrobial resistance and virulence. This study establishes a robust genome‐editing platform for M. abscessus and advances our understanding of efflux pump‐mediated resistance and pathogenicity.
RESULTS
Optimized electrocompetent M. abscessus cells
The current method used for preparing electrocompetent M. abscessus usually yields low transformation efficiency25, 26. Previous studies have shown that adding Ethionamide (ETH) to the culture medium increases transformation efficiency in M. marinum 27. ETH, a chemical drug that inhibits mycolic acid biosynthesis28, may disrupt the synthesis of the thick cell wall, thereby facilitating efficient DNA permeation during electroporation. Since Ethambutol (EMB) is another drug targeting cell wall biosynthesis29, we evaluated the effects of both ETH and EMB on the transformation efficiency of M. abscessus. Addition of ~20× 50% minimum inhibitory concentration (MIC_50_) of EMB (400 μg/ml) increased the transformation efficiency slightly, while addition of 40× MIC_50_ (20 μg/ml) of ETH increased the transformation efficiency by up to 40‐fold when M. abscessus was cultured for the preparation of competent cells (Figure 1).
*ETH improves the transformation efficiency of Mycobacterium abscessus. (A, B) Susceptibility of the indicated M. abscessus to the drug ETH (A) and EMB (B). The MIC50 was defined as the concentration of the compound that inhibited bacterial growth by 50% and was analyzed by a nonlinear fit model in GraphPad Prism 8.0. Data are expressed as the mean ± SD of triplicate samples and are representative of two independent experiments. (C) Transformation efficiency of M. abscessus when different concentrations of ETH or EMB were present in the culture medium used for competent cell preparation. ETH (10, 20, and 40 μg/ml) or EMB (200, 400, and 800 μg/ml) was added to the M. abscessus culture. The transformation efficiency was defined as the total number of CFU generated per transformation. Statistically significant differences were determined using multiple paired t‐tests. EMB, ethambutol; ETH, ethionamide. *p < 0.05; **p < 0.01; ***p < 0.0001.
Optimized CRISPR‐Cas9‐assisted recombineering of M. abscessus using the dual‐reporter detection system
To explore whether the CRISPR‐Cas9 system could assist in recombineering M. abscessus, we constructed an editing system comprising two plasmids: pJV53 expressing recombinant gp60 and gp61 proteins, and the shear plasmid pSGRNA encoding an sgRNA scaffold and the codon‐optimized Cas9 from Streptococcus thermophilus 30, 31. To test the efficiency of this system, we constructed a modified M. abscessus strain integrating the lacZ gene into its chromosome (Figure S1A). First, this system was tested by the introduction of a point mutation into lacZ via the recombination of single‐stranded DNA (ssDNA) oligonucleotide (Figure 2A). Upon successful genome editing, M. abscessus forms white colonies on an Bromo‐4‐chloro‐3‐indolyl‐β‐d‐galactopyranoside (X‐gal) plate; however, less than 11% of the transformants formed white colonies (Figure 2B).
Genome editing using CRISPR‐Cas9‐assisted recombineering in M. abscessus. (A) Schematic showing the lacZ‐targeting sgRNA and oligonucleotides used for recombineering. The oligonucleotides lacZ.lag (79 nt) and lacZ.lead (79 nt, sequences reverse and complementary to lacZ.lag) targeting the lagging strand and the leading strand of DNA replication were designed to introduce a continuous stop codon into the lacZ open reading frame, respectively. (B) Number of CFU per transformation and percentage of white transformant colonies after electroporation of the pSG‐lacZ‐1 plasmids and oligonucleotides into M. abscessus ATCC 19977 harboring the recombinase‐expression plasmid. lacZ‐1 and lacZ‐2 denote sgRNAs targeting two different sites in the lacZ gene. The transformants were plated on X‐gal plates for blue‐white screening. The black bar corresponds to the left Y axis, indicating the number of transformants, and the gray bar corresponds to the right Y axis, indicating the percentage of white colonies. Results represent the average of two independent experiments. (C) Transformation and recombination efficiency of the dual‐reporter detection system resulting from electroporation of the shear plasmid expressing gfp and the oligonucleotides targeting lacZ. Recombination efficiency was defined as the percentage of white colonies on the X‐gal plate. GFP positivity was calculated based on the proportion of GFP‐positive colonies. lacZ‐1 and lacZ‐2 denote sgRNAs targeting two different sites in the lacZ gene. (D) Transformation efficiency and recombination efficiency of gene deletion using CRISPR‐Cas9‐assisted dsDNA recombineering. Deletions of 620 bp were introduced into M. abscessus chromosomal DNA using approximately 1 kb dsDNA fragments. X‐gal, 5‐bromo‐4‐chloro‐3‐indolyl β‐d‐galactopyranoside.
M. abscessus is naturally resistant to a wide range of antibiotics, and bacteria growth cannot be inhibited completely even at a high concentration of antibiotics. We hypothesized that the relatively low recombineering efficiency observed in M. abscessus might be attributed to the fact that some colonies growing on the antibiotic plate did not actually contain the transformed plasmid. To address this issue, the green fluorescent protein (GFP) reporter gene was introduced into the shuttle plasmid to create a new shuttle plasmid pYC2405 (Figure S1B). Successful transformants harboring the shuttle plasmid could be identified easily through GFP fluorescence (Figure S2). Typically, co‐transformation of the oligonucleotide and shuttle plasmid generated ~10³ colony‐forming units (CFU), of which an average of 10% were GFP‐positive. About half of GFP‐positive transformants were genome‐edited successfully when the oligonucleotides targeted the leading strand, whereas 100% of the transformants were genome‐edited successfully when the oligonucleotides targeted the lagging strand (Figure 2C). Next, the system was tested for its ability to generate chromosomal deletions in M. abscessus using linear double‐stranded DNA (dsDNA) substrates. The results showed that 95% of GFP‐positive transformants were genome‐edited successfully using PCR products with ~500‐bp homology upstream and downstream arms of lacZ. Overall, these results indicate the applicability of our system for precise genome editing based on CRISPR‐assisted recombineering in M. abscessus (Figure 2D).
