Targeted Chromosomal Barcoding Establishes Direct Genotype-Phenotype Associations for Antibiotic Resistance in Mycobacterium abscessus

Abstract

A bedaquiline-resistant Mycobacterium abscessus isolate was sequenced, and a candidate mutation in the atpE gene was identified as responsible for the antibiotic resistance phenotype. To establish a direct genotype-phenotype relationship of this mutation which results in a Asp-to-Ala change at position 29 (D29A), we developed a recombineering-based method consisting of the specific replacement of the desired mutation in the bacterial chromosome. As surrogate bacteria, we used two M. abscessus bedaquiline-susceptible strains: ATCC 19977 and the SL541 clinical isolate. The allelic exchange substrates used in recombineering carried either the sole D29A mutation or a genetic barcode of silent mutations in codons flanking the D29A mutation. After selection of bedaquiline-resistant M. abscessus colonies transformed with both substrates, we obtained equivalent numbers of recombinants. These resistant colonies were analyzed by allele-specific PCR and Sanger sequencing, and we demonstrated that the presence of the genetic barcode was linked to the targeted incorporation of the desired mutation in its chromosomal location. All recombinants displayed the same MIC to bedaquiline as the original isolate, from which the D29A mutation was identified. Finally, to demonstrate the broad applicability of this method, we confirmed the association of bedaquiline resistance with the atpE A64P mutation in analysis performed in independent M. abscessus strains and by independent researchers. IMPORTANCE Antimicrobial resistance (AMR) threatens the effective prevention and treatment of an ever-increasing range of infections caused by microorganisms. On the other hand, infections caused by Mycobacterium abscessus affect people with chronic lung diseases, and their incidence has grown alarmingly in recent years. Further, these bacteria are known to easily develop AMR to the few therapeutic options available, making their treatment long-lasting and challenging. The recent introduction of new antibiotics against M. abscessus, such as bedaquiline, makes us anticipate a future when a plethora of antibiotic-resistant strains will be isolated and sequenced. However, in the era of whole-genome sequencing, one of the challenges is to unequivocally assign a biological function to each identified polymorphism. Thus, in this study, we developed a fast, robust, and reliable method to assign genotype-phenotype associations for putative antibiotic-resistant polymorphisms in M. abscessus.

Keywords: Bdq; Bq; barcoding; bedaquiline; chronic obstructive disease; cystic fibrosis; drug resistance; nontuberculous mycobacteria; recombineering.

FIG 1

Workflow scheme followed in this study. (A and B) Design of ssAES carrying silent barcoding (BC) mutations and the antibiotic resistance mutation (A) allowed the selection of resistant colonies after electroporation of M. abscessus carrying the recombineering machinery (B). (C to E) Sanger sequencing (C) or barcoding PCR methods (D and E) were used to confirm genotypes of recombinant colonies recovered. (F) Confirmation of the phenotype-genotype association of the mutation under study was assessed by calculating the MIC to the antibiotic. The panel marked with an asterisk indicates the experimental steps necessary to construct an M. abscessus recombineering strain, which added an extra 8 to 12 days to the times indicated in the figure. It is also important to consider that plasmid transformation into nonlaboratory strains can be challenging.

FIG 2

Chromosomal barcoding method to establish a genotype-phenotype association for the D29A mutation in the atpE gene. (A) Resazurin microtiter assay in 7H10-ADC to determine MICs of bedaquiline against different M. abscessus SL541 strains. Wells with viable bacteria are shown in pink, whereas wells with absence of growth are shown in blue, after addition of resazurin. (B) Graphical representation of the ssAES used to mutate the 29th codon of the atpE gene and the alignment in M. abscessus SL541 and reference ATTC19977 strain genomes. (C) Electroporated bacteria with ssAES atpE wild type (WT), ssAES atpE D29A, or and ssAES BC atpE D29A were plated in the presence (dilution 10⁻¹) or absence (dilution 10⁻⁶) of bedaquiline at 3 μg/mL. (D) MTT in 7H9Tyl-ADC was used to determine the MICs for different isolated recombinant colonies. Wells with viable bacteria are shown in dark purple, whereas wells with absence of growth are shown in yellow, after addition of MTT. (E) Sanger sequencing chromatograms of recombinant colonies of M. abscessus electroporated with ssAES atpE D29A (top) and ssAES BC atpE D29A (bottom).

FIG 3

Development of PCR-based methods to identify chromosomal barcodes. (A and B) Graphical representation of the nonamplification (A) and amplification (B) by barcode PCR of bedaquiline-resistant M. abscessus SL541 recovered after electroporation with ssAES atpE D29A and ssAES BC atpE D29A, respectively. (C) Agarose gels showing barcoding PCR amplifications of bedaquiline-resistant colonies transformed with either a nonbarcoded (gray characters) or a barcoded AES (red characters). Note the presence of the 531-bp specific band in the transformants with the barcoded AES. Amplification of a 609-bp band corresponding to the 16S housekeeping gene is also shown as a positive control for amplification. (D and E) ΔRn values of amplification of the 16S rRNA gene (dashed black) or the specific D29A barcode (continuous red) in ssAES atpE D29A (D) or ssAES BC atpE D29A (E) M. abscessus transformants obtained by barcoded real-time PCR.

FIG 4

Chromosomal barcoding method to establish a genotype-phenotype association for the A64P mutation in the atpE gene. (A) Graphical representation of the ssAES used to mutate the 64th codon of the atpE gene and their alignment in M. abscessus SL541 and reference ATTC 19977 strain genomes. (B) Graphical representation of the amplification by barcoding PCR of bedaquiline-resistant M. abscessus SL541 isolate recovered after electroporation with ssAES BC atpE A64P. (C) Agarose gels of barcoded PCR of bedaquiline-resistant colonies. The presence of a 432-bp band was indicative of the proper exchange of the AES shown in panel A into its specific chromosomal location. (D and E) ΔRn values of amplification of the 16S rRNA gene (dashed black) or the specific A64P barcoded (continuous green) in a WT colony (D) or ssAES BC atpE A64P M. abscessus SL541 transformants (E) obtained by barcode real-time PCR. (F) MTT assay results in 7H9Tyl-ADC to determine the MICs of different isolated recombinant colonies. Wells with viable bacteria are shown in dark purple, whereas wells with absence of growth are shown in yellow, after addition of MTT.