Genome sequencing and analysis of Yersina pestis KIM D27, an avirulent strain exempt from select agent regulation

Abstract

Yersinia pestis is the causative agent of the plague. Y. pestis KIM 10+ strain was passaged and selected for loss of the 102 kb pgm locus, resulting in an attenuated strain, KIM D27. In this study, whole genome sequencing was performed on KIM D27 in order to identify any additional differences. Initial assemblies of 454 data were highly fragmented, and various bioinformatic tools detected between 15 and 465 SNPs and INDELs when comparing both strains, the vast majority associated with A or T homopolymer sequences. Consequently, Illumina sequencing was performed to improve the quality of the assembly. Hybrid sequence assemblies were performed and a total of 56 validated SNP/INDELs and 5 repeat differences were identified in the D27 strain relative to published KIM 10+ sequence. However, further analysis showed that 55 of these SNP/INDELs and 3 repeats were errors in the KIM 10+ reference sequence. We conclude that both 454 and Illumina sequencing were required to obtain the most accurate and rapid sequence results for Y. pestis KIMD27. SNP and INDELS calls were most accurate when both Newbler and CLC Genomics Workbench were employed. For purposes of obtaining high quality genome sequence differences between strains, any identified differences should be verified in both the new and reference genomes.

Figure 1. Comparison of SNP and INDELs detection tools based on different sequence assemblies.

Assemblies of KIM D27 sequences from either 454 only, Illumina only, or hybrid assemblies (454 + Illumina) were analyzed for SNPs and INDELs as described in Materials and Methods. The Venn diagram represents the number of identified differences in either CLC Genomics Workbench or Newbler either singly (purple or blue, respectively) or the overlap between both programs (darkened area).

Figure 2. A. Nucmer alignement of putative invasin gene from KIM 10+ and KIM D27.

Regions of homology are depicted as diagonal lines. The 289 bp repeat aligns with itself numerous times and results in the square pattern. B. PCR amplification of the invasin gene from KIM 10+ and KIM D27. The chromosomal region 4228610 – 4234900 was amplified from genomic DNA from either strain. The predicted 6,290 bp product was observed as the major product from KIM D27 (lanes 1 and 2), but a ladder effect was observed when KIM 10+ DNA was used (lanes 3 and 4).

Figure 3. Instances of poly(A) and poly(T) in sequenced genomes.

All 4–9 bp poly(A) and poly(T) in published genomes were counted and plotted per kilobase in each genome.