IS26 drives the dissemination of bla CTX-M genes in an Ecuadorian community

ABSTRACT The rapid dissemination of extended-spectrum β-lactamase (ESBL)-producing Enterobacterales, mainly Escherichia coli carrying bla CTX-M genes, is a major public health concern due to its successful spread in hospital settings as well as among humans and animals in the community. We characterized ESBL-producing E. coli isolates from children and domestic animals in semirural communities of Ecuador to assess the contribution of horizontal gene transfer of the bla CTX-M genes among E. coli isolates. From 20 selected E. coli isolates (from children and domestic animals) harboring bla CTX-M allelic variants, we identified 16 plasmids carrying bla CTX-M-55 (n = 9), bla CTX-M-65 (n = 5), and bla CTX-M-27 (n = 2), as well as four chromosomes carrying bla CTX-M-65. The backbone structure of plasmids, including replication, maintenance, and plasmid transfer genes, and the synteny was conserved in all plasmids carrying the same bla CTX-M allelic variant. In all plasmids and chromosomes, the bla CTX-M genes were bracketed by two IS26 transposable elements. This study highlights the critical role of the IS26 transposable element for the current mobility of bla CTX-M genes among plasmids or from plasmids to chromosomes, suggesting that IS26-bla CTX-M brackets could be used to study bla CTX-M transmission between humans, domestic animals, and the environment. IMPORTANCE The horizontal gene transfer events are the major contributors to the current spread of CTX-M-encoding genes, the most common extended-spectrum β-lactamase (ESBL), and many clinically crucial antimicrobial resistance (AMR) genes. This study presents evidence of the critical role of IS26 transposable element for the mobility of bla CTX-M gene among Escherichia coli isolates from children and domestic animals in the community. We suggest that the nucleotide sequences of IS26-bla CTX-M could be used to study bla CTX-M transmission between humans, domestic animals, and the environment, because understanding of the dissemination patterns of AMR genes is critical to implement effective measures to slow down the dissemination of these clinically important genes.

The rapid dissemination of bla CTX-M genes deserves close attention.In a previous study, we found 16 E. coli clonal groups (72 E. coli isolates involved, of which 95% carried bla CTX-M genes) associated with either humans or domestic animals in semirural communities in Ecuador, of which 21 E. coli strain pairs (14%) showed evidence of recent transmission between domestic animals and humans (18).In this study, we assessed the contribution of horizontal gene transfer of the bla CTX-M genes among E. coli from humans and domestic animals in these communities.

Genetic environment of bla CTX-M genes
In all the plasmids, the bla CTX-M-55 gene was bracketed by two IS26 transposable elements and located 127 bp downstream of a fragment of ISECp1 insertion sequence (243 bp; 14.7% coverage) truncated by IS26, and 46 bp upstream of the wbuC gene, which codes a cupin fold metalloprotein.Downstream of the wbuC gene, the TnA and bla TEM genes were found, and both were truncated by IS26 (Fig. 4).This structure was the same (99% identity) for all nine bla CTX-M-55 -carrying plasmids (Fig. 4).We use the term IS26-bla CTX-M bracket to indicate the nucleotide sequence containing the bla CTX-M gene that is flanked by two IS26s.In the plasmid from one isolate (ID: 2018082847.3), the IS26bla CTX-M bracket, containing bla CTX-M-55 and identical genes, was in the opposite direction, indicating inversion caused most probably by recombination or transposition of both IS26s (Fig. 1).The nine bla CTX-M-65 gene variants were detected in five plasmids and four chromosomes.Similar to the case of bla CTX-M-55 , the bla CTX-M-65 gene in all cases was bracketed by two IS26s (Fig. 5): in three plasmids (from 2018091135.3, 2018091864.1, and 2018081441.5 isolates) and one chromosome (from 2018102322.3 isolate), IS26-bla CTX-M brackets contained the same genes: fipA gene encoding a conjugal transfer inhibition protein, a hypothetical gene, ISEcp1 fragment, bla CTX-M-65 gene, IS102 insertion sequence, a gene encoding a TonB-dependent receptor, and a gene encoding PAS domain-containing protein; in the two chromosomes (from 2018081453.2 and 201810092.3isolates), the IS26-bla CTX-M bracket contained fewer of the same genes in the same order: fipA gene encoding a conjugal transfer inhibition protein, the hypotheti cal gene, ISEcp1 fragment, and bla CTX-M-65 gene.Although the IS26-bla CTX-M bracket contained the same genes in the same order, some of the genes were located at different distances from each other: IS26 -fipA gene (44 bp: 2018081453.