CRISPR‐NHEJ‐mediated genome editing in M. abscessus
A CRISPR‐assisted NHEJ system for efficient genome editing in mycobacteria was developed previously22. The system was modified by introducing the gfp reporter gene into the shear plasmid. This system was transferred into M. abscessus and its genome‐editing efficiency was evaluated. Consistent with previous findings in M. tuberculosis 22, high expression of the NHEJ machinery, or repression of RecA_mab_‐dependent homologous recombination (HR) by RecX overexpression alone, did not result in efficient genome editing (Figure 3A,B). The genome editing could only be achieved efficiently when the NHEJ machinery and RecX_mab_ were overexpressed together. It has been reported that NHEJ‐assisted genome editing can be achieved without these two additional components in M. abscessus and M. tuberculosis 32, 33. This might be because Cas9 was expressed on the chromosome via an integrating plasmid in these methods, which may prevent leaky expression of Cas9 and enhance editing efficiency. Consistent with this hypothesis, addition of a ssrA‐tag to the C‐terminal of Cas9 to avoid leaky expression can increase NHEJ‐mediated genome‐editing efficiency in M. tuberculosis 22.
Genome editing using CRISPR‐Cas9‐assisted NHEJ in M. abscessus. (A, B) Transformation efficiency (A) and editing efficiency (B) by CRISPR‐Cas9 cleavage in M. abscessus expressing pMV261, pMV261‐RecX, pMV261‐NHEJ, or pNHEJ‐RecX. Shear plasmid expressing lacZ sgRNA or the control was transformed into M. abscessus containing the indicated helper plasmids. Editing efficiency was calculated based on the proportion of white colonies among GFP‐positive cells. The values represent the data from two independent experiments. (C) Genome‐editing efficiency of lacZ with different PAMs. Eleven sgRNAs with different PAM‐targeting template or a non‐template strand of lacZ were transformed into M. abscessus along with the pNHEJ‐RecX. The black bar corresponds to the left Y axis, indicating the number of transformants, and the gray bar corresponds to the right Y axis, indicating the editing efficiency. “T” denotes targeting the template of the lacZ gene, and “RT” denotes targeting the non‐template. The PAMs used by these sgRNA were shown in Figure S3B. (D) Deletion length distribution and mutation length within the indicated genes. The bars on the left represent the median deletion size for each strain. The bars on the right indicate the size of deletions located PAM‐distal to the cleavage site; >95% of all mutations fell within this PAM‐distal category. NHEJ, non‐homologous end joining; PAM, protospacer adjacent motif.
To learn more about the editing characteristics of this system in M. abscessus, eleven sgRNAs were designed, targeting different protospacer adjacent motifs (PAMs) or the template and non‐template strands of the lacZ gene, to assess the editing efficiency (Figure S3A). Consistent with previous findings in M. tuberculosis, CRISPR‐assisted knock‐out showed no strand bias, and genome editing with sgRNA targeting of the template or non‐template strand demonstrated similar efficiency in M. abscessus. The efficiency of genome editing decreased when weak PAM sites were used. More than 58% of GFP‐positive transformants were genome‐edited when PAM sites with a repression‐fold > 42.2 were used (Figures 3C and S3B). The editing efficiency was decreased to ~21% when a PAM site with a repression‐fold of around 20 was used, and no genome editing was detected when a PAM site had a repression‐fold of 7.9 (Figure S3B). Analysis of the end‐joining patterns showed that the CRISPR‐Cas9 assisted NHEJ in M. abscessus usually produced random deletions around the cleavage site that varied in length from 1 to over 10 kb, and more than 95% were PAM‐distal mutations (Figure 3D). Overall, these findings indicate CRISPR‐assisted NHEJ genome editing is a highly effective approach for robust markerless genome editing in M. abscessus.
Investigation of the antibiotic resistance afforded by multidrug efflux pumps in M. abscessus
M. abscessus possesses a variety of multidrug efflux pumps, which might prevent a drug from accumulating within the bacterium, thereby contributing to drug resistance mechanisms. To systematically evaluate the roles of efflux pumps in the drug resistance of M. abscessus, the efflux pump‐encoding genes were knocked out using the above developed CRISPR‐assisted NHEJ genome‐editing method. Seventy efflux pump‐encoding genes, including members of the major facilitator family, ABC transporters, the small multidrug resistance family, and a family of lipophilic drug efflux proteins were selected to construct mutants. Of these, 66 mutants were successfully obtained, while attempts to knock out MAB_1338, MAB_1409, MAB_1429, and MAB_1560 failed. Among them, MAB_1560 is reported to be an essential gene34. Others may also be essential genes or required for normal growth of M. abscessus and thus their mutants were not obtained in our study.
