Article Figures & Data
Figures
- Fig. 1.
Dendrogram showing UPGMA cluster analysis of GCS and GGS. Strain designations at branch tips are listed in the Code for dendrogram column in Table 1, showing group carbohydrate (C or G) followed by emm type. Three branch points showing large clusters of GCS or GGS isolates are indicated by asterisks.
- Fig. 2.
ML gene trees of housekeeping loci. Unrooted radial gene trees, generated by the ML method, contain all reported GAS alleles (14) and all GCS-GGS alleles listed in Table 1. Bootstrap values of ≥80% are indicated; GCS-GGS alleles are highlighted in bold. One highly divergent allele listed in Table 1(recP116) falls outside of sequence similarity clusters I and II. The most appropriate models for DNA substitution were determined to be as follows: HKY85 (for mutS, recP, andyqiL), HKY85+G (for gtr, murI, andxpt), and TrN+I+G (for gki). Scale bars indicate the number of nt substitutions per site. Trees are shown formutS, xpt, and yqiL; the complete set of gene trees for all housekeeping loci is available from the authors upon request.
- Fig. 3.
Polymorphic nt sites among housekeeping alleles of GAS, GCS, and GGS. Polymorphic nt sites are shown for selected GCS and GGS alleles (three digits) and GAS alleles (two digits) (14). GCS-GGS alleles identical to a GAS allele are indicated. GenBank accession numbers for all GCS-GGS alleles listed in Table 1 are as follows: AF332626 to AF3326351 (gki), AF332636 to AF332651 (gtr), AF332797 to AF332808 (murI), AF330232 toAF330237 (mutS), AF332810 to AF332825 (recP),AF332826 to AF33284 (xpt), and AF332841 to AF332856 (yqiL). The gtr alleles are truncated 7 bp at the 5′ end (and 13 bp at the 3′ end) relative to the gtr alleles from GAS, as reported previously. Alignments are shown for some alleles of mutS, recP, xpt, and yqiL; the complete set of alignments for polymorphic nt sites among all GAS and GCS-GGS alleles is available from the authors upon request.
- Fig. 4.
Split decomposition analysis. The effect of recombination on evolutionary relationships was assessed by split decomposition (19, 22, 42). This method depicts multiple pathways linking sequences and allows visualization of the extent of conflicting phylogenetic signals. Thus, recombinational events are depicted as an interconnected network of phylogenetic relationships. A lack of treelike structure was evident for both recP andxpt alleles; splits graphs for recP are available upon request fron the authors. This finding was consistent with the visual inspection of polymorphic nt sites (Fig. 3). Annotated split decomposition graphs are shown for xpt. Graph on the left includes all alleles except xpt112 through xpt115(all cluster II); these alleles were removed because their long branch lengths prevented resolution of central networks. Uncorrected Hamming distances were used; similar results were obtained with other estimated distance measures (Kimura 3-ST model and Jukes-Cantor). Branch lengths are drawn to scale. The fit parameters improved after the additional removal of alleles representing the longer branches, resulting in further resolution of inner branches (splits graph shown to the right). For xpt (right graph), removed alleles are the remaining cluster II alleles (xpt108 through xpt111, xpt28, and xpt29). A fit parameter of 100% indicates that all phylogenetic information is represented by the graph. Splits graphs containing cluster I and II alleles for each of the gki, gtr, murI, mutS, and yqiL loci showed no evidence of networking.
- Fig. 5.
Model for evolution of GAS, GCS, and GGS. Model for evolution based on selectively neutral housekeeping loci shows divergence of S. pyogenes and S. dysgalactiaesubsp. equisimilis into cluster I and II alleles, respectively, followed by more recent interspecies gene flow that is dominated by movement of cluster I alleles from GAS to GCS-GGS. This model also shows that neutral gene flow from GAS donors to GCS-GGS recipients tends to involve larger blocks of DNA (allelic replacements) than the genetic material transferred from the reverse direction (intragenic recombination).
Tables
- Table 1.