Phylogenetic analysis of plasmids and IS26-bla CTX-M brackets
To explore the possibility of IS26-bla CTX-M bracket mobilization among different plasmids, we compared the topology of the maximum likelihood phylogenetic trees of the plasmids and the IS26-bla CTX-M brackets carrying either bla CTX-M-65 or bla CTX-M-55 .Even though some plasmid clustering was concordant with IS26-bla CTX-M brackets (e.g., plasmids p2018091166.4and p2018081441.5 show a common ancestor and p2018082847.3and p201809181.3 also share a common ancestor), there were many cases where the clustering of plasmids and IS26-bla CTX-M brackets were discordant (e.g., plasmids p2018092531.2and p2018081440.2share a recent common ancestor while their IS26-bla CTX-M brackets share a recent ancestor with IS26-bla CTX-M brackets from other plasmids, p2018092531.2with p201809181.3, and p2018081440.2with 2018091176.5) (Fig. 6 and 7).These results suggest that many plasmids have not coevolved for some time with their respective IS26-bla CTX-M brackets.

Plasmid evolutionary rate
To determine the rate of plasmid (carrying bla CTX-M genes) evolution, we took advantage of four clonal E. coli strains (2018090418.2and 2018090458.2:0 SNPs; 2018091135.3 and 2018081441.5:90 SNPs) isolated during the same period in the same community (18).This plasmid comparison showed a highly conserved structure with an extremely high nucleotide identity (28 SNPs).There were no plasmid rearrangements, gene insertions, or deletions for plasmids from E. coli strains with 0 SNPs in their core genomes.The regions with lower similarity (69%) corresponded to duplicated sequences of hypothetical genes and intergenic spaces.The plasmid sizes were 121,366 bp and 121,132 bp for plasmids p2018090418.2and p2018090458.2,respectively (Fig. 3; Table S2).IS26 also bracketed the two bla CTX-M-27 genes, and this region presented 100% identity.The two plasmids carrying bla CTX-M-65 from E. coli strains with 90 SNPs in their core genomes also showed a conserved structure with high nucleotide identity (209 SNPs).There were no rearrange ments found; however, gene insertion and deletion regions were identified.The lower similarity (71%) corresponded to duplicated sequences of IS26 and intergenic spaces.The plasmid sizes were 127,219 bp and 105924 bp from 2018091135.3 and 2018081441.5 isolates, respectively (Fig. 2; Table S2).
To further strengthen the results of this analysis, we determined the number of SNPs and length differences between plasmids sequenced in duplicate, as well as the thresholds to define variations due to inherent variations of sequencing and bioinformat ics analyses.The mean SNP difference was 0.06%, and the mean length difference was 0.10% (Table S2).

ESBL-producing E. coli whole genomes
In a previous study that aimed to study the transmission of cephalosporin-resistant E. coli between domestic animals and humans, we analyzed E. coli strains that showed high chromosomal similarity (18).To study horizontal gene transfer, we selected 125 ESBL-producing E. coli (from domestic animals and humans) carrying the most common bla CTX-M allelic variants identified in these communities (18).We used the ResFinder database (19), with 90% minimum match and 60% minimum length (18).Sixty-nine carried the bla CTX-M-55 allelic variant (children = 22, dogs = 20, chickens = 27) and 56 carried the bla CTX-M-65 allelic variant (children = 7, dogs = 34, chickens = 15).

