Molecular Epidemiological Characteristics of Carbapenem Resistant Aeromonas from Hospital Wastewater

Abstract

Background

Hospital wastewater (HWW) promotes the spread of carbapenem resistance genes (CRGs). Aeromonas carry a large number of CRGs in HWW, they may play a role as a suitable reservoir for CRGs, while resistomes in HWW are still poorly characterized regarding carbapenem resistant Aeromonas. Thus, the aim of the study was to evaluate the molecular epidemiological characteristics of carbapenem resistant Aeromonas in HWW.

Methods

A total of 33 carbapenem resistant Aeromonas were isolated from HWW. Antimicrobial susceptibility testing and polymerase chain reaction (PCR) were used to assess the antimicrobial resistance profiles. Molecular typing was performed using enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) and multilocus sequence typing (MLST). The horizontal transmission mode of bla_KPC was explored through conjugation and transformation experiments. The stability of bla_KPC-IncP-6 plasmids was assessed through plasmid stability and in vitro competition test. The PCR mapping method was used to investigate the structural diversity of bla_KPC.

Results

The detection rates of bla_KPC and cphA in Aeromonas were 97.0% and 39.4% respectively. Aeromonas caviae were grouped into 13 clusters by ERIC-PCR and 12 STs by MLST. Aeromonas veronii were grouped into 11 clusters by ERIC-PCR and 4 STs by MLST. 56.3% bla_KPC were located on mobilizable IncP-6 plasmids. bla_KPC-IncP-6 plasmid showed high stability and low cost fitness.

Conclusion

Carbapenem resistant Aeromonas from HWW mainly carried bla_KPC, which exhibited great structural diversity. Aeromonas might serve as reservoirs for bla_KPC and bla_KPC might spread mainly through transformation in HWW.

Keywords:

• hospital wastewater
• carbapenem resistance
• Aeromonas
• IncP-6
• bla_KPC

Introduction

Carbapenems were generally considered as last resort for treating infections from multi-resistant Gram-negative pathogenic bacteria.¹ But the widespread prevalence of carbapenem resistant bacteria (CRBs) has seriously limited the effectiveness of carbapenems. Carrying carbapenem resistance genes (CRGs) is an important mechanism of carbapenem resistance. CRGs are generally found on mobile plasmids, exhibiting strong dissemination capabilities.² Accordingly, effective control of CRG dissemination is critical to slow the spread of resistance.

Hospital wastewater (HWW) is considered significant reservoirs of CRGs. HWW connects human activities with the ecological environment, promoting the spread of clinically important CRGs such as bla_NDM, bla_KPC, and bla_VIM from clinical bacteria to environmental bacteria.³ The slow flow rate of HWW and high-density bacterial communities promote the horizontal transfer of resistance genes within and between bacterial species. Meanwhile, a large amount of antimicrobial residues promotes the proliferation of antimicrobial resistant bacteria.³ In addition to providing selective advantages, antimicrobial residues may induce the SOS-response and hence upregulate expression of conjugation associated genes and T4SS genes to enhance conjugative transfer.⁴ Furthermore, antimicrobial residues could also promote gene transformation and transduction by increasing the fraction of competent cells and inducing prophage activity.⁴ Hospital wastewater treatment plants could not eliminate antimicrobial resistant genes (ARGs) in wastewater and could even promote conjugations and transformations of ARGs.^5–7 Therefore, compared with other wastewater conditions, HWW may play a more integral role in the dissemination of CRGs.

Aeromonas is a widespread organism in aquatic environment and carry large numbers of CRGs.⁸ Previous studies showed that cphA is prevalent in carbapenem resistant Aeromonas. Notably, more and more CRGs have been detected in Aeromonas, such as bla_KPC or bla_NDM.^9,10 Especially, plasmid-associated bla_KPC-2 is becoming the most prevalent CRG in Aeromonas from wastewater. Given that cphA is located on the chromosome, the spread of plasmid-associated bla_KPC-2 is even more concerning.^11,12 Bacteria of the Aeromonas genus only has a 4–5 Mbp genome, which is smaller than that of other Gram-negative bacteria. The core genome of Aeromonas comprises only 16% of the entire genome, with the remaining 84% consisting of variable represented genes, including plasmids, transposons, genomic islands or integron gene cassettes in order to compensate for its relatively small genome size. Thus, Aeromonas has a significant capacity to acquire various mobile genetic elements.^13,14 Considering the strong ability of Aeromonas to take-up exogenous DNA, it might play a significant role in the transmission of CRGs.

To date, the wastewater resistomes have been poorly characterized regarding carbapenem resistant Aeromonas. A previous study mainly focused on prevalence and antimicrobial resistance characteristics of carbapenem resistant Aeromonas in HWW.¹⁵ The transmission characteristics of CRGs from carbapenem resistant Aeromonas remain relatively unexplored. Therefore, we deciphered the molecular epidemiological characteristics of carbapenem resistant Aeromonas and the potential transmission mechanism of bla_KPC in Aeromonas, helping to gain a more in-depth insight into the contribution of Aeromonas to CRGs dissemination and enhance the public consciousness about resistance-related risks.