To explore the impact of efflux pumps on antibiotic resistance in M. abscessus, drug resistance was assessed in a disk diffusion assay (DDA) comparing the sensitivity of the mutants and the wild type (WT) to various antibiotics, including clarithromycin (CLR), linezolid (LZD), amikacin (AMK), cefoxitin (FOX), bedaquiline (BDQ), and clofazimine (CFZ). Several mutants displayed increased susceptibility to different antibiotics, including MAB_2807, MAB_3142c, MAB_3384, and MAB_1846, which had been identified previously as being associated with intrinsic drug resistance14, 35, 36. Although MAB_2633, MAB_2736, MAB_1275, and MAB_1396 are upregulated upon exposure to subinhibitory concentrations of AMK, CFZ, and FOX37, their mutants did not show increased susceptibility to these antibiotics (Table 1)37. Many genes were identified as being associated with antibiotic resistance for the first time, including MAB_0183c, MAB_1065, MAB_1226, MAB_1359c, MAB_1414c, MAB_1415c, MAB_2217, MAB_4064, and MAB_4849c (Table 1).
Deletion of the MAB_2806–2807 operon results in drug susceptibility
Genetic deletion of MAB_2807 resulted in the most significant increase in antibiotic susceptibility to all six antibiotics (Table 1). MAB_2807 and the lipoprotein LprG (MAB_2806) are encoded by a conserved two‐gene operon. It is suggested that the orthologs of MAB_2806 and MAB_2807 in M. tuberculosis, Rv1411c (lprG) and Rv1410c, respectively, are involved in inserting triacylglycerides and lipoarabinomannans into the outer membrane38. It has been reported that mutation of Rv1410c and lprG increased cell permeability, resulting in greater drug susceptibility and reduced virulence in M. tuberculosis 38, 39. To explore the role of MAB_2806 and MAB_2807, single and double mutants were constructed. Deletion of MAB_2806, MAB_2807, or MAB_2806–2807 resulted in a pronounced increase in antibiotic susceptibility to FOX, LZD, and CLR (Figure 4A). Complementation of the mutant phenotype was achieved only by the entire operon, not by the single gene, indicating that MAB_2806 and MAB_2807 might function together (Figure 4A). Consistent with previous findings in M. tuberculosis, an ethidium bromide (EtBr) uptake assay showed that mutation of MAB_2806–2807 increased uptake of EtBr when compared with the WT, suggesting that the deletion of MAB_2806–2807 increased cell permeability in M. abscessus (Figure 4B). The role of MAB_2807 was further investigated in a clinical M. abscessus isolate. The ortholog of MAB_2807 was successfully deleted from M. abscessus clinical isolate 108 using the CRISPR‐NHEJ method only when competent cells were prepared with the addition of ETH (Figure S4). Deletion of the ortholog of MAB_2807 significantly increased drug susceptibility to all six tested antibiotics in M. abscessus clinical isolate 108 (Figure S5). Collectively, these findings highlight the importance of the MAB_2806–2807 operon in the inherent drug resistance of M. abscessus, providing new insights into potential therapeutic targets.
*MAB_2806–2807 is required for antibiotic resistance of Mycobacterium abscessus. (A) Disk diffusion assay of the MAB_2806‐2807 operon to different antibiotics. The zone of inhibition for FOX, LZD, and CLR was measured for ΔMAB_2806, ΔMAB_2807, and ΔMAB_2806–2807, as well as their single‐ or double‐gene complementary strains. Statistically significant differences were determined using a paired two‐tailed t‐test. All p values were calculated by comparison with the wild type. **p < 0.01; ***p < 0.001; and ***p < 0.0001. (B) Uptake of EtBr by wild type, ΔMAB_2806‐2807, and the complementary strain. CLR, clarithromycin; EtBr, ethidium bromide; FOX, cefoxitin; LZD, linezolid.
MAB_2806–2807 is required for drug resistance and virulence of M. abscessus in Galleria mellonella
G. mellonella is gaining prominence as a cost‐effective, manageable, and ethically simplified model for M. abscessus research40. To explore the function of efflux pumps on the virulence of M. abscessus, the mutants were transformed with barcode‐containing integrative plasmids and then mixed before infecting G. mellonella larvae for 3 and 5 days (Figure 5A). The M. abscessus cells were then collected from infected G. mellonella and analyzed using next‐generation sequencing. It was found that the efflux pump genes MAB_2622c, MAB_2806–2807, MAB_2807, MAB_2958, MAB_3142c, MAB_3871c, and MAB_4500 were required for the survival of M. abscessus in G. mellonella after both 3 and 5 days of infection (Figure 5B and Table S1). Among these, MAB_2806–2807 was the gene most depleted during the screening (Figure 5B). To confirm the role of MAB_2806–2807 in the virulence of M. abscessus, G. mellonella larvae were infected with 10^6^ CFU of the WT strain, ΔMAB_2806, ΔMAB_2807, ΔMAB_2806–2807, and complementation strains, respectively (Figure 6A). The WT and complementation strains showed rapid larval death progression, resulting in 100% mortality within 10 days; however, ΔMAB_2806, ΔMAB_2807, and ΔMAB_2806–2807 infections caused less than 30% mortality over the same period.