Characterization of bla CTX-M carrier contigs
Mobile genetic elements of 125 bla CTX-M gene variant carrier contigs were identified using the command-line version of MobileElementFinder 1.0.3 (20) with 80% minimum match and 10% minimum length.Then, bla CTX-M-55 and bla CTX-M-65 allelic variant carrier contigs were separately aligned in Unipro UGENE (21) to establish them into groups based on the similarity of their nucleotide sequences.The bla CTX-M-55 gene carrier contigs were classified into six groups, whereas the bla CTX-M-65 gene carrier contigs were classified into eight groups.From the established groups in which there were contigs from ESBL-producing E. coli whole genomes isolated from more than one species, we randomly selected one isolate from each species for further analyses.Additionally, FIG 6 Comparative phylogenetic analysis of complete sequences of plasmids carrying bla CTX-M-65 allelic variant with their harbored IS26-bla CTX-M bracket.The evolutionary history was inferred using maximum-likelihood phylogenetics with a general time reversible tree built using the genetic distance.The phylogenetic tree on the left was based on complete sequences of plasmids, whereas the tree on the right was based on IS26-bla CTX-M sequences.Labels show the isolate ID assigned based on the host ID followed by its isolate number.The origin of the isolate harboring the plasmid is indicated by font colors (child: blue; dog: orange; chicken: green).Colored arrows relate the plasmid to its corresponding IS26-bla CTX-M bracket.Bootstrap values (>80) based on 100 replications are shown at the tree nodes.

FIG 7
Comparative phylogenetic analysis of complete sequences of plasmids carrying bla CTX-M-55 allelic variant with their harbored IS26-bla CTX-M bracket.The evolutionary history was inferred using maximum-likelihood phylogenetics with a general time reversible tree built using the genetic distance.The phylogenetic tree on the left was based on complete sequences of plasmids, whereas the tree on the right was based on IS26-bla CTX-M sequences.Labels show the isolate ID assigned based on the host ID followed by its isolate number.The origin of the isolate harboring the plasmid is indicated by font colors (child: blue; dog: orange; chicken: green).Colored arrows relate the plasmid to its corresponding IS26-bla CTX-M bracket.Bootstrap values (>80) based on 100 replications are shown at the tree nodes.
to determine the rate of plasmid (carrying bla CTX-M genes) evolution, we chose four contigs of four different ESBL-producing E. coli whole genomes: two with 0 SNPs in their core genomes (carrying bla CTX-M-27 allelic variant) and two with 90 SNPs in their core genomes (carrying bla CTX-M-65 allelic variant) (18).

DNA extraction of bla CTX-M gene carrier plasmids
Each of the 20 selected bla CTX-M allelic variant carrier E. coli isolates was reactivated on MacConkey Lactose agar (Difco) supplemented with ceftriaxone (2 mg/L) overnight at 37°C, after which one colony was selected and inoculated into 2 mL of Lysogeny Broth (LB) supplemented with ceftriaxone (2 mg/L) with shaking at 250 rpm at 37°C for 9 h.Then, 3 mL of fresh LB media with antibiotic were added to the culture at 37°C for 12-16 h while shaking at 250 rpm.Plasmid extraction from the 20 isolates was performed in duplicate using Pure Yield Plasmid Miniprep System (Promega) according to the protocol provided by the manufacturer.Duplicates were placed into a single microtube before being freeze-dried and resuspended in nuclease-free water to achieve a minimum plasmid DNA concentration of 53 ng/µL.Extracted plasmid DNA concentrations were measured using a Qubit 1× dsDNA High Sensitivity assay kit and a Qubit 4.0 fluorometer (Thermo Fisher Scientific).

Genomic DNA extraction
For the four isolates whose sequences could not be circularized or in which the bla CTX-M allelic variant could not be identified after sequencing and assembly, genomic DNA extraction was performed using 12-16 h cultures obtained as mentioned above, using DNeasy Blood & Tissue kit (Qiagen).DNA was eluted in nuclease-free water, and a minimum DNA concentration of 53 ng/µL was obtained, measured using a Qubit 1X dsDNA High Sensitivity assay kit and a Qubit 4.0 fluorometer (Thermo Fisher Scientific).

