Evolutionary history and introduction of SARS-CoV-2 Alpha VOC/B.1.1.7 in Pakistan through international travelers

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants continue to emerge, and their identification is important for the public health response to coronavirus disease 2019 (COVID-19). Genomic sequencing provides robust information but may not always be accessible, and therefore, mutation-based polymerase chain reaction (PCR) approaches can be used for rapid identification of known variants. International travelers arriving in Karachi between December 2020 and February 2021 were tested for SARS-CoV-2 by PCR. A subset of positive samples was tested for S-gene target failure (SGTF) on TaqPathTM COVID-19 (Thermo Fisher Scientific) and for mutations using the GSD NovaType SARS-CoV-2 (Eurofins Technologies) assays. Sequencing was conducted on the MinION platform (Oxford Nanopore Technologies). Bayesian phylogeographic inference was performed integrating the patients' travel history information. Of the thirty-five COVID-19 cases screened, thirteen had isolates with SGTF. The travelers transmitted infection to sixty-eight contact cases. The B.1.1.7 lineage was confirmed through sequencing and PCR. The phylogenetic analysis of sequence data available for six cases included four B.1.1.7 strains and one B.1.36 and B.1.1.212 lineage isolate. Phylogeographic modeling estimated at least three independent B.1.1.7 introductions into Karachi, Pakistan, originating from the UK. B.1.1.212 and B.1.36 were inferred to be introduced either from the UK or the travelers' layover location. We report the introduction of SARS-CoV-2 B.1.1.7 and other lineages in Pakistan by international travelers arriving via different flight routes. This highlights SARS-CoV-2 transmission through travel, importance of testing, and quarantine post-travel to prevent transmission of new strains, as well as recording detailed patients' metadata. Such results help inform policies on restricting travel from destinations where new highly transmissible variants have emerged.

Keywords: B.1.1.7 variant; Bayesian inference; Markov chain Monte Carlo; Pakistan; phylodynamics; phylogenetics.

Figure 1.

Comparison of viral loads of SGTF and non-SGTF samples. Nasopharyngeal swab samples from travelers arriving from the UK were screened on the cobas® SARS-CoV-2 PCR assay (Roche). Subsequently, all SARS-CoV-2 positive samples were screened for SGTF on the TaqPath COVID-19. For comparison of viral loads on SGTF and non-SGTF sample, the Ct values of cobas Roche SARS-CoV-2 PCR assay (Target 1—ORF1ab gene) and Target 2 (E-gene) were analyzed. There was no statistically significant difference found between Ct values of SGTF and non-SGTF strains for either E or Orf1ab gene amplification.

Figure 2.

Spatial dispersal of B.1.1.221, B.1.36, and B.1.1.7 lineages into Karachi, Pakistan. The spatial dispersal of SARS-CoV-2 viruses from Bahrain (magenta) and the UK (green) into Karachi, Pakistan, was inferred from the respective MCC trees (Figs 4–6). Arrows project the MCC trees’ branches that lead to the seeding events of the B.1.1.212, B.1.36, and B1.1.7 lineages in Pakistan.

Figure 3.

Timeline of introductions of B.1.1.221, B.1.36, and B.1.1.7 lineages into Karachi, Pakistan. Circles represent lineage-specific viral introductions inferred as Pakistani-specific clades in the respective MCC trees (Figs 4–6). The shade depicts if a clade was composed of sequences generated in this study (lighter colour) or previously available (darker colour). The number of sequences in clade is depicted by the number and size of the circles.

Figure 4.

MCC tree inferred for the whole genomes of lineage B.1.1.212. The shade of the branches and tips indicates the inferred location state at the nodes. The ancestral node of run3/s1 (S10) sequence is annotated with the inferred location and probability.

Figure 5.

MCC tree inferred for the whole genomes of lineage B.1.36. The shade of the branches and tips indicates the inferred location state at the nodes. The dark ancestral node of run3/s5 (S5) sequence is annotated with the inferred location and probability. Other sequences collected in Pakistan, but not in this study, are are differentiated by their lighter shade from the ancestral node. Clusters without Pakistan sequences were mostly collapsed to aid visualization of sequences of interest.

Figure 6.

MCC tree inferred for the whole genomes of lineage B.1.1.7. The shade of the branches and tips indicates the inferred location state at the nodes. The ancestral node of run3/s2 (S11), run3/s3 (S6), run3/s4 (S7), and run4/s3 (S20) sequences are annotated with the inferred location and probability. Other sequences collected in Pakistan, but not in this study, are shown by a lighter shade than the ancestral node. Clusters without Pakistan sequences were mostly collapsed to aid visualization of sequences of interest.