Genetic diversity of Bordetella bronchiseptica isolates obtained from primates

Abstract

Bordetella bronchiseptica is a highly contagious bacterial respiratory pathogen with a broad host range of wild and domesticated mammals that can cause a variety of clinical disease outcomes ranging from asymptomatic carriage to severe pneumonia. The goal of this study was to evaluate the genetic diversity of B. bronchiseptica isolates obtained from primates and evaluate the antimicrobial resistance harbored by these isolates. Two isolates were identified as belonging to B. bronchiseptica lineage II and 13 isolates represented new sequence types within B. bronchiseptica lineage I clonal complex 6. The lineage II isolates harbored the lowest sequence identity observed across all genes evaluated and did not contain several well characterized virulence and fimbrial genes. Western blotting revealed no reactivity to a lineage II strain when using antibodies generated against pertactin (PRN) from a lineage I-1 strain or antibodies generated against a domain of filamentous haemagglutinin (FHA) from a lineage I-1 strain. Isolates harbored variation within the wbm locus containing genes encoding for the expression of antigenically distinct O-antigen types and the cya operon was replaced by the ptp operon in several isolates, expanding the phylogenetic distribution of this operon replacement. Thirteen isolates exhibited phenotypic resistance to four antibiotic classes tested, however the Bordetella-specific β-lactamase was the only antimicrobial resistance gene identified. Collectively, the data in this report expands the known phylogenetic diversity and genetic variation of B. bronchiseptica isolates.

Keywords: Bordetella bronchiseptica; antimicrobial resistance (AMR); comparative genomics; primate; virulence factor; whole-genome sequencing (WGS).

Figure 1

Hierarchically clustered heatmap displaying the relatedness of Bordetella isolates based on ANI. Pairwise ANI values for analyzed genomes were used to generate a distance matrix heatmap clustered by hierarchical clustering using a complete linkage method with Euclidean distance. Isolate names are provided at the bottom and right side of heatmap. ANI is represented using the color scale legend shown at the left of the heatmap. Dendrograms are on the left and top sides of the heat map. Lineage designations B. bronchiseptica (Bb I-1, Bb I-4, Bb II), B. parapertussis (Bpp I-3), and B. pertussis (Bp I-2) are shown alongside the isolate names on the right side of the heatmap colored according to the legend shown at the left of the heatmap.

Figure 2

Hierarchically clustered heatmap displaying the relatedness of Bordetella isolates based on FHA AA identity. Pairwise AA percent identity of unique FHA sequences from isolate genomes containing a full-length fhaB gene was determined and used to generate a distance matrix heatmap clustered by hierarchical clustering using a complete linkage method with Euclidean distance. Isolate names are provided at the bottom and right side of heatmap. FHA AA identity is represented using the color scale shown left of the heatmap. Dendrograms are on the left and top sides of the heat map. Lineage designations B. bronchiseptica (Bb I-1, Bb I-4, Bb II), B. parapertussis (Bpp I-3), and B. pertussis (Bp I-2) are shown alongside the isolate names on the right side of the heatmap colored according to the legend shown left of the heatmap.

Figure 3

FHA alignment. MAFFT alignment of FHA protein sequences from Bb isolates M2020-1, C2020-8, C2020-7, and M2020-2. FHA from strain KM22 served as the reference sequence. Orange arrows indicate the FHA region corresponding to residues 400–1,025 used to generate polyclonal α-FHA antibodies that recognize the HBD and the region corresponding to residues 2,000–2,525 used to generate polyclonal α-FHA antibodies that recognize the MCD. In the aligned sequences, vertical black lines represent residues that differ from the KM22 FHA reference. The top bar represents mean pairwise identity over all pairs in that column of the alignment; green is 100% identical, yellow is 30 - <100% identical, and red is <30% identical. The height of the bar represents conservation of sequence at that position; a lower height indicates low sequence conservation at that position.

Figure 4

Western blot analysis using α-MCD and α-HBD KM22 antibodies. Western blot analysis of whole-cell bacterial lysates from isolates KM22, TN27 (ΔfhaB), C2020-7, M2020-1, M2020-2, and C2020-8. Membranes were probed with α-MCD (A) and α-HBD (B) antibodies. Sizes are indicated in kilodaltons on the left and the large ~250 kDa polypeptides indicative of multiple processed FHA proteins are indicated.