Conjugation experiments
Conjugation assays were performed to evaluate the conjugative capacity of bla CTX-M carrier plasmids.The 20 selected bla CTX-M allelic variant carrier E. coli isolates were used as donors and E. coli TOP10 (Invitrogen) resistant to rifampin as the recipient (22).Prior to conjugation experiments, the phenotypic AMR profile of each donor strain was confirmed against the same 12 antimicrobials used in our previous study (18) using the disk diffusion method according to the Clinical and Laboratory Standards Institute guidelines (23).Among the 12 antimicrobials, we used ceftazidime (CAZ; 30 µg), CTX (30 µg), cefepime (FEP; 30 µg), and AMC (20 per 10 µg), with which we carried out the double-disk synergy test (24).Phenotypic expression of ESBL was evaluated by placing a disk of AMC surrounded by disks of CAZ, CTX, and FEP (30 mm apart, center to center).An extension of the edge of the CAZ, CTX, or FEP inhibition zone toward AMC disk as a keyhole effect was interpreted as positive for the ESBL phenotype (24,25).For each conjugation experiment, the donor and recipient strains were grown in LB at 37°C for 18 h, and the strains in the logarithmic growth phase were mixed and incubated at 37°C for 18 h.Transconjugants were selected by the spread plate method onto LB agar-containing ceftriaxone (2 mg/L) and rifampin (100 µg/mL) as previously described (22).The phenotypic expression of ESBL by DDST and antimicrobial phenotypic profile by disk diffusion of transconjugants were evaluated to determine the acquired AMR.

MinION library preparation and sequencing
According to the manufacturer's instructions, library preparation was performed using the Rapid Barcoding Sequencing Kit (SQK-RBK004) (Oxford Nanopore Technologies).The constructed libraries were loaded into R9.4.1 (FLO-MIN106D) flow cells and sequenced on a MinION Mk1B sequencing device for approximately 24 h using the MinKNOW software 22.03.5 (Oxford Nanopore Technologies).We sequenced a random selection of three plasmid DNA samples twice, obtained from the same bacterial cultures, to determine the intrinsic variations of sequencing and bioinformatic analyses.Basecall ing was carried out with Guppy 6.0.6 (https://community.nanoporetech.com) in a fast basecalling model.Raw data were demultiplexed, and adapters and barcodes were trimmed using Porechop 0.2.4 with default parameters (26).Then, raw reads were filtered with Filtlong 0.2.1 using a minimum read length threshold of 1 kbp and keeping 95% of the best reads (measured by bases).Filtered reads metrics were assessed using NanoPlot 1.40.0 (27).

Assembly of plasmids and chromosomes carrying bla CTX-M gene
De novo assembly of complete plasmids and chromosomes with filtered reads was carried out using Flye assembler 2.8.1-b1676 (28).Different assembly parameters were evaluated to optimize the assembly of the circular sequences of interest, due to the unknown plasmid size and possible contamination with chromosomal DNA in the plasmid DNA samples.The genome-size option was set at 0.1, 1, 2, 3, 4, and 5 m each, with the asm-coverage option set at 10, 15, 20, 30, 40, and 50, with all combinations.The plasmids option was specified to allow recovery of unassembled short plasmids.Additionally, the meta-assembly option (29) was also assessed.

