Municipal wastewater monitoring revealed the predominance of bla GES genes with diverse variants among carbapenemase-producing organisms: high occurrence and persistence of Aeromonas caviae harboring the new bla GES variant bla GES-48

ABSTRACT We characterized carbapenemase-producing organism (CPO) detected in municipal wastewater to better understand the epidemiology of CPOs in the community. In total, 36 samples were collected at six sampling sites every other month from December 2020 to October 2021. CPOs were not recovered from influent taken from inlet A receiving separated sewer line, treated effluents, and river water samples upstream and downstream of the effluent outlet. By contrast, 75 CPOs were detected in all influent samples taken from inlets B and C receiving combined sewer lines collecting both domestic/industrial wastewater and rainwater runoff. Aeromonas caviae was the dominant species (25/75, 33.3%), and the other 11 Aeromonas spp. together accounted for 48% of CPOs. The remaining 39 Enterobacterales strains mainly comprised 17 Klebsiella spp. and 10 Raoultella spp. CPOs carrying bla GES carbapenemase genes were overwhelmingly dominant, accounting for 72 of 75 isolates, including two isolates harboring both bla GES-24 and bla IMP-1 (96%), followed by three bla IMPs-positive isolates, where those carbapenemase genes were mainly carried in diverse class 1 integrons. Among bla GES variants, including six new variants (bla GES-47, bla GES-48, bla GES-49, bla GES-50, bla GES-51, and bla GES-54), bla GES-5 was detected in 28 CPOs, with Aeromonas spp. accounting for 53.6% of these organisms. Quantitative analysis revealed that the repeated detection of bla GES-48-positive A. caviae ST1056 from both inlets B and C ranked the total number of this bacterial clone highest in the wastewater influent. In summary, our study revealed the high prevalence and persistence of diverse bla GES carbapenemase genes among CPOs isolated from influent inlets connected to combined sewer systems. IMPORTANCE The emergence and spread of carbapenemase-producing organisms (CPOs) represent a global health threat because they are associated with limited treatment options and poor clinical outcomes. Wastewater is considered a hotspot for the evolution and dissemination of antimicrobial resistance. Thus, analyses of municipal wastewater are critical for understanding the circulation of these CPOs and carbapenemase genes in local communities, which remains scarcely known in Japan. This study resulted in several key observations: (i) the vast majority of bla GES genes, including six new bla GES variants, and less frequent bla IMP genes were carbapenemase genes encountered exclusively in wastewater influent; (ii) the most dominant CPO species were Aeromonas spp., in which a remarkable diversity of new sequence types was observed; and (iii) CPOs were detected from combined sewer wastewater, but not from separate sewer wastewater, suggesting that the load of CPOs from unrecognized environmental sources could greatly contribute to their detection in influent wastewater.


IMPORTANCE
The emergence and spread of carbapenemase-producing organisms (CPOs) represent a global health threat because they are associated with limited treatment options and poor clinical outcomes.Wastewater is considered a hotspot for the evolution and dissemination of antimicrobial resistance.Thus, analyses of municipal wastewater are critical for understanding the circulation of these CPOs and carbapene mase genes in local communities, which remains scarcely known in Japan.This study resulted in several key observations: (i) the vast majority of bla GES genes, including six new bla GES variants, and less frequent bla IMP genes were carbapenemase genes encountered exclusively in wastewater influent; (ii) the most dominant CPO species were Aeromonas spp., in which a remarkable diversity of new sequence types was observed; and (iii) CPOs were detected from combined sewer wastewater, but not from separate sewer wastewater, suggesting that the load of CPOs from unrecognized environmental sources could greatly contribute to their detection in influent wastewater.environments (16).Nonetheless, outbreaks of GES-5-producing Pseudomonas aeruginosa and Serratia marcescens have been reported in Japan (17,18).In a very recent study, we found that bla GES-24 and bla GES-5 , which are mainly carried by Klebsiella and Enterobacter spp., are the only carbapenemase genes detected in wastewater effluent from a hospital with no history of detection of clinical GES carbapenemase-producing isolates (19).Notably, these bla GES genes were mostly embedded in new class 3 integrons, although they are generally associated with class 1 integrons in clinically relevant Enterobacterales (19,20).These findings prompted us to investigate the epidemiology of CPOs in local communities.
In this study, quantitative monitoring of CPOs in influent from a municipal wastewater treatment plant (WWTP) was conducted to investigate their presence, distribution, and dynamics.This WWTP receives sewer water in a combined sewer system that primarily collects domestic/industrial wastewater mixed with urban runoff rainwater, as well as in the separated sewer system that carries only domestic/industrial wastewater via different inlets.Therefore, the influent of these inlets was analyzed to assess the environmental impact of the CPO load from the combined sewer system.Additionally, wastewater effluent and river water samples (upstream and downstream of the effluent outlet) were also included in the analysis to better understand whether the effluent is a potential source of CPO release into the environment.
A total of 75 CPOs were selected as representative strains of bacterial populations, estimated as colony-forming units (CFU) per milliliter, sharing similar colony morpholo gies, the same bacterial species, and the same carbapenemase genes.Table 1 shows bacterial species based on average nucleotide identity based on MUMmer (ANIm) analyses, carbapenemase genes, sequence types (STs), and strain numbers for 75 CPOs identified in the raw influent of inlets B (28 isolates) and C (47 isolates).Aeromonas caviae was the dominant species (25/