Screening of efflux pump genes affecting the survival of M. abscessus in Galleria mellonella larvae. (A) Screening of an M.abscessus efflux pump mutant library in G. mellonella larvae. Efflux pump mutants and the wild‐type ATCC 19977 strain of M. abscessus harboring barcodes were mixed and used to infect G. mellonella larvae. At 3 and 5 days postinfection, the larvae were homogenized and plated. The plated bacteria were collected, and the barcode sequences were amplified for next‐generation sequencing analysis. Created in BioRender. Duan, D. (2025) https://BioRender.com/bi8qu91. (B) Volcano plots depict the log2FC values and p values of each mutant from the library at 3 days (left) and 5 (right) days postinjection of G. mellonella larvae. Blue dots represent mutants depleted from G. mellonella larvae, while red dots represent those that are enriched. Gray dots indicate mutants showing no significant difference. The horizontal dashed line indicates a p‐value of 0.05, while the left and right vertical dashed lines indicate log2FC of −1 and 1, respectively.
*The MAB_2806–2807 operon affects the survival of M. abscessus in G. mellonella larvae. (A) Survival of G. mellonella larvae injected with 106 CFU of WT, ΔMAB_2806, ΔMAB_2807, ΔMAB_2806–2807, or the complementary M. abscessus strain, respectively (n = 10). (B) Assessment of the effect of ΔMAB_2806‐2807 deletion on the efficacy of antibiotics in the G. mellonella model. The effect of treatment on bacterial counts (CFU) was measured on Day 5 postinfection. Statistically significant differences were determined using a paired two‐tailed t‐test. **p < 0.01 and p < 0.1.
To explore the role of MAB_2806–2807 in drug efficacy in vivo, G. mellonella larvae were infected with WT, ΔMAB_2806–2807, or complementation strains and then treated with different drugs on Day 2 postinfection. The larvae were kept at 37°C afterward and killed on Day 5 for CFU counting (Figure 6B). Deletion of MAB_2806–2807 strongly sensitized M. abscessus to antibiotics, particularly LZD and CFZ, leading to a more pronounced reduction in CFU. Taken together, these data suggested that MAB_2806 and MAB_2807 may serve as drug targets to enhance antibiotic effectiveness against M. abscessus.
The survival of the MAB_2806–2807 mutant is attenuated in mouse
To further investigate the role of MAB_2806–2807 in the virulence of M. abscessus, experiments were carried out using infected human THP‐1 macrophages. The ΔMAB_2806–2807 mutant showed significantly lower survival rates within macrophages compared to the WT and complementary strains at 24 h postinfection (Figure 7A). Next, the role of MAB_2806–2807 was evaluated using the severe combined immunodeficiency (SCID) mouse model, which was chosen for its permissiveness to M. abscessus infection41. Mice were infected with M. abscessus via aerosolization, and ethically euthanized on Day 1 to determine the initial bacterial load, or on Days 21 and 28 to evaluate bacterial burden in the lungs. No bacteria were detected in the lungs of mice infected with ΔMAB_2806–2807 at 21 and 28 days postinfection (dpi); by contrast, there was only a slight reduction in the CFU in the lungs of mice infected with the WT or complementation strains at 21 and 28 dpi compared with 1 dpi (Figure 7B). The histopathological characterization revealed that alveolar septal thickening and hyperemia of capillary were observed in the lung tissue of WT group after 28 days of infection (Figure S6). However, the alveolar structure in the lung tissue was clear in the group of ΔMAB_2806–2807 (Figure S6). Overall, the results indicate that the MAB_2806–2807 operon is crucial for the virulence of M. abscessus.
*The MAB_2806–2807 operon is required for the virulence of M. abscessus. (A) MAB_2806–2807 is required for the survival of M. abscessus in macrophages. THP‐1 macrophages were infected with WT, ΔMAB_2806–2807, or the complementary strain (MOI = 10). Intracellular growth in macrophages was assessed by counting CFU after 6 and 24 h. Data represent the mean ± SD from four independent experiments. Statistically significant differences were determined using one‐way ANOVA with Tukey's multiple comparisons test. **p < 0.01; ns, not significant. (B) Comparative analysis of bacterial burden (at 1, 21, and 28 dpi) in the lungs after autopsy of SCID mice infected with 105 CFU of WT, ΔMAB_2806–2807, and its complementary strain. Data points represent individual mice. **p < 0.001.
DISCUSSION
Traditional methods used for the genetic manipulation of M. abscessus are typically time‐consuming and labor‐intensive. Recently, genome editing based on CRISPR/Cas9 has been developed for M. abscessus33, 42. However, this approach necessitates the use of an integrative plasmid harboring a Cas9 expression cassette, as well as a kanamycin‐resistance gene. This plasmid is integrated into the chromosome of M. abscessus and cannot be removed from the genome‐edited strain. The CRISPR/Cas9‐assisted recombineering strategy is inefficient due to high rates of spontaneous background antibiotic resistance, necessitating the selection of numerous colonies to obtain the correct mutants43. In addition, the electrotransformation efficiency of M. abscessus is relatively low. In this study, we optimized the method for preparing competent cells, which allows for high‐efficiency electrotransformation of M. abscessus. We also optimized the CRISPR‐assisted genome‐editing tools to avoid their disadvantages, thereby achieving efficient recombineering‐based precision genome editing and NHEJ‐based template‐free genome editing in M. abscessus.
Mycobacteria possess a thick cell wall, which affects transformation efficiency, as it is dependent on the permeability of the competent cells to foreign DNA. Various methods have been applied to increase the permeability of the cell wall. Addition of glycine improves the transformation efficiency of M. tuberculosis, Mycobacterium avium, and Mycobacterium aurum. Glycine affects peptidoglycan synthesis and decreases cell wall cross‐linking, thereby increasing cell permeability44, 45, 46, 47. However, the addition of glycine did not significantly affect the transformation of M. abscessus (data not shown). The addition of ETH increased the transformation efficacy of M. marimum by ~2‐fold. Interestingly, the addition of 20 μg/ml ETH increased transformation efficiency by up to ~40 fold, achieving more than 10^4^ transformants per electroporation. This level of transformation efficiency enables CRISPR library construction and screening of M. abscessus, thereby facilitating functional genome research in this organism.
