Prior to the COVID-19 pandemic, tuberculosis (TB) was the world leading cause of human death due to a single infectious agent. In México, TB is a serious public health issue with contrasting prevalence among the different geographic regions (Sánchez-Barriga, 2015). Chiapas, a Southern state located in the Mexico-Guatemala border, is one of the poorer states and has one of the highest TB morbidity and mortality rates in the country (Sánchez-Barriga, 2015). Factors that contribute to this situation include poverty, low educational levels, inadequate health service coverage, failure to comply the strict directly observed treatment (DOT) for TB, and a growing epidemic of resistance to first-line drugs (Sánchez-Pérez et al., 2001; Nájera-Ortiz et al., 2008; Sánchez-Pérez et al., 2010; Nájera-Ortiz et al., 2012).

Overall, the Mexican population comprised divergent ancestry such as a mixture of the original descendants of American indigenous populations, the Amerindians, and Europeans, mostly of Spanish heritage (Moreno-Estrada et al., 2014), but the regional ethnic composition may vary greatly. For example, in Nuevo León, a state located in the North of México, pure Amerindians account for only 1% of the population, while in Chiapas, the Amerindian population may constitute up to 28%, mainly from Mayans descendants (Moreno-Estrada et al., 2014).

Little is known about the genetic diversity of the M. tuberculosis complex (MTBC) isolates circulating in Mexican-Amerindian settlements. The presence of MTBC species in America prior to the arrival of Europeans (pre-Columbian) has been a source of controversy. It has been proposed that the presence of M. pinnipedii in some South American mummies could be explained by the migration of sea lions from the Cape of Good Hope in Africa, first to South America and later to Central and North America (Bos et al., 2014). If such a hypothetical scenario is correct, the MTBC isolates found in America should have different spoligotypes than those typically found in European descendants. In order to identify the predominant genetic patterns of MTBC isolates, we conducted a study in an Amerindian rural population settled in the mountains of the Highlands region of Chiapas using spoligotyping and whole-genome sequencing (WGS) analyses.

Chiapas is a Southern Mexican state with approximately 7.3% of Mexico’s indigenous population (INEGI, 2015). The state was a part of a large region dominated by the Mayans during the Classic period (A.D. 250 to 900), and people currently living in this region are predominantly their descendants. The most numerous of these Mayan groups include the Tzeltal, Tzotzil, Ch’ol, Zoque, Tojolabal, Lacandon, and Mam (INEGI, 2004). Most communities are found in the municipalities of the Centro, Altos, Norte, and Selva regions, with many considered to be 100% indigenous (INEGI, 2015).

To date, several studies have described M. tuberculosis genetic diversity around Mexico, but not in Amerindian populations. Previous studies have shown SIT53 (T1) to be the most predominant subclade found in Monterrey (24%), Guadalajara (25%), and San Luis Potosí (28.6%) (Molina-Torres et al., 2010; López-Rocha et al., 2013; Flores-Treviño et al., 2015). In our results in Chiapas, SIT53 represented 19.3% of the total isolates examined.

Interestingly, SIT53 isolates from Chiapas presented seven IS6110 bands; in our previous study in Monterrey, SIT53 strains showed four bands or less (Molina-Torres et al., 2010). In Mexico City (Zenteno-Cuevas et al., 2015), isolates corresponding to SIT53 were multi-banded strains but the band pattern was not found to be similar to the Chiapas profile. Clusters of strains with the same spoligotype can present different IS6110 profiles because it has been estimated that the rate of change in the DR region is slower than that of IS6110 (Niemann et al., 1999). IS6110 is a genetic clock that shows recent changes in the genome (Niemann et al., 1999; Warren et al., 2002). In our study, the identical IS6110 pattern between SIT53 isolates indicates a recent transmission (Cave et al., 2005). Different and assorted IS6110 patterns among SIT53 isolates from the entire country indicate a common ancestor probably long time ago when lineage 4 was introduced by Europeans into the country, about 1800 A.D. (Brynildsrud et al., 2018). A wider WGS analysis of SIT53 isolates from Mexico will help to trace the evolution of these strains over the 500 years since the arrival of these M. tuberculosis isolates, and how they have been adapted to the Mexican population.

Lineage 4 M. tuberculosis strains are originally from Europe; from there, they migrated to Vietnam in the 13th century (Brynildsrud et al., 2018), to the Republic of Congo, Africa, in the 15th century, and then to Uganda and Malawi in the late 17th century. The three earliest migration events to America were inferred to have occurred between 1466 and 1593 to South America and between 1566 and 1658 to North America (Brynildsrud et al., 2018). Most of our SIT 53 isolates belonged to lineage 4.8, which is not commonly found in America, where 4.3.3 dominates (Hadi et al., 2021). This might be explained by groups of people from Spain, where Lineage 4 is predominant (Madrazo Moya et al., 2021), or Portugal who migrated to Mexico or South America carrying different sub-lineages, or to the adaptation through the years to the local hosts, as has been observed in in vivo experiments (Reiling et al., 2013). When comparing our 4.8 isolates with other isolates of the world (Stucki et al., 2016), we observed a closer relationship with strains from Brazil, strengthening this possibility. Interestingly, we found a lineage 4.9 isolate among the SIT53 isolates, which has been rarely reported in the world (Napier et al., 2020).

SIT42 seems to dominate central and south America. In neighboring Guatemala, the most abundant spoligotypes are similar to those found in Mexico: SIT33 (23%), SIT53 (14%), and SIT42 (7%) (Castellanos et al., 2020). In Panama, SIT42 produces 68% of the MDR cases (Realpe et al., 2014; Castellanos et al., 2020) and is also the spoligotype dominant in Colombia (30%) (Rosas et al., 2013). Alternatively, SIT53 in Colombia represents less than 3.4% of the isolates (Rosas et al., 2013).