bla CTX-M gene variant carrier plasmid and chromosome annotation
AMR genes and plasmid types were identified with the Resfinder (19) and PlasmidFinder (30) databases, respectively, using ABRicate tool 1.0.1 (31) in all plasmids and chromo somes circularized.Each bla CTX-M gene variant carrier plasmid and chromosome was rotated with the task fixstart of Circlator tool 1.5.5 (32) and a fasta file with 7171 ancestral sequences of the most common replication initiators (33) to fix the start position of each plasmid and chromosome.As plasmids usually have more than one replication initiator, they were aligned with MAFFT algorithm and manually modified to establish the same replication origin in cases where possible in Unipro UGENE 40.1 (21).For the plasmid sequences of the DNA samples sequenced twice, we used Unipro UGENE to align them and obtain the consensus sequences using the Levitsky algorithm.The number of SNPs between plasmids sequenced twice and between plasmids from clonal E. coli strains was determined using Snippy 4.6.0(34).Each bla CTX-M gene variant carrier plasmid and chromosome was annotated with the National Center for Biotechnology Information (NCBI) Prokaryotic Genome Annotation Pipeline (PGAP) (35).The output GenBank file was manually curated using data obtained from different annotation tools.Mobile genetic elements were identified using the command line version of MobileEle mentFinder 1.0.3 (20), and AMR genes and plasmid types were again predicted after rotation of sequences with the Resfinder and PlasmidFinder databases, respectively.The genomic structure comparison among plasmids and chromosomes and among IS26-bla CTX-M brackets was performed according to BLASTn using Easyfig 2.2.2 (36).

Similarity of IS26-bla CTX-M brackets
The IS26-bla CTX-M bracket sequences similarity search was carried out with the IS26bla CTX-M-55 bracket (from p2018081440.2),and with the two most common IS26-bla CTX- M-65 brackets identified (from p2018091135.3and p201810092.3,respectively), using BLASTn without inclusion or exclusion parameters.All 100 match sequences for each of the three IS26-bla CTX-M brackets, excluding synthetic constructs, were selected.We used Unipro UGENE 40.1 to convert sequences to their reverse complement as neces sary to ensure that all selected sequences are in the same orientation with respect to the IS26-bla CTX-M brackets.Then, the number of SNPs between each of the selected sequences and their respective IS26-bla CTX-M brackets was determined using Snippy 4.6.0.

Phylogenetic analyses
To investigate the possibility of IS26-bla CTX-M bracket mobilization among different plasmids, we constructed maximum likelihood phylogenetic trees of the plasmids and the IS26-bla CTX-M brackets carrying bla CTX-M-65 or bla CTX-M-55 .Due to inverted sequen ces in plasmids that concealed their phylogenetic relationships, we identified inverted DNA sequences using Easyfig 2.2.2 (36), and we manually placed these sequences in the same direction using Unipro UGENE 40.1 (21) before phylogenetic tree construction.We also carried out BLASTn analyses of one representative plasmid of each phylogenetic tree cluster obtained to identify the best match plasmids.Additionally, BLASTn analyses of IS26-bla CTX-M-55 bracket, and the most common IS26-bla CTX-M-65 bracket identified, were performed to establish the best match plasmid harbored by Kluyvera spp.Then, from the Kluyvera plasmid sequence more similar to the IS26-bla CTX-M-65 bracket, we carried out a new BLASTn comparison to select the four best match plasmid sequences to include them in a phylogenetic tree based on all of our plasmids carrying either bla CTX-M-65 , bla CTX-M-55 , or bla CTX-M-27 .All maximum likelihood phylogenetic trees were performed with the general time reversible model using RaxML-NG 0.6.0(37).The visualization and edition of phylogenetic trees were carried out using iTOL v6 (38) and GIMP 2.10 (https:// www.gimp.org),respectively.