bla GES-48 -positive A. caviae ST1056 occurred at the highest total densities in the influent
The abundance (CFU/mL) of CPOs measured bimonthly at inlets B and C is shown in Fig. 2. The bacterial counts of CPOs ranged from 1.0 × 10 −1 to 3.8 × 10 2 CFU/mL and from 5.0 × 10 −1 to 2.5 × 10 3 CFU/mL for the influent from inlets B and C, respectively, and significant differences in the overall CPO abundance were detected between samples from these two inlets (Wilcoxon signed-rank sum test, P = 0.00056).When examining seasonal differences in the abundance of CPOs, no significant differences were detected in samples at each inlet (Friedman test).On a total abundance level, A. caviae ST1056 lineage isolates harboring class 1 integron-associated bla GES-48 , which were detected in almost every sampling month at both inlets, represented the most abundant clone (Fig. 2).The occurrence of class 1 integron-associated bla GES-48 was observed among A. caviae ST1182, K. pneumoniae subsp.pneumoniae ST76, and K. quasipneumoniae subsp.similipneumoniae ST6102 lineage isolates in the last two sampling months (August and October 2021).Our conjugation experiments between A. caviae ST1056 or A. caviae ST1182 as donor strains and K. pneumoniae subsp.pneumoniae ATCC13883 (Rif r ) or E. coli χ1037 (Rif r ) as the recipient strain failed to obtain any transconjugants harboring bla GES-48 .
Figure 4 shows the network consisting of 78 nodes and 777 edges depicting the co-occurrence patterns among bla GES and bla IMP gene cassettes in class 1 and 3 integrons and bacterial lineages.The topological analysis revealed that the top 15 ranking genes (intI1, sul1, qacEΔ1, aacA4, bla GES-5 , aacA31, aadA1, bla GES-24 , bla GES-48 , bla GES-4 , aadA8, bla OXA-4 , catB10, intI3, and aadA2) with node degrees exceeding the average value and higher betweenness centrality and closeness centrality values (greater than the median of the nodes in the network) were considered hub nodes.Clustering analysis by MCODE identified three important cluster subnetworks of highly intraconnec ted nodes.The first cluster, consisting of 77 nodes and 774 edges with an MCODE score of 9.692, mainly depicted densely connected nodes of class 1 integron-associated genes (intI1, qacEΔ1, and sul1), carbapenemase genes, other resistance genes, and diverse host species and lineages.The second cluster (32 nodes and 99 edges with an MCODE score of 3.879) featured a significant relationship among carbapenemase genes (bla GES-24 , bla GES-48 , bla GES-54 , and bla IMP-1 ), several other resistance genes, and mainly Aeromonas and Enterobacter spp. as host species.The third cluster (11 nodes and 27 edges with an MCODE score of 3.333) mainly featured a relationship among bla GES-48 , aadA2, and Klebsiella spp.