M. abscessus shows significant spontaneous resistance to diverse antibiotics, particularly when its genome is edited using CRISPR‐assisted recombineering methods (Figure 2C,D). This could explain the low effectiveness of CRISPR‐assisted recombineering in genome editing. Introducing the GFP reporter gene into the shear plasmid solved this issue and simplified the procedure for obtaining the correct recombinants. The efficiency of CRISPR‐NHEJ‐mediated genome editing is relatively high in M. abscessus ATCC 19977, and thus it has not been significantly affected by spontaneous resistance (Figure 3A,B). However, the efficiency of CRISPR‐NHEJ‐mediated genome editing might be lower in M. abscessus clinical isolates. Therefore, employing this strategy based on the GFP reporter could be useful for obtaining the desired mutants.
To systematically investigate the roles of genes in vivo and in vitro, transposon insertion sequencing (Tn‐Seq) and CRISPR screening have been used widely in bacteria. Tn‐Seq can only be used for whole‐genome screening, while CRISPR screening can investigate genes associated with specific cellular pathways. In this study, a small mutant library containing 66 efflux pump gene mutants was constructed in M. abscessus using the CRISPR‐NHEJ‐mediated genome‐editing method, followed by the investigation of their roles. Several efflux pumps were identified as being involved in intrinsic drug resistance. Consistent with previous findings that efflux pumps contribute to drug resistance by preventing intracellular accumulation of antibiotics in M. abscessus and other bacteria14, 48, 49, 50, 51, each efflux pump mediated intrinsic resistance to one or several drugs; however, MAB_2807 strongly affected cell permeability and was therefore involved in resistance to all tested drugs. Similar to Rv1410 and lprG, the corresponding orthologs in M. tuberculosis, MAB_2807 and MAB_2806, are required for drug resistance and virulence. Although efflux pumps are usually associated with drug resistance, none of the others have been noted to contribute to the virulence in M. abscessus. The data from the G. mellonella model suggest that several other efflux pumps are required for the virulence of M. abscessus, although they still need to be verified in mouse models.
In summary, we constructed and optimized tools for transformation and genome editing that may facilitate molecular biology studies in M. abscessus. A small efflux pump mutant library was constructed in M. abscessus using the developed genome‐editing tools, leading to the identification of several efflux pump genes involved in drug resistance and virulence. Among them, MAB_2807 was the most sensitizing gene for various drugs and essential for the virulence of M. abscessus. Our study suggests that lprG and MAB_2807 might be potential targets for new drugs to treat M. abscessus. Additionally, these findings may help us to better understand the roles of efflux pumps in drug resistance and virulence of M. abscessus.
MATERIALS AND METHODS
Strains, media, and growth conditions
Two strains of M. abscessus were utilized in this study: the ATCC 19977 strain and the Clinical isolate 108, both archived in the Beijing Key Laboratory for Drug Resistance Tuberculosis Research of Beijing Chest Hospital. The bacteria were cultured at 37°C with shaking in Middlebrook 7H9 broth (Difco) containing 0.05% Tween 80, 0.2% glycerol, and OADC (oleic acid‐albumin‐dextrose‐catalase). Alternatively, they were cultured on 7H10 agar plates (Difco) supplemented with 0.5% glycerol and OADC, and appropriate antibiotics (100 μg/ml kanamycin, 100 μg/ml bleomycin, or 500 μg/ml hygromycin). An appropriate concentration of Anhydrotetracycline (ATc) (100 ng/ml) was added to the M. abscessus cultures when necessary.
Plasmids
Plasmids pSGRNA, pJV53, pMV261‐RecX, pMV261‐NHEJ, pNHEJ‐Recx, and pYC3189 were used. The shear plasmid pSGRNA was constructed to express the codon‐optimized Cas9, regulated by a TetR‐controlled promoter, while the sgRNA cassette was regulated by an optimized TetR‐controlled promoter30. The sgRNA cassette features two BbsI restriction sites, enabling target sequence insertion. The MmNHEJ machinery genes (MMAR_4573, MMAR_4574, and MMAR_4575) were amplified from the M. marinum chromosome and cloned into pMV261, yielding pMV261‐NHEJ. Additionally, the recX allele from the M. abscessus chromosome was cloned into pMV261 and pMV261‐NHEJ to generate pMV261‐RecX and pNHEJ‐RecX, respectively, via seamless cloning.
The dual‐reporter detection system
To facilitate screening of M. abscessus mutants, the lacZ gene was integrated into the MAB_1579‐MAB_1580 locus via the L5 integrating plasmid, under the control of the Psmyc promoter. To identify successful transformants, the GFP reporter gene was cloned into pSGRNA by seamless cloning to yield plasmid pYC2405. Successful transformants of pYC2405 could be easily identified through green fluorescence excited by a fluorescence flashlight at 440–460 nm.
Preparation of recombinogenic DNA
Recombinogenic 79‐nt oligonucleotides, harboring targeted mutations centrally within their sequences, were designed for ssDNA recombineering studies. Cumulative skew diagrams were used to identify the leading and lagging strands of M. abscessus chromosomes52. In conducting the dsDNA recombineering experiments, a 620‐bp deleted lacZ fragment, containing 500‐bp homologous arms of the flanking regions, was amplified to generate recombinogenic dsDNA products.
