Multilocus sequence typing as a replacement for serotyping in Salmonella enterica

Abstract

Salmonella enterica subspecies enterica is traditionally subdivided into serovars by serological and nutritional characteristics. We used Multilocus Sequence Typing (MLST) to assign 4,257 isolates from 554 serovars to 1092 sequence types (STs). The majority of the isolates and many STs were grouped into 138 genetically closely related clusters called eBurstGroups (eBGs). Many eBGs correspond to a serovar, for example most Typhimurium are in eBG1 and most Enteritidis are in eBG4, but many eBGs contained more than one serovar. Furthermore, most serovars were polyphyletic and are distributed across multiple unrelated eBGs. Thus, serovar designations confounded genetically unrelated isolates and failed to recognize natural evolutionary groupings. An inability of serotyping to correctly group isolates was most apparent for Paratyphi B and its variant Java. Most Paratyphi B were included within a sub-cluster of STs belonging to eBG5, which also encompasses a separate sub-cluster of Java STs. However, diphasic Java variants were also found in two other eBGs and monophasic Java variants were in four other eBGs or STs, one of which is in subspecies salamae and a second of which includes isolates assigned to Enteritidis, Dublin and monophasic Paratyphi B. Similarly, Choleraesuis was found in eBG6 and is closely related to Paratyphi C, which is in eBG20. However, Choleraesuis var. Decatur consists of isolates from seven other, unrelated eBGs or STs. The serological assignment of these Decatur isolates to Choleraesuis likely reflects lateral gene transfer of flagellar genes between unrelated bacteria plus purifying selection. By confounding multiple evolutionary groups, serotyping can be misleading about the disease potential of S. enterica. Unlike serotyping, MLST recognizes evolutionary groupings and we recommend that Salmonella classification by serotyping should be replaced by MLST or its equivalents.

Figure 1. General overview of the current classification of Salmonella enterica.

Figure 2. Minimal spanning tree (MSTree) of MLST data on 4257 isolates of S. enterica subspecies enterica.

Each circle corresponds to one of 1,095 STs, whose size is proportional to the number of isolates. The topological arrangement within the MSTree is dictated by its graphic algorithm, which uses an iterative network approach to identify sequential links of increasing distance (fewer shared alleles), beginning with central STs that contain the largest numbers of isolates. As a result, singleton STs are scattered throughout the MSTree proximal to the first node that was encountered with shared alleles, even if equal levels of identity exist to other nodes that are distant within the MSTree. The figure only show links of six identical gene fragments (SLVs; thick black line) and five identical gene fragments (DLVs; thin black line) because these correlate with eBGs, which are indicated by grey shading. The serovar associated with most of the isolates in each eBG or singleton ST is indicated by color coding for the 28 most frequent serovars (see legend at lower right). Within each ST, isolates of a different serovar or for which information is lacking are shown in white, except for monophasic variants.

Figure 3. Venn diagram of numbers of eBGs whose STs were assigned to a single cluster by ClonalFrame, BAPS and gene by gene bootstrapping.

Other details are in Table 1.

Figure 4. MSTree of Typhimurium plus its serological variants.

Each circle represents one ST, subdivided into one sector per isolate, flanked by the ST number in small print. The primary links between STs within the MSTree are indicated by straight lines and additional cross-links at the same level of identity are indicated by lines that are terminated by bars. eBG designations are indicated by rounded white boxes. White sectors indicate a lack of serological information. Serological formulas are summarized in Table S3. Other details are as in Fig. 2.

Figure 5. MSTree of Enteritidis, Dublin, Paratyphi B and their serological variants.

Serological formulas are summarized in Tables S4 and S5. Other details are as in Fig. 4. Additional information on Paratyphi B and Java isolates can be found in Tables 2 and S6.

Figure 6. MSTree of 6,7:c:1,5 isolates.

Details are as in Fig. 4 and additional information can be found in Tables 3 and S7.

Figure 7. Variant nucleotides in the fliC gene and variant amino acids in the FliC protein.

Sequences of fliC (Table S8) from isolates investigated here and additional sequences from GenBank with ≥97% BLAST identity and ≥97% coverage were trimmed to a length of 1344 bp, beginning at nucleotide 73 of the fliC gene of strain LT-2. The translated 448 amino acid sequences begin at amino acid 24. The figure shows all differences relative to the uppermost sequence (strain 6631/88), with nucleotide differences at the left and amino acid differences at the right. FliC protein groups were assigned with the help of a UPGMA tree (Fig. S4) and are indicated at the far right. Strain and serovar designations are in the center, followed by MLST ST and eBG designations for the strains investigated here.

Figure 8. Variant amino acids in the FljB protein.

Sequences of FljB (Table S8) from isolates investigated here and additional sequences from GenBank with ≥95% BLAST identity and 100% coverage were trimmed to a length of 440 amino acids, beginning at amino acid 36 in the FljB protein from strain LT-2. The figure shows all differences relative to the uppermost sequence (strain SL3261). FljB protein groups were assigned with the help of a UPGMA tree (Fig. S7) and are indicated at the right, together with the serological factors in the phase 2 flagellar antigen. Strain and serovar designations are at the left, followed by MLST ST and eBG designations for the strains investigated here.