A collection of 814 M. tuberculosis isolates from years 2012 to 2016 stored at “Centro Nacional de Referencia para Micobacterias” from “Instituto Nacional de Salud Pública e Investigación Leopoldo Izquieta Pérez” in Guayaquil (Ecuador) were included in the study (Figure 1). MTBC isolates are routinely processed at INSPI laboratories, where culture and antibiotic resistance profiling for first-and second-line drugs used in M. tuberculosis therapy is performed for MTBC cultures following Pan American Health Organization guidelines (31–33). The samples were previously inactivated and stored for research purposes following the guidelines from this government center. The access to this MTB strains collection was approved by IRBs from “Instituto Nacional de Salud Pública e Investigación Leopoldo Izquieta Pérez” and University San Francisco de Quito (code 2017-023IN), both certified by Ministry of Public Health from Ecuador following guidelines from Declaration of Helsinki. All samples were anonymized, and no data of the patients were made available.

A small sample from cultures of MTBC isolates was collected and resuspended in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0), then inactivated by boiling at 95°C for 45 min. After this process, samples were centrifuged for 5 min at 10,000 g and the supernatant was directly used for genotyping, as reported elsewhere (34, 35).

A multiplex PCR developed for the detection of the RD^Rio pattern was performed using a two set of primers (Supplementary Figure S1): (1) a set of primers that target the IS1561’ fragment, located inside the region between the genes Rv3346c and Rv3355c, and used as a marker for wild type M. tuberculosis (530 bp); (2) another set of primers is used to flank the RD^Rio locus and bridge the deletion in RD^Rio strains (1175 bp) (36). The PCR reaction was prepared as follows: 12.5 μL of GoTaq^® Green Mastermix (Promega, Wisconsin, United States) (1X), 0.5 μL of each primer (0.2 μM), 5 μL of mycobacterial DNA and completed with Nuclease-Free water to a final volume of 25 μL. The multiplex PCR program was established as follows: initial denaturation at 94°C for 3 min, 35 cycles at 94°C for 30 s, 60°C for 30 s, 72°C for 1 min 30 s, and a final extension step of 72°C for 10 min. PCR products were run in 1.5%UltraPure^™ Agarose (Invitrogen, California, United States) gels of 15 cm × 10 cm in 0.5X Tris-boric acid-EDTA (TBE) buffer at 100 V for 2 h using a ladder 100 bp Plus Opti-DNA Marker (Cat. No.: G016, Applied Biological Materials Inc., British Columbia, Canada) for size determination (22, 36).

The method is PCR-based and allows the detection of different Mycobacterial Interspersed Repetitive Units (MIRU) located at multiple loci in the MTBC genome. Each MIRU allele is identified by a number; thus generating a numerical profile which is used for genotyping studies (35, 37). The PCR reaction was prepared individually to a final volume of 15 μL: 7.5 μL of GoTaq^® Green Mastermix (Promega, Wisconsin, United States) (1X), 0.75 μL of MIRU primers (0.5 μM) (38), 1 μL of mycobacterial DNA and completed with Nuclease-Free water. The following PCR program was established for all MIRUs: initial denaturation at 94°C for 5 min, 35 cycles at 94°C for 1 min, 54°C for 1 min 30 s, 72°C for 1 min 30 s, and a final extension step of 72°C for 10 min. Amplicons were run in 2% UltraPure^™ Agarose (Invitrogen, California, United States) gels of 15 cm × 10 cm in 0.5X Tris-boric acid-EDTA (TBE) buffer at 100 V for 3 h using a ladder 100 bp Plus Opti-DNA Marker (Cat. No.: G016, Applied Biological Materials Inc., British Columbia, Canada) for size determination. MIRU allele identification was performed according to Supply et al. (38).

Results obtained were analyzed using the MIRU-VNTRplus web application¹ (39). Lineage designation was performed by similarity search using 24-loci MIRU-VNTR data (Supplementary Data 1) and calculation of Minimum Spanning Tree (MST) and Neighbour-joining Tree (NJT) was performed to determine the population structure of Ecuadorian M. tuberculosis isolates.

