Full text loading...
Chapter 11 : Strain Typing Using Multiple “Variable Number of Tandem Repeat” Analysis and Genetic Element CRISPR
Category: Clinical Microbiology; Bacterial Pathogenesis
This chapter reviews the current status and future prospects of two very highly promising high-throughput genotyping approaches, i.e., multiple-locus VNTR analysis (MLVA) and clustered regularly interspaced short palindromic repeats (CRISPR) analysis. Tandem repeats (TRs) often are found in families of intergenic elements. Such TRs are frequently used for genotyping because of their high degree of polymorphism. For example, MLVA schemes for M. tuberculosis, Streptococcus pneumoniae, P. aeruginosa, S. aureus, and Leptospira interrogans use TRs within intergenic elements. One potential problem in trying to make strain typing using variable number of tandem repeat (VNTR) readily available in multiple laboratories is that many of the TR alleles are too large for routine capillary electrophoresis. In species of more recent emergence, such as Y. pestis, almost all strains possess the same number of CRISPRs, although the number of spacers and DRs varies, and their position on the chromosome changes due to rearrangements. Comparison of the four available Y. pestis sequenced genomes shows the existence of three CRISPRs, two of which present a different number of spacers. The analysis of a large collection of strains showed that all three loci could vary in size, although one was more polymorphic than the others. In this study, a large group of isolates from outbreaks in Vietnam in the 1960s revealed that the CRISPRs acquired new spacers in a polarized fashion and that these spacers originated either from a bacteriophage or from chromosomal genes.
(A) Analysis of six S. pneumoniae strains with the two VNTR loci Spneu36 and Spneu37 by agarose gel electrophoresis. The repeat unit is 45 bp long. The number of repeats can be evaluated visually and is indicated under each PCR product. (B) Left panel, result of the multiplex amplification of 6 VNTRs in eight S. aureus strains as visualized by agarose gel electrophoresis; right panel, capillary electrophoresis analysis of sample 2, in which each locus is individualized owing to colors and known locus size range.
Result of a query in TRF. A TR containing 7.5 copies of a 24-bp sequence was found in the 2,956-bp LegC5/Lgt3 glucosyltransferase gene ( 5 ) of L. pneumophila 130b (accession number EU495330), between nucleotides 2080 and 2262. This tandem repeat is polymorphic in the species and was investigated under the name Lpms13 ( 77 ).
Query of tandem repeats in S. aureus strain Mu50 genome using the Microorganism Tandem Repeat database. (A) Result for a query of repeat units longer than 50 bp, repeated at least four times, with an internal conservation of 80% or more. Six TRs possess the required characteristics. The last column on the right of the table indicates the possible existence of a TR family (with a similar repeat sequence) by activating the Group button. (B) Result obtained by activating the Group button for the TR in position 855 kb in panel A. A TR family present in multiple locations is identified in which the repeat unit is 55 to 59 bp long (the TR unit at position 847705 corresponds to two repeats, 58 and 59 bp long).
Clustering of Brucella reference strains using the categorical distance coefficient and the UPGMA aggregation method with different MLVA panels. (Right) Eight highly variable loci are used, representing panel 2A (3 loci) and 2B (5 loci) ( 39 ). Although the panel is highly discriminatory, the suggested clustering is not very meaningful. For instance, three strains derived from marine mammals (arrows), which are known to be closely related by MLST data, are dispersed ( 58 ). (Center) Eight loci (panel 1) with a stronger phylogenetic value than those in the right-hand panel are used. Clustering makes much more sense (arrows). (Left) The whole dataset (16 primers) is used, but a lower weight is given to the most variable loci as described ( 39 ). Panel 1, panel 2A, and panel 2B loci are each given a relative weight of respectively 2, 1, and 0.2 when calculating the distance matrix. As a result, the assay is very highly discriminatory, while keeping phylogenetic value.
Methods for showing the clustering of multiple strains. The data set is the same as used to produce Fig. 4 ; numbers refer to strains as they are listed from top to bottom in the left panel of Fig. 4 . (A) Maximum parsimony analysis. Strain 6 is incorrectly associated with the B. suis, B. ovis, B. canis cluster. (B) Minimum spanning tree analysis. Long branches connecting distantly related strains or clusters are drawn in dotted lines, since they are usually of little value in this kind of analysis.
Structure of a CRISPR and mode of evolution. (A) The basic structure of a CRISPR locus is illustrated with the leader region from which the locus may be transcribed and four spacers. (B) A new spacer has been acquired, next to the leader. (C) Illustration of interstitial losses, new acquisition, IS element insertion.
Output of CRISPRfinder for strain Actinobacillus succinogenes 130Z (accession number NC 009655). Six spacers are present at CRISPR locus NC_009655_1. The DR is 36 bp long.