DISCUSSION
This study revealed the occurrence and persistence of CPOs comprising a diverse range of bacterial species and lineages in wastewater influent from inlets B and C connected to combined sewer systems collecting both primarily domestic/industrial wastewater and urban runoff rainwater.CPOs carrying bla GES carbapenemase genes were overwhelm ingly dominant, accounting for 72 of 75 isolates (96%).Our finding is congruent with previous reports documenting the predominance of bla GES genes among CPOs in raw wastewater samples from WWTPs (15,22).However, our results are unique in that various bla GES variants, including six newly identified variants, were associated with CPOs, and Aeromonas spp.were the most dominant CPO species.In addition, CPOs were not detected in any raw influent from inlet A connected to the separated sewer systems carrying only domestic/industrial wastewater.Thus, the load of water washing the ground surface, including rainwater, contaminated with CPOs carrying bla GES carbapene mase genes from unrecognized environmental sources could contribute significantly to their detection in the wastewater influent from inlets B and C. In particular, in recent years, the volume of runoff has often exceeded the capacity of the combined sewer system due to localized heavy rainfall resulting from abnormal weather conditions.This can lead to the backflow of wastewater, resulting in the spread of CPOs throughout the entire system.However, one limitation of the study is the lack of data on this type of water monitoring.The bla GES carbapenemase genes (bla GES-5 , bla GES-6 , bla GES-16 , and bla GES-24 ) carried by diverse Enterobacterales and non-Enterobacterales species have been documented in aquatic environments, such as urban streams ( 23), river water intI1-bla GES-24 -aadA2-aacA31-qacEΔ1-sul1 (24,25), coastal water (26,27), urban ponds (28), lake water (29), wastewater (30), and hospital sewage (19,31,32).In this study, we could not identify potential sources of the bla GES carbapenemase genes.However, the possible presence of these CPOs of environmental origin is also supported by the findings that Aeromonas spp., which are ubiquitous in aquatic environments and soils and capable of rapidly colonizing various niches and hosts (33), permitting them to serve as potential environmental reservoir bacteria for carbapenemase genes, were the most predominant CPO species harboring bla GES carbapenemase genes.bla GES genes remain rare in clinical isolates, accounting for 2.1% (5/240 isolates) among isolates positive for carbapenemase genes according to the carbapenem-resistant Enterobacteriaceae surveillance in the Infectious Agents Surveil lance System, 2020, in Japan (available online at https://www.niid.go.jp/niid/ja/cre-m/ cre-iasrd/11520-511d01.html).Thus, the normal human microbiota is unlikely to be a source of the bla GES carbapenemase genes primarily detected in wastewater influent in our study.
In this study, we described the occurrence of multiple bla GES variants, with bla GES-5 being the most prevalent in 36 Aeromonas spp.strains mostly belonging to 19 newly identified STs and in 39 Enterobacterales strains including Enterobacter, Klebsiella, and Raoultella spp.Furthermore, six variants of bla GES carbapenemase genes, namely bla GES-47 , bla GES-48 , bla GES-49 , bla GES-50 , bla GES-51 , and bla GES-54 , were newly identified.Compared to the sequence of GES-1, GES-49 and GES-54 shared the amino acid substitution Glu104Lys; GES-47, GES-48, GES-50, and GES-54 shared the amino acid substitution Met62Thr; and an amino acid change at position 6 was shared by GES-49 and GES-51 (Ala6Val) and GES-50 (Ala6Thr), in addition to the Gly170Ser substitution identified in all six bla GES variants.The Gly170Ser substitution within the Ω-loop region displays carbapenem-hydrolyzing activity (34).This substitution has also been associated with increased catalytic efficiency against cephamycin (cefoxitin), whereas Glu104Lys was linked to enhanced hydrolytic activity toward oxyimino-cephalosporins (35).The Met62Thr substitution does not significantly affect the MICs of β-lactams for E. coli clones (36).The Ala6Thr substitution in the leader peptide was associated with higher MICs for β-lactams (cefotaxime, aztreonam, and imipenem) (37).The phenotypic effects of these resistance-related substitutions did not vary substantially, as all E. coli transformants gaining these new bla GES variants showed susceptibility or intermediate resistance to imipenem and meropenem with MIC ranges of 0.5-2 and 0.25-1 µg/mL, respectively.