Preparation of electrocompetent cells
Initially, M. abscessus ATCC 19977 was recovered on 7H10 solid medium. Subsequently, a single colony was inoculated into 3 ml of 7H9 broth containing 0.05% Tween 80 and cultured to saturation at 37°C with shaking at 200 rpm. The saturated cultures were then shaken at 70 rpm until they reached an optical density at 600 nm (OD_600_) of 0.7 after being diluted in 100 ml of 7H9 broth to an OD_600_ of 0.03. The cultures were supplemented with various concentrations of ETH (10, 20, or 40 μg/ml) or EMB (200, 400, or 800 μg/ml), and incubated continuously for 5 h. Among these concentrations, the addition of ETH at 20 μg/ml was found to enhance the editing efficiency most significantly. The cultures were then centrifuged at 3700 g at 4°C for 10 min, and the harvested cells were washed three times in ice‐cold 10% glycerol supplemented with 0.05% Tween 80 (centrifugation at 3700g for 10 min each time). Finally, the washed cells were gently resuspended in 1/100 the original volume (i.e., 1 ml) of 10% glycerol. The prepared competent cells were aliquoted (200 μl per aliquot) and stored at −80°C.
Assessment of transformation efficiency in M. abscessus
To detect the transformation efficiency in M. abscessus, ~200 ng DNA of plasmid pYC3189 containing the GFP reporter and zeocin resistance gene was electroporated into the competent cells. Consequently, the successful transformants were identified visually as green colonies on the plate with zeocin.
CRISPR‐Cas9‐assisted recombineering
Cells harboring pJV53 were cultivated in 5 ml of 7H9 containing 0.05% Tween 80, 0.2% glycerol, OADC, and 100 μg/ml kanamycin. The resultant culture was inoculated into 7H9 medium supplemented with 0.2% (w/v) succinate at an OD_600_ of 0.03. The culture was incubated at 37°C with shaking at 70 rpm until the OD_600_ reached 0.6, at which point 0.2% (w/v) acetamide was added. Incubation was continued for an additional 5 h until a final OD_600_ of approximately 1.0 was reached. The preparation of competent cells of M. abscessus was described as previously mentioned26. For recombineering, 200 ng of shear plasmid and 600 ng of oligonucleotides were mixed and electroporated into pJV53‐containing competent cells. Electroporation was performed under the following conditions: 2500 V, 25 µF, and 800 Ω. After electroporation, the cells were immediately mixed with 1 ml of 7H9 containing OADC and incubated overnight at 37°C. Then, the samples were plated on 7H10 medium supplemented with 50 ng/ml ATc, corresponding antibiotics, and OADC. Recombination of lacZ was assessed by examining the white colonies on 7H10 agar supplemented with X‐gal, and then confirmed by PCR and sequencing analysis. To cure plasmids, the recombinant colony was inoculated in 7H9 containing OADC but without antibiotics and cultured at 37°C to saturation. After that, the cultures were diluted, plated on 7H10 medium supplemented with OADC, and incubated for 5 days. Single‐colony streaking on plates with or without the corresponding antibiotics was used to check for plasmid loss.
CRISPR‐NHEJ‐mediated genome editing
Cells harboring pMV261‐RecX, pMV261‐NHEJ, or pNHEJ‐RecX were used to prepare competent cells as described above. Electroporation of 600 ng of the shear plasmid into the competent cell was performed as described above. Following electroporation, the cells were recovered and cultured overnight at 37°C in 1 ml of 7H9 supplemented with OADC. Subsequently, these cultures were plated onto 7H10 medium containing 50 ng/ml ATc, 100 μg/ml zeocin, 100 μg/ml kanamycin, and OADC. The successful transformants were identified by GFP fluorescence and selected for PCR and sequencing. The method to cure plasmids from the mutant was performed as described above.
Determination of MICs for ETH and EMB
ETH and EMB were prepared in 90% DMSO, and subjected to twofold serial dilutions in 96‐well microtiter plates (Falcon, #3072). Bacterial cultures were grown to the mid‐logarithmic phase and diluted to an OD_600_ of 0.0035. Subsequently, 100 μl of each adjusted culture was dispensed into wells containing the indicated drug concentrations and incubated for 48–72 h. Bacterial growth was assessed by measuring OD_600_ using an EnVision multimode microplate reader (PerkinElmer). Relative growth was calculated relative to the DMSO vehicle control for each strain and MIC_50_ values were determined by curve fitting in GraphPad Prism 10. All measurements were performed in triplicate.
Disk diffusion assay
The susceptibility of M. abscessus strains to various antibiotics was assessed using the DDA. The M. abscessus ATCC 19977 WT and efflux pump mutant strains were cultured in 7H9 without OADC until the OD_600_ reached ~1.0, followed by dilution to an OD_600_ of ~0.7. Clinical isolate 108 and its mutant, which showed growth deficiency, were cultured in 7H9 medium supplemented with OADC. Before plating onto 7H10 medium, 500 μl of bacterial culture was mixed with 2500 μl of 7H9 and 3 ml of 7H9 soft agar. Sterile filter paper disks saturated with 10 μl of CLR (125 μg/ml), LZD (200 μg/ml), AMK (1000 μg/ml), FOX (2500 μg/ml), BDQ (250 μg/ml), and CFZ (1250 μg/ml) were placed onto the agar. The plates were incubated at 37°C for 72 h. The diameter of the zones of growth inhibition was recorded and analyzed statistically.