The method is based on the presence/absence of 43 DNA spacer sequences, which are interspersed between 36 conserved loci located in the Direct Repeat (DR) region (37, 40). Spoligotyping was performed according to Kamerbeek et al. (41): PCR amplification of DR loci was performed using DRa and DRb primers and products were biotinylated and hybridized to a membrane containing oligonucleotides for each spacer sequence. After hybridization, the membrane is washed and then incubated in diluted streptavidin-peroxidase conjugate (Roche, United States). Membrane was exposed to chemiluminescent Amersham ECL reagents (GE Healthcare, United Kingdom) and located in an X-Ray cassette on a Hyperfilm ECL (Merck, United States). After the reaction, the film was inserted into a film developer solution in a dark room after which it is moved to the fixer solution. Thereafter, the film is dried and ready for the interpretation of the result. After developing the film a positive/negative signal is recorded in binary or octal formats for genotyping interpretations (Supplementary Data 1). The results are compared against the Fourth International Spoligotyping Database (SpolDB4) (42).

A total of 814 M. tuberculosis isolates were analyzed by multiplex PCR for MTB RD^Rio identification (Figure 1). Of those, 17 strains generated a band pattern in the electrophoresis gel compatible with MTB RD^Rio (Figure 1; Table 1; Supplementary Figure S1), representing a 2.09% of the total MTBC population analyzed. Those MTB RD^Rio strains were assigned to LAM MTB subfamily by MIRU-VNTR (Table 1; Figure 2; Supplementary Data 1). Those 17 isolates were subsequently analyzed by spoligotyping, with 6 strains with undetermined lineage and 11 of them associated to LAM sublineage: five belong to LAM9, one to LAM5-LAM6, two belong to LAM3, one to LAM2 and two were identified as LAM1, according to the SpolDB4 (Table 1; Figure 2; Supplementary Data 1).

Regarding the geographic location of the 17 MTB RD^Rio strains, all of them were isolated from provinces in the Coastal Region of Ecuador: Guayas, Los Rios, Santo Domingo and Santa Elena (Figure 1).

Drug susceptibility testing of MTB RD^Rio isolates was performed for first line antibiotics: 8/17 were resistant to isoniazid (47.1%), 1/17 was resistant to streptomycin (5.9%), 6/17 were resistant to rifampicin (35.3%) and no samples showed resistance to ethambutol and pyrazinamide; 9/17 isolates were sensitive to all first-line antibiotics (52.9%). Finally, 5/11 (29.4%) isolates were resistant to isoniazid and rifampicin, that means multi drug resistance (MDR) strains (Table 2). For the whole collection of 814 MTB strains included in the study, the MDR strains represented 10.9% (Table 3).

The population structure of the MTB RD^Rio isolates was determined through the analysis of 24-loci MIRU-VNTR and spoligotyping using the MIRU-VNTRplus platform (Supplementary Data 1), as it is detailed in the Neighbour-Joining Tree (NJT) presented in Figure 2. Although several spoligotype patters were repeated for 2 or 3 MTB RD^Rio strains, none of the strains shared the same MIRU-VNTR pattern. Based on total identity in 24 loci for MIRU-VNTR, no actively transmitted MTB RD^Rio clones were found (Figure 2).

Additionally, a phylogenetic analysis was done using 12-loci MIRU-VNTR, including all the MTB RD^Rio strains reported in the bibliography (18, 43), as well as the strains identified in the current study for Ecuador. This analysis is detailed in the Minimum Spanning Tree (MST) in the Figure 3, where the maximum locus difference within a clonal complex was set at 2. MST included 17 MTB RD^Rio isolates from Ecuador, 8 from Argentina, 31 from Paraguay, 51 from Venezuela and 15 from Brazil. Four clonal complexes of MTB RD^Rio isolates are well defined in the MST: two clonal complexes comprising isolates exclusively from Ecuador, one clonal complex comprising isolates exclusively from Paraguay, and one large clonal complex including MTB RD^Rio isolates from Ecuador, Argentina, Brazil, Paraguay and Venezuela. Interestingly, 12 of 17 MTB RD^Rio strains from Ecuador are well differentiated from strains from the other South American countries.