Class 1 integrons, being capable of capturing AMR genes, play a particularly important role in the acquisition and dissemination of multidrug resistance among Gram-negative bacteria (38).bla GES and bla IMP genes are frequently associated with class 1 integrons in Enterobacterales, Pseudomonas, and Aeromonas spp. of clinical and aquatic environmental origin (17,20,30,39).These carbapenemase genes have been less commonly associated with class 3 integrons, but our recent study revealed the linkage of bla GES-24 , which was predominant among CPO isolates in hospital wastewater, with rare Tn402-like class 3 integrons (19).In the present study, bla GES and/or bla IMP genes carried by 29 Aeromonas spp.and 39 Enterobacterales strains were embedded in class 1 integrons, whereas bla GES genes (bla GES-5 , bla GES-6 , and bla GES-24 ) carried by the remaining seven Aeromonas spp.strains were mostly embedded in Tn402-like class 3 integrons.Although structural diversity of gene cassette arrays was noted in the class 1 integrons carried by these 68 CPOs, several shared gene cassettes were persistent in wastewater environment, including one between A. veronii bv.veronii ST1054 and A. caviae ST1178 strains, one among K. pneumoniae subsp.pneumoniae ST2791, R. planticola, and R. ornithinolytica, and one between E. cloacae subsp.cloacae ST1821 and E. roggenkampii ST1059 (R. ornithinolytica also carried an extremely similar one). A. caviae ST1015 and A. taiwanensis ST1055 strains shared the class 3 integron tniABQR-intI3-bla GES-5 -aacA4.Thus, these integrons, by association with transposons and plasmids, might contribute to the dissemination of bla GES carbapenemase genes through horizontal gene transfer events (40) within different species of Aeromonas or Enterobacter/Klebsiella/Raoultella.
Quantitative analysis of CPOs revealed that the repeated detection of bla GES-48 -posi tive A. caviae ST1056 from both inlets B and C ranked the total number of this bacterial clone highest in the wastewater influent, suggesting the constant presence of this clone in the combined sewer wastewater.Biofilm formation mediated by polar flagella, lateral flagella, T6SS, etc. might also allow this clone to become a resident organism in the sewer system.However, there might be a missing link to explain the occurrence of the bla GES-48 -positive A. caviae ST1056 strains and the bla GES-5 -positive A. caviae ST1059, A. caviae ST1178, and K. pneumoniae subsp.pneumoniae ST2791 strains in influent from both inlets B and C. Interestingly, eight of nine A. caviae ST1056 strains harbored identical or extremely similar class 1 integron cassettes around the bla GES-48 genes (intI1bla GES-48 -aacA31-aadA1-aadA1-qacEΔ1-sul1), which were different from the structure of class 1 integron with bla GES-48 carried by the remaining A. caviae ST1056 strain (intI1bla GES-48 -ISPa59-qacEΔ1-sul1) and by one each of K. pneumoniae subsp.pneumoniae ST76 and K. quasipneumoniae subsp.similipneumoniae ST6102 strains (intI1-bla GES-48bla OXA-129 -aadA2-qacEΔ1-sul1). bla GES-48 could not be transferred via conjugation between A. caviae ST1056 and K. pneumoniae subsp.pneumoniae ATCC13883 or E. coli χ1037; thus, bla GES-48 harbored by A. caviae ST1056 is likely associated with an Aeromo nas-specific plasmid.
In this study, a representative colony was selected from a minimum of three colonies with the same colony morphology, the same bacterial species, and the same carbapene mase genes.This selection process of representative colonies can result in the successful detection of different ST types within the same CPO species sharing the same carbape nemase genes and the successful tracking of the presence of CPOs belonging to multiple STs in wastewater influents.However, these representative colonies are not guaranteed to be an exact match to the corresponding bacterial populations in terms of bacterial species and genetic types, which may limit the quantitative detection of CPOs under the conditions studied.
In summary, our study revealed the high prevalence and persistence of bla GES carbapenemase genes among CPOs isolated from influent inlets connected to combined sewer systems.Aeromonas spp.carrying bla GES-5 , bla GES-6 , bla GES-24 , bla GES-48 , and bla GES-49 embedded in class 1 or 3 integrons stood out as the most abundant CPOs, followed by Klebsiella spp.carrying bla GES-4 , bla GES-5 , bla GES-47 , bla GES-48 , and bla GES-50 embedded in class 1 integrons.The presence of diverse integrons, bla GES variants, and host clones of Aeromonas spp., Enterobacter spp., Klebsiella spp., and Raoultella spp.represents the plasticity of these genetic elements, which might allow integrons to capture and disseminate such bla GES variants and facilitate the adaptability of CPOs to the environment.This study shed light on the great potential of the environment in holding bla GES carbapenemase genes and promoting their genetic variability.