In Mexico, SIT42 isolates have been found in Monterrey and San Luis Potosi, in the north, at 1.11% and 5.9%, respectively (INEGI, 2015; Flores-Treviño et al., 2015); in Acapulco, in south central Mexico, it produces 5.6% of the cases (Nava-Aguilera et al., 2011); and in Guadalajara, in the west (López-Rocha et al., 2013), 7.4%, compared to 13.7% in Chiapas. This amount is more similar to that found in central and south America.

We found in Chiapas 4.3 lineages to be abundant, particularly sub-lineages 4.3.3 and 4.3.4.2. The Latino-American family (Lineage 4.3) has been observed to predominate in South America (Brynildsrud et al., 2018). Interestingly in a recent study using samples from the Guarani-Kaiowaí indigenous people living in Paraguay and the Brazilian state of Mato Grosso do Sul, Hadi et al. (2021) found that 81% of the isolates belong to sub-lineage 4.3.3, followed by 4.3.4.2 (13%), in a similar way to our findings (Hadi et al., 2021). In both cases, Amerindian people were the hosts, which makes possible the adaptation of these sub-lineages to people with this genetic background, as has been seen in Tanzania where L3.1.1 is highly successful in Dar es Salam (Zweyer et al., 2021). The comparison of our 4.3.3 genomes with those from Hadi et al.’s (2021) study resulted in the formation of a separated cluster with two Amazonian isolates in the group, which strengthen the possible relationship with South American isolates. These Chiapas’ 4.3.3 isolates, even if they were not identical, showed a similar pattern in mutations in genes associated with antimicrobial resistance. When aligned with other 294 lineage 4 isolates, these isolates clustered together with an isolate from Sierra Leona, supporting the African origin of these isolates brought by people enslaved by Portuguese to Brazil. Three more isolates were located close to MTB isolates from Portugal, strengthening this possibility.

The pre-Columbian presence of M. tuberculosis has been suggested by the finding of IS6110-positive samples from Peruvian mummies (Klaus et al., 2010). More recently, M. pinnipedii DNA has been isolated in South America from human remains, supporting the presence of MTBC infections before the contact with Europeans (Bos et al., 2014). If pre-Columbian M. tuberculosis would exist, then it should be demonstrated in Amerindian settlements which have little contact with the rest of the Mexican population. However, we did not find a new lineage or other members of the MTBC complex. Our results support the theory of the origin of TB in Mexico as through contact with Europeans and that 200 years was sufficient time for the organism to spread across the whole country and adapt to the new naïve population.

According to our results, Chiapas 4.3.3 isolates seem to come from Africa via Brazil. It seems like from this country this TB lineage spread to the rest of Latin America where it is abundant, including south of Mexico. Interestingly, SIT42 is not predominant in north of Mexico, perhaps due to the recent migration of 4.3.3 to Chiapas or due to differences in host susceptibility to MTB lineages. It is necessary to study the lineages prevailing in the rest of Mexico in order to determine the dissemination and adaptation of the different MTB families.

The dataset presented in this study can be found in online repositories. The Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the following accession numbers: M. tuberculosis LIID-CH2 (JAIFJQ000000000), M. tuberculosis LIID-CH3 (JAIFJP000000000), M. tuberculosis LIID-CH6 (JAIFJO000000000), M. tuberculosis LIID-CH14 (JAIFJN000000000), M. tuberculosis LIID-CH15 (JAIFJM000000000), M. tuberculosis LIID-CH17 (JAIFJL000000000), M. tuberculosis LIID-CH18 (JAIFJK000000000), M. tuberculosis LIID-CH19 (JAIFJJ000000000), M. tuberculosis LIID-CH21 (JAIFJI000000000), M. tuberculosis LIID-CH24 (JAIFJH000000000), M. tuberculosis LIID-CH28 (JAIFJG000000000), M. tuberculosis LIID-CH30 (JAIFJF000000000), M. tuberculosis LIID-CH31 (JAIFJE000000000), M. tuberculosis LIID-CH33 (JAIFJD000000000), M. tuberculosis LIID-CH40 (JAIFJC000000000), M. tuberculosis LIID-CH42 (JAIFJB000000000), M. tuberculosis LIID-CH53 (JAIFJA000000000), M. tuberculosis LIID-CH60 (JAIFJZ000000000), M. tuberculosis LIID-CH68 (JAIFJY000000000), M. tuberculosis LIID-CH70 (JAIFJX000000000), M. tuberculosis LIID-CH76 (JAIFJW000000000), M. tuberculosis LIID-CH93 (JAIFJV000000000), M. tuberculosis LIID-CH109 (JAIFJU000000000), and M. tuberculosis LIID-CH117 (JAIFJT000000000). All these sequences belong to BioProject PRJNA751891. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

Prior to the start of the study, ethical approval was obtained from the Ecosur Research Ethics Committee (CEI-O-068/14) and from the Ministry of Health for Chiapas (5003/00063) under the project “Mycobacterium tuberculosis spoligotypes from clinical isolates from Chiapas, Mexico.” The microbiological records and basic demographic information for patients were anonymized and de-identified prior to analysis. Informed consent for participation was obtained from patients in a written form, and sputum samples were used only for the purpose of this study.

LV-C, FQ, and CM-T contributed to conception and design of the study. JC-G and JO-C organized the database. AG-V, AB-A, HS-P, SM-J, and AR contributed to the collection, culture and identification of clinical samples. LV-C, AA, MS, and PS performed the bioinformative analysis. LV-C wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

This work was supported by PAICYT grant SA152-15 from the Universidad Autónoma de Nuevo León.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The handling editor KKM declared a shared affiliation with the author(s) AA, MS, and PS at the time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.