(A) Circular representation of the L. welshimeri genome. The first circle represents the scale in kilobases starting with the origin of replication at position 0. The second circle shows the distribution of CDS (gray) in the leading and lagging strand. rRNA operons are colored in blue, a putative prophage region in red, and tRNAs in magenta. The third circle indicates the clusters of orthologous groups (COG) classification for each gene and the distribution of COG classes within the genome. The color scheme and categories are according to the conventions of the COG database. The fourth circle indicates the deviation of the GC content average, with values greater than zero in red and those less than zero in blue. The innermost circle displays the GC skew ([G+C]/[G−C]), with values greater than zero in purple and those less than zero in orange. For the GC skew analysis, we applied a sliding window of 1,000 bp with an overlap of 500 bp. The figure was generated by using the program GenomeViz (16). (B) Identification of deleted gene regions in L. welshimeri and L. innocua corresponding to positions in the L. monocytogenes EGD-e genome. The first circle represents the scale in kilobases, starting with the origin of replication at position 0. The second circle shows the distribution of CDS in L. monocytogenes (gray). Genes that are colored are those deleted in L. welshimeri and L. innocua compared to those in L. monocytogenes EGD-e: red, common deletion in both apathogenic strains; yellow, specific deletion in L. welshimeri; and orange, specific deletion in L. innocua. Core genes were analyzed by GECO by using bidirectional pairs, including prophages and paralogous genes. The third circle indicates the locations of genes coding for internalins (green), LPXTG-motif-harboring proteins (black), and phage-related proteins, pseudogenes, transposase genes, and integrase genes (blue). The fourth circle shows “alien genes” (light blue), which are genes of deviant codon usage identified by SIGI (24). The fifth circle indicates the deviation of the GC content, with values greater than zero in red and those less than zero in blue. The innermost circle displays the GC skew ([G+C]/[G−C]), with values greater than zero in purple and those less than zero in orange. For the GC skew analysis, we applied a sliding window of 1,000 bp with an overlap of 500 bp. The figure was drawn using GenomeViz software (16).

Confirmation of deleted gene regions in L. welshimeri compared to those in L. monocytogenes EGD-e using PCR. Lanes: M, 1 kb Plus DNA ladder (Gibco); 1, L. welshimeri SLCC5877 (1/2b); 2, L. welshimeri SLCC5828 (6b); 3, L. welshimeri SLCC6199 (6a); 4, L. welshimeri SLCC5334 (6b); 5, L. welshimeri SLCC5333 (6b); 6, L. welshimeri 18121; and 7, L. welshimeri 18217. Colony PCR was performed by using the Expand high-fidelity PCR system (Roche) by following the protocol supplied by manufacturer. Primer sequences for all amplicons are available in the supplemental material. All amplified regions reveal clonality of the species L. welshimeri.

Analysis of six chromosomal regions of L. monocytogenes EGD-e in comparison to those of L. innocua, a close relative of L. monocytogenes EGD-e, and L. welshimeri showing the reductive evolution of Listeria spp. With increasing distance from the common ancestor, deletion events occurred more frequently. Comparative analysis was performed using the GECO application of bidirectional pairs.

Comparative representation of the chromosomal “hot spot” region between lwe0390 and lwe0431 of L. welshimeri, including cell wall-associated genes, which are marked. Putative regions of horizontal gene transfer and a rearrangement of the F_oF₁-ATP synthase region are boxed. Bioinformatic analysis was performed with GECO by using bidirectional pairs.

Comparative analysis of six major surface protein classes for L. monocytogenes EGD-e, L. innocua, and L. welshimeri as predicted by Augur. Paralogs are not included. Orthologs were identified by unidirectional BLASTP analysis.

Representative overview of the ami region in L. monocytogenes EGD-e, L. innocua, and L. welshimeri. GW modules were identified by using the multiple sequence alignment program CLUSTALW 1.8.3. Repeated GW modules are indicated as R1 to R4 (black boxed) as previously described by Milohanic et al. (27). The flanked regions of the ami locus are highly conserved as identified with GECO by using bidirectional pairs.

An integrated view of the metabolic pathway comparison between L. monocytogenes EGD-e, L. innocua, and L. welshimeri. The map illustrates biochemical pathways for main energy productions, such as glycolysis, starch metabolism, pentose phosphate pathway, and tricarboxylic acid cycle, which are necessary to maintain the survival of the strains. Several amino acid pathways and nucleic acid metabolism are involved as well (pathway names are colored blue). The lipid and fatty acid pathways are omitted since they are incompletely present. Closed red arrowheads describe the corresponding enzymes/reactions absent in L. welshimeri; however, they exist in both L. monocytogenes and L. innocua, whereas open red arrowheads show the enzymes/functions specifically present in L. monocytogenes. Closed blue arrowheads indicate the reactions present in only L. welshimeri, not L. monocytogenes or L. innocua. The gray line means the reaction is believed to occur spontaneously. Each gene product with a predicted function in ion or solute transport is illustrated in the figure. Proteins are grouped by substrate specificity, with transporters for cations (green), anions (pink), carbohydrates and organic acids (orange), and amino acids and peptides (blue). Ion-coupled permeases are represented by ovals, while ABC transport systems are shown as ovals. This construction is concluded from the latest annotation data (NCBI) (C. Liang, unpublished data). The functional orthologs in silico are defined using a threshold of 50% identity and 75 to 125% alignment coverage rate. The resulting strain-specific genes were inspected using InterPro and ExPASy protein domain analysis platforms. AICAR, 5-amino-1-(5-phospho-d-ribosyl)imidazole-4-carboxamide; E4P, erythrose-4-phosphate; DHF, dihydrofolate; G-1-P, glucose-1-phosphate; Dahp, 3-deoxy-d-arabino-heptulosonate-7-phosphate; F-6-P, fructose-6-phosphate; G-6-P, glucose-6-phosphate; GAP, glyceraldehyde-3-phosphate; OAA, oxaloacetate; 6PGluco, 6-phosphogluconate; PRPP, 5-phosphoribosyl 1-pyrophosphate; R5P, ribose-5-phosphate; Ru5P, ribulose-5-phosphate; S7P, sedaheptulose-7-phosphate; TCA, tricarboxylic acid; TCOylcine, 3α,7α,12α-trihydroxy-5β-cholan-24-oylglycine; X5P, xylulose-5-phosphate; The enzymes encoded by the indicated genes are as follows: lwe0243, xylose isomerase; lwe0244, glycerol kinase; lmo1081, glucose-1-phosphate thymidyl transferase; lmo1082, dTDP-sugar epimerase; lmo1083, dTDP-d-glucose 4,6-dehydratase; lmo1084, DTDP-l-rhamnose synthetase; lmo2023, l-aspartate oxidase; lmo2067, conjugated bile acid hydrolase; lmo2143, mannose-6-phosphate isomerase; and lmo2848, l-rhamnose isomerase.

Phylogenomic tree of three listerial species, L. welshimeri, L. monocytogenes EGD-e, and L. innocua, and B. subtilis. The phylogenomic analysis is based on whole-genome sequence data using MAVID (3). The gain (left arrow) and loss (right arrow) of genes of L. welshimeri and L. innocua were calculated in relation to their putative ancestor, L. monocytogenes EGD-e, by using cluster analysis, including prophages and paralogous genes.