Sample collection and bacterial isolation
One sample each of raw influent from three different inlets, namely, inlet A receiving separated sewer lines and inlets B and C receiving combined sewer lines, and treated effluent from the WWTP outlet were collected once every 2 months on rain-free days from December 2020 to October 2021 from the municipal WWTP located in Matsumoto City, Nagano Prefecture, Japan (Fig. S3).The WWTP processes approximately 82,200 m 3 of wastewater per day from 124,700 inhabitants, comprising 43,980, 24,540, and 56,180 inhabitants for inlets A, B, and C, respectively.Simultaneously, river water samples from sites 1.35-km upstream (36°13′54.7″N,137°57′11.0″E)and 0.56-km downstream (36°14′56.0″N,137°57′4.7″E)from the effluent outlet were collected (Table S1).Each sample (approximately 1,000 mL) was placed in cleaned polypropylene bottles and immediately transported in a cooler box with sufficient ice packs to our laboratory for processing within 2 h after obtaining the samples.A volume of 250 mL from the influent was concentrated 100-fold via centrifugation at 3,500 rpm for 20 min, after which the pellet was suspended in 2.5 mL of sterile phosphate-buffered saline (PBS, pH 7.4).For effluent and river water samples, 600 mL were filtered through a 0.45 µM membrane (HAWP04700; Millipore).Then, the filter was cut into small pieces and placed in 6 mL of sterile PBS to prepare a bacterial cell suspension.Finally, a 1,000× concentrated suspension was obtained via centrifugation at 15,000 rpm for 10 min.
The 100× concentrated influent samples were serially diluted using sterile PBS, and 100 µL of each dilution was spread onto MacConkey agar (Eiken Chemical Co., Tokyo, Japan) containing 1 µg/mL imipenem (Sigma-Aldrich Japan, Tokyo, Japan) and CHROMagar mSuperCARBA (Kanto Chemical, Tokyo, Japan), followed by incubation at 37°C overnight.In the same manner, 100 µL aliquots of 1,000× concentrated suspensions obtained from effluent and river water samples were spread directly on the same agar plates.
For each sample, bacterial colonies exhibiting similar morphological features were counted using only agar plates yielding approximately 300 visible colonies, enabling us to calculate the number of CFU per milliliter.Then, a minimum of three well-isolated colonies from the bacterial population with similar colony morphology were individually subcultured onto Mueller−Hinton agar (Eiken) and subjected to MALDI-TOF MS with the Bruker BioTyper database and software version 3.1 (Bruker Daltonics Japan, Yokohama, Japan) using a cutoff of ≥2.00 for species-level identification.Intrinsically carbapenemresistant bacterial species, including Morganella morganii, Providencia rettgeri, and Stenotrophomonas maltophilia, were excluded from further analysis after PCR confirmation of the absence of acquired carbapenemase genes.

