A robust SNP barcode for typing Mycobacterium tuberculosis complex strains.

Coll F, McNerney R, Guerra-Assunção JA, Glynn JR, Perdigão J, Viveiros M, Portugal I, Pain A, Martin N, Clark TG. · 2014

Read A robust SNP barcode for typing Mycobacterium tuberculosis complex strains. by Coll F, McNerney R, Guerra-Assunção JA, Glynn JR, Perdigão J, Viveiros M, Portugal I, Pain A, Martin N, Clark TG. (2014) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4166679.txt.

Introduction

Francesc Coll, Taane G. Clark, and eight co-authors set out in Nature Communications in 2014 to build a stable typing code for the bacteria that cause tuberculosis. Strain type can shape how sick a patient gets, how fast the germ spreads, and whether drugs still work. Older typing schemes sorted strains in ways that did not always agree.

Scanning 1,601 genomes from around the world, the group found about 92,000 single-letter DNA changes. From the strain-specific ones they picked 62 markers that form a barcode. The barcode sorts isolates into all the main lineages and more sub-lineages than older tools, and it matches the standard classification well. Health labs can use it to type clinical samples fast.

Catalog record

CREATORColl F, McNerney R, Guerra-Assunção JA, Glynn JR, Perdigão J, Viveiros M, Portugal I, Pain A, Martin N, Clark TG.
YEAR2014
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4166679

