Tracking a tuberculosis outbreak over 21 years: strain-specific single-nucleotide polymorphism typing combined with targeted whole-genome sequencing

Abstract

Background: Whole-genome sequencing (WGS) is increasingly used in molecular-epidemiological investigations of bacterial pathogens, despite cost- and time-intensive analyses. We combined strain-specific single-nucleotide polymorphism (SNP) typing and targeted WGS to investigate a tuberculosis cluster spanning 21 years in Bern, Switzerland.

Methods: On the basis of genome sequences of 3 historical outbreak Mycobacterium tuberculosis isolates, we developed a strain-specific SNP-typing assay to identify further cases. We screened 1642 patient isolates and performed WGS on all identified cluster isolates. We extracted SNPs to construct genomic networks. Clinical and social data were retrospectively collected.

Results: We identified 68 patients associated with the outbreak strain. Most received a tuberculosis diagnosis in 1991-1995, but cases were observed until 2011. Two thirds were homeless and/or substance abusers. Targeted WGS revealed 133 variable SNP positions among outbreak isolates. Genomic network analyses suggested a single origin of the outbreak, with subsequent division into 3 subclusters. Isolates from patients with confirmed epidemiological links differed by 0-11 SNPs.

Conclusions: Strain-specific SNP genotyping allowed rapid and inexpensive identification of M. tuberculosis outbreak isolates in a population-based strain collection. Subsequent targeted WGS provided detailed insights into transmission dynamics. This combined approach could be applied to track bacterial pathogens in real time and at high resolution.

Keywords: genomic epidemiology; outbreak; screening; tuberculosis; whole genome sequencing.

Figure 1.

Overview of patient isolates and whole genome sequences generated. A total of 1642 isolates collected between 1991 and 2011 were available for single-nucleotide polymorphism (SNP) genotyping. Three isolates showed ambiguous SNP-typing results and were excluded. One additional patient isolate (P028; isolated in 1987) reported in the original publication [17] and predating the systematic collection of isolates in 1991 was included in the study. For the key patient, a second isolate (P006B; isolated in 1991 [17]) was available and was included in the genomic analyses.

Figure 2.

Initial neighbor joining phylogeny of Mycobacterium tuberculosis isolates. Three whole-genome sequences from the historic outbreak and 4 control isolates were used to identify single-nucleotide polymorphisms (SNPs) specific to the outbreak genotype. Node support was assessed by bootstrapping over 1000 pseudo-replicates and is indicated as a percentage.

Figure 3.

Epidemic curve of the 68 patients identified as tuberculosis cluster cases. Gray boxes indicate patients associated with the social milieu (homeless individuals and/or substance abusers). One additional patient isolate (P028; isolated in 1987) reported in the original publication [17] and predating the systematic collection of isolates in 1991 was included in the study. For 1 patient from 1991 [17], a second isolate (P006A; isolated in 1988) was available and was therefore backdated.

Figure 4.

Distribution of cluster patients with tuberculosis in the milieu of substance abusers and homeless people. The 4 main hotspots of transmission that were identified by social contact tracing are shown (a short-term homeless shelter, a long-term social integration home, a meeting point for substance abusers, and a bar). Milieu patients are associated with a particular social context (homeless, substance abuser scene). Solid lines indicate confirmed epidemiological links, and dashed lines indicate suspected social links. Presumptive individual links between milieu patients are not shown because these patients are highly interlinked.

Figure 5.

Median Joining network using 133 variable single nucleotide positions (SNP) among whole genome sequences of Mycobacterium tuberculosis cluster isolates of the “Bernese cluster” outbreak. Branch lengths correspond to number of SNPs. Circle sizes correspond to number of isolates, median vectors (mv) are hypothetical genotypes. Position of “mv6” is the root of the network. A, Network showing the four identified “sub-clusters. Darker circles in each subcluster indicate patients that were associated with the particular social milieu (homeless, substance abuser scene), lighter circles are patients in the non-milieu population. Underlined labels represent isolates that were identified in the original publication [17]. B, Network colored according to time period when the M. tuberculosis strains were isolated. C, Network showing patients associated with a particular hotspot.

Figure 5.

Median Joining network using 133 variable single nucleotide positions (SNP) among whole genome sequences of Mycobacterium tuberculosis cluster isolates of the “Bernese cluster” outbreak. Branch lengths correspond to number of SNPs. Circle sizes correspond to number of isolates, median vectors (mv) are hypothetical genotypes. Position of “mv6” is the root of the network. A, Network showing the four identified “sub-clusters. Darker circles in each subcluster indicate patients that were associated with the particular social milieu (homeless, substance abuser scene), lighter circles are patients in the non-milieu population. Underlined labels represent isolates that were identified in the original publication [17]. B, Network colored according to time period when the M. tuberculosis strains were isolated. C, Network showing patients associated with a particular hotspot.