Figure 5

Hierarchically clustered heatmap displaying the relatedness of Bordetella isolates based on PRN AA identity. Pairwise AA percent identity of unique PRN sequences from isolate genomes containing a full-length prn gene was determined and used to generate a distance matrix heatmap clustered by hierarchical clustering using a complete linkage method with Euclidean distance. Isolate names are provided at the bottom and right side of heatmap. PRN AA identity is represented using the color scale shown left of heatmap. Dendrograms are on the left and top sides of the heat map. Lineage designations B. bronchiseptica (Bb I-1, Bb I-4, Bb II), B. parapertussis (Bpp I-3), and B. pertussis (Bp I-2) are shown alongside the isolate names on the right side of the heatmap colored according to the legend shown left of heatmap.

Figure 6

PRN alignment. MAFFT alignment of PRN protein sequences from Bb isolates M2020-1, C2020-8, C2020-7, and M2020-2. FHA from KM22 served as the reference sequence. In the aligned sequences, vertical black lines represent residues that differ from the KM22 FHA reference. The top bar represents mean pairwise identity over all pairs in that column of the alignment; green is 100% identical, yellow is 30 - <100% identical, and red is <30% identical. The height of the bar represents conservation of sequence at that position; a lower height indicates low sequence conservation at that position.

Figure 7

Western blot analysis using α-PRN antibodies. Western blot analysis of whole-cell bacterial lysates from isolates KM22, TN28 (Δprn), C2020-7, M2020-1, M2020-2, and C2020-8. Membranes were probed with α-PRN antibodies. Sizes are indicated in kilodaltons on the right.

Figure 8

Genomic organization and alignment of the region containing the prn gene. MAFFT alignment of the region containing the prn gene for isolates KM22, MO149, D420, and 12,822 (A) and I328 and M2020-2 (B). Bb isolate and (lineage) shown at left. In the aligned sequences, the prn gene (yellow) and flanking genes (blue) are shown with vertical black lines that represent residues that differ from the KM22 reference. Base pair numbers are displayed at top. The top bar represents mean pairwise identity over all pairs in that column of the alignment; green is 100% identical, yellow is 30 - <100% identical, and red is <30% identical. The height of the bar represents conservation of sequence at that position; a lower height indicates low sequence conservation at that position.

Figure 9

Organization of fimNX locus. Predicted fimbrial protein genes are represented as arrows; gene names refer to names used in Table 3 and are color-coded by nucleotide sequence identity. fimNX locus for isolates (A) M2020-1, M2020-5, M2020-3, C2020-8, C2020-1, C2020-2, C2020-3, C2020-4, C2020-5, C2020-9, C2020-10, C2020-11, C2020-12, C2020-13, C2020-14, C2020-15, C2020-7 and (B) M2020-4 and M2020-2.

Figure 10

Organization of wbm locus from rhesus monkey Bb isolates. Alignment of the wbm locus of C2020-8 to KM22 serving as reference O-antigen serotype O2 (A) and alignment of the wbm locus of C2020-8 to Bpp_ov strain Bpp5 serving as reference (B). wbm gene name designations are provided at bottom of each alignment. Predicted genes within wbm locus are represented as blue arrows and genes flanking the wbm locus are indicated by grey arrows. Unique gene, which is not present in O-antigen serotype O2, is indicated by yellow arrows. Pseudogenes are indicated by pink arrows. Base pair numbers are displayed above black bar at top. In the aligned sequences, vertical black lines represent residues that differ from the reference. The top bar represents mean pairwise identity over all pairs in that column of the alignment; green is 100% identical, yellow is 30 - <100% identical, and red is <30% identical. The height of the bar represents conservation of sequence at that position; a lower height indicates low sequence conservation at that position.

Figure 11

M2020-2 and M2020-4 harbor a novel O-antigen serotype. Alignment of the wbm locus for isolates M2020-2, M2020-4, AU3139I124, and I328. The wbm locus for these isolates includes 16 predicted genes, with functional annotations including predicted formyltransferase genes (green arrows), predicted aminotransferase genes (yellow arrows), predicted methyltransferase genes (purple arrows), predicted asparagine synthase genes (dark blue arrows), predicted glyceroltransferase genes (pink arrows), predicted sugar amidotransferase genes (red arrows), and other genes of undetermined function (light blue arrows). Genes flanking the wbm locus are indicated by grey arrows. Base pair numbers are displayed above black bar at top. In the aligned sequences, vertical black lines represent residues that differ between sequences. The top bar represents mean pairwise identity over all pairs in that column of the alignment; green is 100% identical and yellow is 30 - <100% identical.