Melting Curve-based Minimum Multilocus Sequence Typing Of Pathogenic Microorganisms

The infectious disease caused by pathogenic microorganisms is a challenging issue for global public health. Typing of pathogenic microorganisms is helpful for the epidemiological investigation, source tracking, and transmission intervention, which are essential to control the outbreak of the diseases. An ideal typing method should have such advantages as high resolution, ease of use, good repeatability, high efficiency, easy automation, and cost-effectiveness. In this dissertation, we proposed a novel molecular typing method termed as melting curve-based minimum multilocus sequence typing (minim McMLST) as an alternative tool for molecular typing.In chapter one, we first discussed the significances of typing of pathogenic microorganisms. We reviewed current typing technologies with emphasis on those DNA-based methods. A summary of the advantages and disadvantages of these methods was given. The features of an ideal molecular typing were discussed. We then proposed a new multilocus sequence typing analysis based on multicolor melting curve analysis, minim McMLST, by using a minimal number of single nucleotide polymorphisms (SNPs) loci within the sequence types (STs) of multilocus sequence typing (MLST) method.In chapter two, we first detailed the working principle of minim McMLST. The feasibility of this method was then studied using Salmonella as a model pathogen. According to minim McMLST, we retrieved six significant SNPs from1560STs according to the Minimum SNPs software. A seventh SNP site was added in order to differentiate two endemic STs in China. In silico analysis showed that above seven SNPs could classify1560STs of MLST into59melting types (MTs) by minim McMLST. A dual reaction, triplex real-time PCR assay was established to detect the seven SNPs. The assay could accurately and reproducibly type the seven SNPs with a limit of detection of5genomic copies per reaction. Using this assay,167clinical isolates could be separated into17MTs. Of these167isolates,52were validated by sequencing analysis. A100%agreement was achieved between the two methods within the covered range of minim McMLST. Phylogenetic trees analysis of the52isolates using unweighted pair group method with arithmetic mean (UPGMA) algorithm showed that17MTs were obtained by our method while23STs were obtained by sequencing analysis. The Simpson’s index of diversity (D) of minim McMLST was0.9818with respect to the sequencing method.In chapter three, we further increased the number of SNPs loci to improve the discrimination power of the minim McMLST. For this purpose, Staphylococcus aureus was used as a model pathogen owing to its wide diversity. Up to15SNP loci were chosen, which could divide2151STs of S. aureus into288MTs. A4-plex real-time PCR assay was established in three reactions and each was used for5SNPs analysis. The established assay was used to analyze376clinical isolates, from which29MTs were obtained. Of376samples,70samples were analyzed in parallel by sequencing, which resulted in a100%agreement with minim McMLST within its covered range. Phylogenetic trees analysis of the70isolates using UPGMA showed that24MTs were obtained by our method while29STs were obtained by sequencing analysis. The calculated D value of the minim McMLST was0.9842with respect to MLST. Additional clonal complex (CC) analysis of the70isolates using goeBURST-1.2.1software showed that those isolates not discriminated by minim McMLST, although assigned to different STs, were nevertheless classified as identical CC group in MLST. The typing accuracy of minim McMLST was calculated to be97.19%. Finally, we determined the drug susceptibility status with respect to methicillin for each strain. Interestingly, we observed that there was a strong relationship between the MTs and the drug susceptibility status. These results indicated the potential usefulness of molecular typing in surveillance of methicillin-resistant S. aureus.