Detection of carbapenemase genes
The carbapenemase genes bla IMP , bla NDM , bla KPC , bla GES , and bla OXA-48 were screened by PCR and identified by DNA sequencing (19).After confirming by MALDI-TOF MS that at least three colonies representing a population of bacteria with similar colony morphology were the same bacterial species and harbored the same carbapenemase genes, a representative strain was selected from these colonies.Plural representative isolates identified as K. pneumoniae sharing the same carbapenemase genes obtained from the same agar plate were further differentiated by analyzing rpoB sequences (41).

Antimicrobial susceptibility testing
The MICs of the carbapenemase-producing isolates were determined by the broth microdilution method recommended by the Clinical and Laboratory Standards Institute (CLSI) using Dry Plate DP41 (Eiken), and the results were interpreted using CLSI document M100-ED32 (42).The MICs of faropenem (Sigma-Aldrich), ertapenem (Fujifilm Wako Pure Chemical Co., Osaka, Japan), colistin (Fujifilm), and tigecycline (Tokyo Chemical Industry Co., Tokyo, Japan) were determined using in-house prepared panels according to the CLSI broth microdilution method.E. coli ATCC25922 was used as a quality control strain.

Transformation experiments
Transformation of NEB 10-beta electrocompetent E. coli (DH10B derivative, New England Biolabs, Tokyo, Japan) with plasmid DNA extracted from each of the carba penemase-producing isolates harboring new variants of the bla GES genes (bla GES-47 , bla GES-48 , bla GES-49 , bla GES-50 , bla GES-51 , and bla GES-54 ) was performed by electropora tion.Transformants were selected on LB agar plates containing ampicillin (50 µg/mL), and the presence of bla GES genes and the absence of other β-lactamase genes were confirmed by PCR and sequencing.The MICs for the transformants were determined in the same manner described above.

Transferability of bla GES-48 from A. caviae to Enterobacterales
The broth mating assay was used to investigate the transferability of bla GES-48 from the parental isolates, A. caviae ST1056 strain 2BC3 and A. caviae ST1182 strain 8CC4 to K. pneumoniae subsp.pneumoniae ATCC13883 (Rif r ) or E. coli χ1037 (Rif r ).Conjugation was performed at 25°C and 37°C, and transconjugants were selected on rifampicin-contain ing (100 µg/mL) LB agar plates supplemented with faropenem (16 µg/mL) and ampicillin (50 µg/mL) for the former and latter recipients, respectively.

WGS and bioinformatics
Genomic DNA was extracted using lysozyme and proteinase K and purified using an Agencourt AMpure XP kit (Beckman-Coulter Life Sciences, Brea, CA, USA) according to the manufacturer's instructions.To construct DNA libraries, 5× whole-genome sequenc ing (WGS) Fragmentation Mix and 5× WGS Ligase Mix (Enzymatics, Beverly, MA, USA) were used.Pooled libraries were subjected to 350-to 800-bp size selection using the BluePippin system (Sage Science, Inc., Beverly, MA, USA).The pooled libraries were sequenced on the Illumina HiSeq X Five platform (Illumina Inc., San Diego, CA, USA) using the 150-bp paired-end method.A small portion of the CPOs were subjected to WGS by the 150-bp paired-end method using the NovaSeq6000, MiSeq, and MiniSeq platforms (Illumina).The de novo assembly of short reads was conducted using Shovill v1.1.0(https://github.com/tseemann/shovill).
Genome annotation was achieved using Prokka 1.14.6 (https://github.com/tseemann/prokka).Whole-genome clustering based on k-mer mash distances among genomes was conducted using PopPUNK (43) on a Galaxy ARIES-based platform (Galaxy Version 1.1).iTOL v6 (https://itol.embl.de/)was used to annotate and visualize the tree.The genetic context of the carbapenemase gene was investigated by PCR mapping and sequencing.The co-occurrence relationships among integrons, resistance genes, and bacterial strains were visualized in a network using Cytoscape version 3.9.1 (44).The Cytoscape plugin cytoHubba (v0.1) (45) and MCODE (v2.0.2) (46) were used to predict hub nodes based on topological parameters and the identification of highly intraconnected clusters in a network, respectively.New STs were assigned by the Institut Pasteur MLST database or PubMLST database.New allele numbers for bla GES carbapenemase genes were assigned by NCBI.

4 TABLE 1 a
Seventy-five strains harboring carbapenemase genes identified in raw influents of inlets B and C a GES variants newly identified in this study.b ST, sequence type.c ST newly assigned in this study.

FIG 1 7 TABLE 2
FIG 1 Phylogenetic relationship of 75 CPOs.Core genome neighbour-joining tree (left) was generated by PopPUNK.Strain numbers and their STs are indicated at the right of the tree, with new STs marked in red.The presence (filled squares) or absence (open squares) of antimicrobial resistance-associated genes and plasmid Inc groups among strains is shown.

FIG 2
FIG 2 Babble plots showing the abundance (CFU/mL) of CPOs detected in raw influents at inlets B and C. The CFU/mL values were obtained by quantifying organisms of the same bacterial species belonging to the same STs and harboring the same carbapenemase genes.

FIG 4
FIG4 Network analysis depicting co-occurrence relationship among bla GESs and bla IMPs (pink nodes), class 1 and 3 integron-integrase genes, intI1 and intI3 (green nodes), gene cassette composition in these integrons (beige nodes), and species and lineages of Enterobacterales (orange nodes) and the genus Aeromonas (blue nodes).

TABLE 1
Seventy-five strains harboring carbapenemase genes identified in raw influents of inlets B and C a

TABLE 3
Genetic contexts of bla GESs and bla IMPs in carbapenemase-producing organisms

TABLE 3
Genetic contexts of bla GESs and bla IMPs in carbapenemase-producing organisms (Continued)