Epidemiological properties and MLST analysis of GCS and GGS
Strain Tissuea Diseasea Yr Locationb emm type Group carbohydrate ST Allele assignmentsc Code in dendrogram gki gtr murI mutS recP xpt yqiL 4236 Sterile Invasive ND USA stG643 G 1 101 106 104 101 116 101 107 GstG643 4031 Blood Bacteremia 1998 Argentina stG4831 G 2 102 102 103 106 104 114 106 GstG4831 4949 URT None 1996 India stC839 C 3 103 104 109 106 103 107 105 Cst839-1 4241 Sterile Invasive ND USA stC839 C 4 103 104 112 106 101 107 112 Cst839-2 4231 Sterile Invasive ND USA stG653 C 5 103 105 111 106 101 107 116 CstG653 5341 URT ND 1920s UK stG93464 C 6 103 111 109 106 101 107 110 Cst93464 SS1069 URT ND 1974 N.J. stC839 C 7 103 111 109 106 101 107 112 Cst839-3 MGAS338 Sterile Invasive 1980s USA stC839 C 8 104 102 109 106 101 107 104 Cst839-4 4966 URT Pharyngitis 1996 India stC839 C 9 104 111 109 106 101 107 107 Cst839-5 4277 Sterile Invasive ND USA stC36 G 10 105 106 109 103 104 111 109 GstC36 5344 Skin Wound 1980s UK stC5344 C 11 105 106 112 106 113 105 111 CstC5344 4232 Sterile Invasive ND USA stG10 G 12 105 109 109 106 113 104 113 GstG10-1 1891 Sterile Cellulitis 1999 Ill. stG480 G 13 106 101 101 106 113 104 102 Gst480-1 D421 Skin Impetigo 1971 Trinidad stG480 G 14 106 103 102 106 113 104 103 Gst480-2 4030 Blood Bacteremia 1998 Argentina stG11 G 15 106 104 105 106 113 104 106 GstG11-1 4234 Sterile Invasive ND USA stG480 G 16 106 108 105 104 113 104 111 Gst480-3 4255 Sterile Invasive ND USA stG653 G 17 106 108 105 106 114 104 107 GstG653 5353 Skin Ulcer 1980s UK stG11 G 18 106 110 109 106 113 104 113 GstG11-2 5357 Skin Ulcer 1980s UK stG652 G 19 107 112 108 105 111 109 108 GstG652 1778 URT ND 1993 Wyo. stC36 C 20 108 106 105 102 102 102 110 CstC36-1 4288 Sterile Invasive ND USA stG485 C 21 108 106 105 106 107 104 114 Cst485 4265 Sterile Invasive ND USA stC36 C 22 108 106 106 106 108 106 108 CstC36-2 4242 Sterile Invasive ND USA stG485 G 23 108 107 109 105 102 104 115 Gst485-1 4276 Sterile Invasive ND USA stG10 G 24 108 109 107 105 105 110 116 GstG10-2 3296 ND Unknown 1994 Ariz. stG485 G 25 108 109 109 105 102 104 116 Gst485-2 5354 Blood Bacteremia 1980s UK stG62467 G 26 108 109 109 106 107 113 113 Gst62467 SS957 ND ND 1969 ND stC957 C 27 108 111 105 106 109 104 109 CstSS957 SS188 URT ND 1941 USA stC74a C 28 108 113 109 106 108 112 111 CstC74a 4286 Sterile Invasive ND USA stG6 G 29 108 113 109 106 110 115 109 GstG6 1775 URT ND 1993 Wyo. stC36 C 30 108 114 110 102 107 101 108 CstC36-3 5345 Blood Bacteremia 1983 UK stG5345 C 31 108 115 109 105 112 108 110 CstG5345 3110 Blood Invasive 1994 Ariz. stG643 C 32 108 116 109 105 115 103 115 CstG643 4247 Sterile Invasive ND USA stC74a G 33 109 109 109 105 102 104 110 GstC74a 4951 URT ARF 1996 India stG4974 G 34 110 101 109 106 106 115 101 GstG4974 ↵a Isolation of a GCS or GGS organism from a subject with noninvasive disease does not necessarily imply that the organism caused that disease. Sterile, a normally sterile site; URT, upper respiratory tract; ARF, acute rheumatic fever. ND, not determined.
↵b UK, United Kingdom.
↵c Allele assignments in bold are assigned to cluster II, whereas the allele in bold italics (recP116) is highly divergent from both sequence similarity clusters I and II (see Table 3 and Fig. 2).
- Table 2.