DISCUSSION
Our results suggest that IS26 mobilizes bla CTX-M-65 , bla CTX-M-55 , and bla CTX-M-27 allelic variants among different plasmids (Fig. 1 to 3).Even though some plasmids carrying the same bla CTX-M gene share a more recent ancestor, which may suggest plasmid co-evolution with bla CTX-M genes (Fig. S1), the phylogeny of the IS26-bla CTX-M did not correspond to plasmid phylogeny (Fig. 6 and 7).Additional evidence of IS26 contribution in bla CTX-M mobility is the presence of the identical IS26-bla CTX-M-65 bracket in the plasmids of three different isolates and the chromosome of another isolate (Fig. 5).We also found evidence of different evolutionary trajectories of bla CTX-M genes and plasmids; genes bla CTX-M-27 and bla CTX-M-65 belong to phylogenetic group 1, whereas gene bla CTX-M-55 belongs to phylogenetic group 9 (8).However, our results show that plasmids carrying bla CTX-M-27 share a more recent common ancestor with plasmids carrying bla CTX-M-55 than with plasmids carrying bla CTX-M-65 (Fig. S1).These observations suggest that the plasmids have exchanged IS26-bla CTX-M brackets (through transposition or recombination) throughout their evolution.The large divergence in plasmids carrying bla CTX-M genes is consistent with the notion that bla CTX-M genes were associated with different plasmids that existed before the use of third-generation cephalosporins (14) (Table 1).
Our findings are also consistent with recent reports showing that IS26 is extremely active, as transposable elements, mobilizing many important AMR genes (39).In our study, however, the IS26-bla CTX-M-55 brackets carried a fragment of the bla TEM gene (in addition to the bla CTX-M gene); seven of the nine plasmids carrying bla CTX-M-55 allelic variant showed the fosA3 gene (coding for fosfomycin resistance) in the vicinity, whereas other plasmids showed a fragment of the mef(B) gene (coding for a macrolide efflux pump) in the vicinity (Fig. 1).Similarly, IS26-bla CTX-M-65 brackets showed more AMR genes: fosA3, floR (coding for chloramphenicol), aph (4)-Ia (coding for an aminogly coside phosphotransferase), aac (3)-Iva (coding for gentamicin), and ant(3″)-Ia (coding for streptomycin) in the vicinity (Fig. 2).
Even though we were not able to observe direct transmission of a plasmid between E. coli from domestic animals and humans, all the plasmids carrying bla CTX-M genes were conjugable, and domestic animals and humans shared many (65%, 13 of 20) of the IS26-bla CTX-M brackets.These findings suggest that horizontal gene transfer events of diverse plasmids and bla CTX-M genes outnumber clonal transmission events (among domestic animals and humans), as we found 14% of bla CTX-M E. coli strains presented evidence of recent transmission between humans and domestic animals (18).These results suggest that the IS26-bla CTX-M bracket involves a complex multi-step process of horizontal gene transfer in which transposons mobilize the bla CTX-M among plasmids or from plasmids to chromosomes.These results agree with previous observations that some bla CTX-M gene variants (and their contiguous regions) were associated with specific environments in Ecuador (40).In these cases, the only evidence of AMR-gene transmis sion is the presence of the highly similar IS26-bla CTX-M brackets in different isolates in an epidemiological context compatible with this transmission.We acknowledge that because E. coli isolates were obtained from the same geographic region in Ecuador (18), the results may not be generalizable to other countries, although our IS26-bla CTX-M brackets were highly similar to the sequences found in plasmids and chromosomes from GenBank, suggesting that IS26-bla CTX-M brackets may play an important role in the dissemination of the bla CTX-M gene through several bacterial species in different geographic regions.
In conclusion, the prevalence of CTX-M enzymes has increased dramatically since the mid-to late 2000s (3).ESBL-encoding genes were identified in plasmids present in E. coli strains isolated before the use of third-generation cephalosporins in 1981, suggesting that the E. coli acquisition of these genes had occurred in multiple independ ent events (14).The IS26 transposable element is critical for the current mobility of these and other clinically crucial AMR genes.Our study suggests that the nucleotide sequences of IS26-bla CTX-M brackets could be an important genetic structure to study bla CTX-M transmission between humans, domestic animals, and the environment.We provide evidence for the complexity of the bla CTX-M horizontal gene transfer and how this understanding can be applied to determine the dissemination of these genes in any community, animals, or environment.The amplification and sequencing of the DNA inside the brackets may be used to monitor the bla CTX-M dynamics (increasing rates, allelic variant replacement, dissemination, etc.).Understanding the dissemination patterns of AMR genes is critical to implementing effective measures to slow down the dissemination of these clinically important genes.