G. mellonella larvae infection and treatment
G. mellonella larvae were grown at 37°C before inoculation. For inoculation, the left hind leg of the larvae was disinfected using 75% alcohol and then injected with 10^6^ CFU (or as indicated) using a 4 G needle (10 μl). After inoculation, the infected larvae were maintained at 37°C and the survival rate was recorded daily. Typically, 1 to 3 larvae were killed immediately after inoculation to verify the actual inoculum. Antibiotics (BDQ, 10 mg/kg; FOX, 25 mg/kg; LZD, 40 mg/kg; AMK, 25 mg/kg; CFZ, 10 mg/kg; or CLR, 25 mg/kg) were administered by injection into the right hind leg 2 days postinoculation. After 3 days, larvae were killed and plated individually to determine the CFU. Larvae were killed by immersion in water for 10 min, decontaminated twice with 75% ethanol, and washed with sterile water before being homogenized in a 5 ml conical tube using a grinding mill. Serial dilutions were prepared and plated to determine the CFU number.
Screening of the M. abscessus mutant library in G. mellonella larvae
Plasmids containing a 10‐base‐pair barcode sequence were constructed and transformed into M. abscessus ATCC19977 WT and the efflux pump mutants. The mixed barcode‐containing M. abscessus were cultured, followed by an injection of ~10^5^ CFU into the last right pseudopodia of G. mellonella larvae. After inoculation, the infected larvae were incubated at 37°C. At the indicated time points (3 and 5 days), 10 randomly selected larvae were individually euthanized, homogenized, and plated. Each larva was euthanized using 5 ml of pre‐cooled PBS and decontaminated (sequentially) with 3% hydrogen peroxide, 70% ethanol, and 95% ethanol. The selected larvae were then homogenized in 2 ml of PBS using a homogenizer comprising five metal beads in a 5 ml lysing tube (70 Hz for 20 s). All homogenates were filtered through a cell sieve to remove insect residue. Then, the liquid was centrifuged at 4000 rpm for 5 min to eliminate insect oil components. The remaining liquid was used for plating. Plates were incubated at 37°C for 4 to 5 days. Approximately 2 × 10^5^ CFU cells were scraped and collected for extraction of genomic DNA. The barcode sequences were amplified using the extracted genomic DNA as a template and analyzed by next‐generation sequencing.
Macrophage infection
RPMI 1640 medium, supplemented with 10% heat‐inactivated fetal bovine serum, l‐glutamine, and 25 mM HEPES, was used to cultivate THP‐1 cells at 37°C under 5% CO₂. Cells were collected in 50 ml centrifuge tubes by centrifugation at 200 rpm for 2 min, and THP1 monocytes were induced to differentiate into THP1‐derived macrophages by the addition of PMA (1:20,000; 50 ng/ml). The cells were then inoculated into 12‐well cell culture plates. After 36–48 h of PMA‐induced cell differentiation, the cell culture medium was removed, and the cells were washed three times with fresh RPMI 1640 medium. Subsequently, 1 ml of RPMI 1640 medium was added to each well. The macrophages were infected with WT and mutant strains at a multiplicity of infection (MOI) of 10, under standard conditions (37°C, 5% CO_2_). After 6 or 24 h of infection, the medium was removed and the cells were washed three times with sterile PBS before being lysed with 200 μl of 0.5% triton lysis buffer. Then, the lysate was doubly diluted and spread on 7H10 agar plates supplemented with OADC. Three parallel wells were set up at a time, and all experiments were repeated at least four times.
M. abscessus infection of model mice
SCID mice were infected via aerosol challenge. WT, ΔMAB_2806–2807, and a complementary strain were cultured to an OD_600_ ~0.7, corresponding to approximately 2.5 × 10^8^ CFU/ml. The 2 ml bacterial suspension was subjected to three sterile PBS buffer washes, followed by centrifugation for bacterial collection. The bacteria were resuspended in 5 ml of PBS containing 0.05% Tween‐20. SCID mice aged 5–7 weeks were infected with a wild‐type, ΔMAB_2806–2807, or complementary strain via aerosol infection. Infected mice were killed after 24 h to assess the initial lung bacterial load. On Days 21 and 28 postinfection, lungs and spleens were collected, homogenized in sterile‐distilled water, and subjected to 10‐fold serial dilutions before plating on 7H10 agar plates for CFU counting.
AUTHOR CONTRIBUTIONS
Sishang Li: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; software; supervision; validation; visualization; writing—original draft; writing—review and editing. Aofei Duan: Conceptualization; data curation; formal analysis; investigation; methodology; supervision; validation; visualization; writing—original draft; writing—review and editing. Lanyue Zhang: Conceptualization; data curation; investigation; methodology; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Chunliang Wang: Writing—review and editing. Meiyi Yan: Writing—review and editing. Gai‐Xian Ren: Writing—review and editing. Li‐Ping Pan: Conceptualization. Yi‐Cheng Sun: Conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; project administration; resources; software; supervision; validation; visualization; writing—original draft; writing—review and editing.
ETHICS STATEMENT
Four‐ to six‐week‐old female CB17‐SCID mice, a model for severe combined immunodeficiency, were sourced from Beijing Vital River Laboratory Animal Technology Co. Ltd. These mice were maintained in a controlled, pathogen‐free environment throughout the study. The research procedures adhered to ethical standards and received formal approval from the Ethics Committee of Beijing Chest Hospital, affiliated with Capital Medical University (Permit number 2022‐075).