Housekeeping alleles of GCS and GGS and comparison to GAS
Locusa Length of sequence (bp) No. of GAS alleles (from 212 isolates) No. of GCS-GGS alleles (from 34 isolates) Maximal % divergence Within GAS Within GCS-GGS Between GAS and GCS-GGS gki 498 34 10 2.6 11.7 12.6 gtr 430 32 16 2.6 22.8 24.5 murI 438 24 12 1.4 32.2 32.6 mutS 405 21 6 2.2 4.9 4.9 recP 459 35 16 6.1 6.4 7.6 xpt 450 29 15 2.9 4.9 4.9 yqiL 434 22 16 1.4 26.6 26.9 ↵a For GCS and GGS isolates, thegtr locus was truncated to 430 bp (20 bp smaller than that originally reported for GAS). The allele recP116 from GGS strain 4236 formed a third cluster (not assigned to either cluster I or II) and is excluded from the maximal divergence calculations.
- Table 3.
Summary of sequence clusters for the housekeeping locia
Locus % of GAS alleles represented % of GCS-GGS alleles represented Maximal % divergence Sequence cluster I Sequence cluster II Sequence cluster I Sequence cluster II Within cluster I Within cluster II Between clusters I and II gki 100 0 10 90 2.7 2.5 12.6 gtr 100 0 25 75 2.6 2.4 24.5 murI 100 0 33 67 1.4 2.1 32.6 mutS 100 0 33 67 2.2 0.7 4.9 recP 89 11 38 56 3.1 6.4 7.6 xpt 93 7 47 53 1.8 4.9 4.9 yqiL 100 0 38 62 1.4 4 26.9 ↵a Includes all alleles assigned to cluster I or II. The allele recP116 from GGS strain 4236 formed a third cluster (not assigned to either cluster I or II) and is excluded from this analysis. Cluster II alleles of xpt form two distinct groups (cluster IIa and IIb); the maximal divergence between clusters I and IIa is 3.0%, and that between clusters I and IIb is 3.7%.
- Table 4.
Housekeeping alleles shared by GAS and GCS-GGSa
Locus Shared GCS-GGS allele GCS-GGS emm types represented Group carbohydrate represented No. of GCS-GGS isolates represented GCS-GGS allelic frequency Shared GAS allele GAS emm types represented No. of GAS isolates represented GAS allelic frequency gtr 101 stG4974, stG480 G 2 0.0588 03 1, 87, st4935 18 0.0849 102 stC839, stG4831 C, G 2 0.0588 06 3, 28, 49, 53, 60, 76, 77, 91, st4973, st64/14, stD633, stNS5 40 0.1887 103 stG480 G 1 0.0294 13 93 1 0.0047 104 stC839, stG11 C, G 3 0.0882 26 57 1 0.0047 murI 101 stG480 G 1 0.0294 04 1 22 0.1038 102 stG480 G 1 0.0294 08 3, 4, 8, 19, 24, 26, 29, 34, 52, 66, 78, 89, 93, st1RP31, st2370.1, st4973, st64/14 55 0.2594 103 stG4831 G 1 0.0294 24 70 1 0.0047 104 stG643 G 1 0.0294 02 12, 14, 39, 49, 51, 55, 63, 77, 81, 83, 90, 91, st3365, st4529, st4935, st833, stD432, stD633 30 0.1415 mutS 101 stG643 G 1 0.0294 07 9, 11, 25, 42, 49, 57, 60, 81, 89, stD432 14 0.0660 recP 105 stG10 G 1 0.0294 02 2, 3, 33, 43, 53, 83, 90, 91, 92, 93, st2370.1, st4935, st4973, stNS5 46 0.2170 106 stG4974 G 1 0.0294 04 1, 70,st4529 22 0.1038 xpt 103 stG643 C 1 0.0294 02 1, 4, 12, 33, 42, 52, 53, 60, 76, 87, 91, 92,stNS5 55 0.2594 yqiL 101 stG4974 G 1 0.0294 12 51,st2370.1, st3365 3 0.0142 102 stG480 G 1 0.0294 04 1, 2, 4, 8, 11, 22, 25, 49, 50, 62, 68, 73, 76, 44/61, 90,st833 43 0.2029 103 stG480 G 1 0.0294 15 88, 92 3 0.0142 105 stC839 C 1 0.0294 01 1, 4, 9, 11, 13, 32, 39, 49, 66, 67, 78, 44/61, 82, 89, st2346, st4592, st4935, st64/14 48 0.2264 ↵a Analysis includes 34 isolates of GCS and GGS (S. dysgalactiae subsp. equisimilis) and 212 isolates of GAS. Minimum possible allelic frequencies are 0.0294 and 0.0047 for GCS-GGS and GAS, respectively.