FIG 1
FIG 1 Comparison of nine plasmids carrying bla CTX-M-55 gene variant of extended-spectrum β-lactamase-producing Escherichia coli isolates from children, chickens, and dogs.Labels show the plasmid ID assigned based on the host ID followed by its isolate number and length of the plasmid carrying the bla CTX-M-55 allelic variant.The origin of the isolate harboring the plasmid is shown by a figure in black (child, chicken, and dog).Each plasmid is represented by linear visualization, and coding sequences (CDSs) are represented by arrows.The direction of the arrow indicates the transcription direction of each CDS.CDSs are colored based on their functions.Blue and red shading areas between plasmids indicate the similarity of regions in the same and inverted directions, respectively, according to BLASTn.The percentage of sequence similarity is shown according to a color gradient.

FIG 2
FIG 2 Comparison of four chromosome fragments (100,000 pb) and five plasmids carrying bla CTX-M-65 gene variant of extended-spectrum β-lactamase-produc ing Escherichia coli isolates from children, chickens, and dogs.Labels show the plasmid and chromosome ID assigned based on the host ID followed by its isolate number and length of chromosome* and plasmid carrying bla CTX-M-65 allelic variant.The origin of the isolate harboring the plasmid is shown by a figure in black (child, chicken, and dog).Each chromosome fragment or plasmid is represented by linear visualization, and coding sequences (CDSs) are represented by arrows.The direction of the arrow indicates the transcription direction of each CDS.CDSs are colored based on their functions.Blue and red shading areas between sequences indicate the similarity of regions in the same and inverted directions, respectively, according to BLASTn.The percentage of sequence similarity is indicated according to a color gradient.

FIG 3
FIG 3 Comparison of two plasmids carrying bla CTX-M-27 gene variant of extended-spectrum β-lactamase-producing Escherichia coli isolates that were part of a clonal relationship with 0 SNPs in their core genomes.Labels show the plasmid ID assigned based on the host ID followed by its isolate number and length of the plasmid carrying the bla CTX-M-27 allelic variant.The origin of the isolate harboring the plasmid is shown by a figure in black (child and dog).Each plasmid is represented by linear visualization and coding sequences (CDSs) represented by arrows.The direction of the arrow indicates the transcription direction of each CDS.CDSs are colored based on their functions.Blue and red shading areas between plasmids indicate the similarity of regions in the same and inverted directions, respectively, according to BLASTn.The percentage of sequence similarity is indicated according to a color gradient.

FIG 4
FIG 4 Comparison of nine IS26-bla CTX-M-55 brackets of extended-spectrum β-lactamase-producing Escherichia coli isolates from children, chickens, and dogs.Labels show plasmid ID (harboring the IS26-bla CTX-M bracket) assigned based on the host ID followed by its isolate number and length of the IS26-bla CTX-M-55 bracket.The origin of the isolate harboring the plasmid is shown by a figure in black (child, chicken, and dog).Each IS26-bla CTX-M-55 bracket is represented by linear visualization, and coding sequences (CDSs) are represented by arrows.The direction of the arrow indicates the transcription direction of each CDS.CDSs are colored based on their functions.Blue shading areas between plasmids indicate the similarity of regions in the same direction according to BLASTn.The percentage of sequence similarity is indicated according to a color gradient.

FIG 5
FIG 5 Comparison of nine IS26-bla CTX-M-65 brackets of extended-spectrum β-lactamase-producing Escherichia coli isolates from children, chickens, and dogs.Labels show the plasmid or chromosome ID (harboring the IS26-bla CTX-M bracket) assigned based on the host ID followed by its isolate number and length of the IS26-bla CTX-M-65 bracket.The origin of the isolate harboring the plasmid or chromosome is shown by a figure in black (child, chicken, and dog).Each IS26-bla CTX-M-65 bracket is represented by linear visualization, and coding sequences (CDSs) are represented by arrows.The direction of the arrow indicates the transcription direction of each CDS.CDSs are colored based on their functions.Blue shading areas between plasmids indicate the similarity of regions in the same direction according to BLASTn.The percentage of sequence similarity is indicated according to a color gradient.

TABLE 1
Length, plasmid types, and origin of plasmids and chromosomes carrying bla CTX-M

genes Sequence ID Origin of E. coli isolate Allelic variant bla CTX-M Size (bp) Plasmid type
a Two copies of this plasmid type are carried by the plasmid.