CONFLICT OF INTERESTS
The authors declare no conflict of interests.
Supporting information
Supplementary figure 1. Dual‐reporter detection system in Mycobacterium abscessus. (A) Schematic showing integration of lacZ (under control of the psmyc promotor) into the genome of M. abscessus. (B) Schematic of the shear plasmid pYC2405. The plasmid contains a zeocin resistance gene. The plasmid expresses Cas9 from Streptococcus thermophilus, and the GFP reporter gene under the control of constitutive promoter psmyc. The plasmid also expresses an sgRNA cassette. Supplementary figure 2. GFP reporter detection system. Successful transformation of the plasmid into Mycobacterium abscessus was determined by measuring GFP fluorescence. The red arrows point to the clones with a GFP signal, indicating that the plasmid was transformed successfully. The black arrows point to clones without a GFP signal. Supplementary figure 3. The PAM and sgRNA sequences used to target the target * lacZ * gene. (A) Schematic illustrating 11 sgRNAs with distinct PAMs used to target the lacZ gene. The arrows depict sgRNAs and the orientation of the arrows corresponds to the orientation of the sgRNA sequences. "T" denotes targeting the template of the lacZ gene, and "RT" denotes targeting the non‐template. (B) List of PAM sequences used to target the lacZ gene. Supplementary figure 4. CRISPR‐Cas9‐assisted NHEJ genome editing in * M. abscessus * clinical isolate 108. (A, B) Transformation (A) and editing efficiency (B) of Mycobacterium abscessus clinical isolate 108 expressing pNHEJ‐RecX via CRISPR‐Cas9 cleavage targeting MAB_2807, with or without the addition of 20 μg/ml ETH to the culture medium during competent cell preparation. Editing efficiency was determined via PCR and sequencing analysis. Supplementary figure 5. Disk diffusion Assay of the * MAB_2807 * mutant in Mycobacterium abscessus clinical isolate 108. Plate images show the sensitivity of M. abscessus clinical isolate 108 (upper) and its MAB_2807 mutant (lower) to six antibiotics using a disc diffusion assay. Supplementary figure 6. Pulmonary pathology of lungs in SCID mice. The lungs of uninfected SCID mice, and after 28 days of infection with the wild type or ΔMAB_2806‐2807. The images show the slides of lungs from control and infected mice stained with hematoxylin–eosin. The black arrow points to the thickened tissue. Magnification: ×10 on the left and ×40 on the right.
Supplementary Table 1. Log2FC values and P values of Mutant strains from the library with statistically decreased abundance (p < 0.05, log2FC < − 1) at 3 and 5 days post‐injection of Galleria mellonella larvae.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Wallace, Jr. RJ , Brown BA , Griffith DE . Nosocomial outbreaks/pseudo‐outbreaks caused by nontuberculous mycobacteria. Annu Rev Microbiol. 1998;52:453–490.9891805 10.1146/annurev.micro.52.1.453 · doi ↗ · pubmed ↗
- 2Newman MI , Camberos AE , Clynes ND , Ascherman JA . Outbreak of atypical mycobacteria infections in U.S. patients traveling abroad for cosmetic surgery. Plast Reconstr Surg. 2005;115:964–965.15731722 10.1097/01.prs.0000153818.99552.cc · doi ↗ · pubmed ↗
- 3Esther, Jr. CR , Esserman DA , Gilligan P , Kerr A , Noone PG . Chronic Mycobacterium abscessus infection and lung function decline in cystic fibrosis. J Cyst Fibros. 2010;9:117–123.20071249 10.1016/j.jcf.2009.12.001PMC 3837580 · doi ↗ · pubmed ↗
- 4Bryant JM , Grogono DM , Rodriguez‐Rincon D , Everall I , Brown KP , Moreno P , et al. Emergence and spread of a human‐transmissible multidrug‐resistant nontuberculous mycobacterium. Science. 2016;354:751–757.27846606 10.1126/science.aaf 8156 PMC 5142603 · doi ↗ · pubmed ↗
- 5Floto RA , Olivier KN , Saiman L , Daley CL , Herrmann JL , Nick JA , et al. US Cystic Fibrosis Foundation and European Cystic Fibrosis Society consensus recommendations for the management of non‐tuberculous mycobacteria in individuals with cystic fibrosis: executive summary. Thorax. 2016;71:88–90.26678435 10.1136/thoraxjnl-2015-207983 PMC 4717423 · doi ↗ · pubmed ↗
- 6Nessar R , Cambau E , Reyrat JM , Murray A , Gicquel B . Mycobacterium abscessus: a new antibiotic nightmare. J Antimicrob Chemother. 2012;67:810–818.22290346 10.1093/jac/dkr 578 · doi ↗ · pubmed ↗
- 7Chen J , Zhao L , Mao Y , Ye M , Guo Q , Zhang Y , et al. Clinical efficacy and adverse effects of antibiotics used to treat Mycobacterium abscessus pulmonary disease. Front Microbiol. 2019;10:1977.31507579 10.3389/fmicb.2019.01977 PMC 6716072 · doi ↗ · pubmed ↗
- 8Kurz SG , Zha BS , Herman DD , Holt MR , Daley CL , Ruminjo JK , et al. Summary for clinicians: 2020 clinical practice guideline summary for the treatment of nontuberculous mycobacterial pulmonary disease. Ann Am Thorac Soc. 2020;17:1033–1039.32870060 10.1513/Annals ATS.202003-222CME · doi ↗ · pubmed ↗