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A robust SNP barcode for typing Mycobacterium tuberculosis complex strains By Francesc Coll, Ruth McNerney, José Afonso Guerra-Assunção, Judith R Glynn, João Perdigão, Miguel Viveiros, Isabel Portugal, Arnab Pain, Nigel Martin, Taane G Clark Published in: Nature Communications DOI: 10.1038/ncomms5812 ABSTRACT Abstract Strain-specific genomic diversity in the Mycobacterium tuberculosis complex (MTBC) is an important factor in pathogenesis that may affect virulence, transmissibility, host response and emergence of drug resistance. Several systems have been proposed to classify MTBC strains into distinct lineages and families. Here, we investigate single-nucleotide polymorphisms (SNPs) as robust (stable) markers of genetic variation for phylogenetic analysis. We identify ~92k SNP across a global collection of 1,601 genomes. The SNP-based phylogeny is consistent with the gold-standard regions of difference (RD) classification system. Of the ~7k strain-specific SNPs identified, 62 markers are proposed to discriminate known circulating strains. This SNP-based barcode is the first to cover all main lineages, and classifies a greater number of sublineages than current alternatives. It may be used to classify clinical isolates to evaluate tools to control the disease, including therapeutics and vaccines whose effectiveness may vary by strain type. Infection with bacteria of the Mycobacterium tuberculosis complex (MTBC) results in a variety of outcomes including latent infection and/or progression to pulmonary or extra-pulmonary manifestations of disease. Such diversity has been historically attributed to host and environmental factors, and the MTBC was previously considered genetically monomorphic in nature1. However, the development of typing methods that discriminate strains into distinct lineages has demonstrated previously unrecognized diversity. The first attempts to differentiate strains were based on phage typing2. Geographic patterns of strain distribution were identified and in the south of India, an association with virulence was noted3. Phage typing has since been superseded by genetic methods. Initial efforts were focussed on large sequence polymorphisms4. Six major global MTBC lineages have been defined (1 Indo-Oceanic, 2, East-Asian including Beijing, 3 East-African-Indian, 4 Euro-American, 5 West Africa or Mycobacterium africanum I, 6 West Africa or M. africanum II), distinct from a M. bovis clade. Lineages 1, 5 and 6 are considered ‘ancient’, and 2 to 4 ‘modern’. A novel phylogenetic lineage of MTBC that appears to be intermediate between the ancient and modern has been described in Ethiopia and the Horn of Africa5,6, referred to as lineage 7. Additional genotyping methods have been developed based on detection of insertion elements, variable number of repeats or the presence or absence of spacer oligonucleotides (spoligotyping)7. Although mainly used for public health purposes or epidemiological studies, spoligotype families have been used to provide further resolution within the lineages8. Lineage is of importance to tuberculosis control as it has been shown that strain type may play a role in disease outcome, variation in vaccine efficacy9 and emergence of drug resistance10. Different strains of MTBC have produced distinct biological responses in experimental models and can affect clinical presentation11,12,13. Strain type may also influence disease epidemiology as in some settings it is associated with the presence or absence of clustering due to recent transmission14. Lineage-specific differences in the virulence of clinical isolates have been reported across independent experimental systems with modern lineages, such as Beijing and Euro-American Haarlem strains believed to exhibit more virulent phenotypes compared with ancient lineages, such as East-African-Indian and M. africanum strains15. The molecular mechanisms and genetic factors responsible for the described differences remain largely unknown. Their investigation requires transparent, easily applied, reliable methods for determining stain type. Several sets of single-nucleotide polymorphisms (SNPs) have previously been proposed16,17,18,19,20 but limited numbers and variation of strains were used in their construction and they describe a restricted number of lineages and sublineages. In this study, we have sought to provide an improved system by examining whole-genome data from a large number of strains (n=1,601) from a diverse geographic spread. We identify a single panel of 62 SNPs that may be used to resolve all seven lineages and a total of 55 sublineages. RESULTS Population structure A genomic analysis was performed on whole-genome sequences of 1,601 MTBC isolates from eleven independent sequencing studies from different areas of the world, with representation of all seven major lineages (1, n=121, 7.6%; 2, n=390, 24.3%; 3, n=189, 11.8%; 4, n=856, 53.5%; 5, n=17, 1.1%; 6, n=11; 7, n=6), as well as M. bovis (n=11; Supplementary Table 1). A total of 91,648 SNPs were identified; 54.6% were observed in a single sample, that is, not found in other samples (Supplementary Fig. 1), 89.2% were in coding regions, and 63.5% resulted in non-synonymous changes in amino acids. The SNP-based phylogenetic tree demonstrated a clustering largely congruent with published MTBC phylogenies (Fig. 1). MTBC main lineages (1–7 and M. bovis) and sublineages were subsequently identified based on the spoligotype and regions of difference (RD) composition of the clades in the SNP-based phylogeny (see Methods). The phylogeny revealed the presence of new clades for which RDs do not discriminate. In particular, there were gaps in the Euro-American lineage for which molecular fingerprint classifications are less accurate16. [Figure: Global phylogeny of 1,601 MTBC isolates.A total of 91,648 SNPs spanning the whole genome were used to reconstruct the phylogeny of 1,601 MTBC isolates. All seven main MTBC lineages are indicated at the inner area of the tree. The main sublineages are annotated at the outer arc along with lineage-specific RDs. Identified clades are colour coded.] Although estimates of genetic diversity may be influenced by sampling bias, the greatest nucleotide diversity was observed in lineages 1 (π=0.0103) and 6 (π=0.0093), and the least within lineage 2 (π=0.0039) and 3 (π=0.0040) strains (Supplementary Table 2). Although spoligotypes tended to cluster within specific clades, there was some evidence of homoplasy, particularly in lineage 4. These anomalies arise from convergent evolution of CRISPR-based spoligotyping polymorphisms16. All previously reported lineage-specific RDs4 were detected, and their distribution was consistent with clades in the SNP-based phylogeny. There was no evidence of homoplasy events using large sequence polymorphisms, further demonstrating their robustness as phylogenetic markers. As expected, isolates from lineage 1 harboured the RD239 deletion and had EAI-like spoligotypes. Two natural sublineages designated 1.1 and 1.2 contained distinctive spoligotype compositions (Supplementary Data 1). The Beijing-specific RD105 deletion was restricted to lineage 2, whilst others (RD207, RD181, RD150 and RD142) were observed downstream from the common ancestor (Fig. 1), defining sublineages within lineage 2.2. Isolates belonging to lineage 3 harboured the CAS-specific RD750 deletion, and included sublineages for CAS1-Delhi (33.9%), CAS1-Kili (53.4%), CAS (6.3%) and CAS2 (5.3%) spoligotypes. All non-CAS1-Delhi samples were grouped into the same clade (sublineage 3.1), which was subdivided into two clades harbouring CAS1-Kili and CAS/CAS2 samples, respectively. The Euro-American lineage has been the most poorly characterized historically21, and as expected using spoligotypes there was evidence of non-homogeneous sublineages (in particular T, H and LAM families), potentially due to homoplasy events21. The phylogeny revealed 36 distinctive clades for lineage 4. All 10 Euro-American lineage RDs were consistently located within one of these clades, further subdivision was achieved and clades with unreported RD were identified (for example, sublineages 4.2, 4.4, 4.7 and 4.9; Fig. 1). Representatives of Haarlem (sublineage 4.1.2.1), Cameroon (4.6.2), LAM (4.3), S-type (4.4.1.1), TUR (4.2.2.1), Uganda (4.6.1), Ural (4.2.1) and X-type (4.1.1) strains were all identified (Supplementary Data 1). Consistent with previous phylogenetic studies, strains belonging

Read the complete text