Materials and Method

Aeromonas Isolates

The HWW samples were collected from the raw sewage influent of on-site HWW treatment plant in Fujian medical university union hospital (Fujian, China). A total of three HWW samples were taken from November 3 to November 25, 2020. Each HWW sample was independent and was cultured individually. The HWW samples were diluted and then plated on LB agar (Hopebio, China) containing Imipenem (IPM) at 2 μg/mL to select imipenem-non-susceptible Gram-negative bacteria at 35 °C. All imipenem-non-susceptible Gram-negative bacteria were cultured separately and identified through MALDI-TOF MS (Autobio, China). All isolates, identified as Aeromonas, were then reconfirmed by housekeeping gene sequencing (gyrB).

Antimicrobial Susceptibility Testing

Antimicrobial susceptibility testing was performed using the VITEK 2 Compact system (Vitek-AST-GN16) (bioMerieux, France) and the disk diffusion method (K-B method) (Oxoid, UK) according to the Clinical and Laboratory Standards Institute File.¹⁶ The antimicrobials included Piperacillin/Tazobactam (TZP), Cefoxitin (FOX), Ceftriaxone (CRO), Cefepime (FEP), Aztreonam (ATM), Ertapenem (ETP), IPM, Amikacin (AN), Gentamicin (CN), Ciprofloxacin (CIP), Levofloxacin (LVX) and Trimethoprim/Sulfamethoxazole (SXT). E.coli ATCC 25922 was used as a quality control.

Detection of Carbapenemase Genes

All isolates were subjected to a polymerase chain reaction (PCR) for confirmation of carbapenemase genes, including bla_KPC, bla_NDM, bla_VIM, bla_IMP, bla_OXA-48, cphA, bla_SME, bla_GIM, bla_IMI and bla_GES. The sequences and product lengths of primers are summarized in Table S1. The positive products were confirmed by sequencing (Sangon Biotech, China).

Enterobacterial Repetitive Intergenic Consensus PCR (ERIC-PCR)

The sequences of primers and PCR procedures were described previously.¹⁷ The results of ERIC-PCR were analyzed using Quantity One software (version 4.6.2). The phylogenetic tree was determined using the unweighted pair-group method with arithmetic means (UPGMA) clustering method and Dice similarity.

Multilocus Sequence Typing (MLST)

Six housekeeping genes (gyrB, groL, gltA, metG, ppsA, recA) were amplified using PCR. The sequences of primers and the PCR procedures were described previously.¹⁸ The products were sent to Sangon Biotech for DNA sequencing, and the results were compared with known alleles in the MLST database. The new STs were deposited in the Aeromonas PubMLST database (http://pubmlst.org/aeromonas). The phylogenetic trees were inferred using the UPGMA clustering method with MEGA11 software.

Conjugation

Conjugation experiments were performed to test the transferability of bla_KPC using filter mating. Rifampin-resistant E.coli EC600 was used as the recipient. Donors and recipients were cultured to log-phase in LB broth (Hopebio, China), respectively. Donors and recipients were mixed in an initial ratio of 1:4 and then incubated on a 0.22 μM filter, which was placed on a Columbia blood agar plate, and left overnight at 35 °C. The next day, the mating mixture was washed from the filter and spread onto LB agar containing Rifampin at 500 μg/mL and IPM at 1 μg/mL. The transconjugants harboring bla_KPC were identified by PCR.

Transformation and Plasmid Replicon Types Analysis

Plasmid DNA was isolated using the alkaline lysis extraction procedure and introduced to competent E.coli DH5α (Weidibio, China) through chemical transformation. The transformants were selected on LB agar supplemented with IPM at 1 μg/mL. The transformants harboring bla_KPC were identified by PCR. Plasmids were classified according to their incompatibility group using the PCR-based replicon typing scheme (PBRT)¹⁹ and the sequences of IncP-6 primers are listed in Table S2.

Plasmid Stability Assay and Fitness Assessment

The stability of plasmids was assessed as previously described with minor modifications.²⁰ Briefly, bla_KPC-carrying E.coli DH5α was inoculated into LB broth. Subcultures were diluted daily, and each dilution was plated on LB agar every 3 days. The plates were then incubated overnight at 35 °C. A total of 50 colonies were randomly collected from the LB agar and the presence of bla_KPC was identified using PCR.

The fitness costs of plasmids were evaluated through in vitro growth competition experiment. Plasmid-carrying E. coli DH5α (IPM-R) and plasmid-free E. coli DH5α (IPM-S) were cultured in an initial ratio of 1:1. The subcultures were then transferred to fresh LB broth. After 1 day, cultures were inoculated overnight on LB agar. Then colonies were collected from the LB agar and the presence of bla_KPC was identified by PCR. The fitness costs were calculated using the following equation:²¹

where Fit represents the fitness cost of the bla_KPC-carrying plasmid in the host E. coli DH5α; R denotes number of the bla_KPC-carrying E. coli DH5α; S represents the number of bla_KPC-free E. coli DH5α; R₀ indicates the initial number of bla_KPC-carrying E. coli DH5α in the competition experiments, and S₀ represents the initial number of bla_KPC-free E.coli DH5α in the competition experiments.

bla_KPC Genetic Region Analysis

The sequences and lengths of the primer products are summarized in Table S3. These primers were designed based on the sequence of the template plasmid p10265-KPC.²² Plasmid DNA was extracted using the GeneJET Kit (Thermo Scientific, US). PCR products were purified using the FastPure Gel DNA Extraction Mini Kit (Vazyme, China). Then, the purified DNA samples were digested using different enzymes. Fragment I, II and III were digested by HindIII (NEB, England). Fragment IV and V were digested by BamHI (NEB, England). Fragment VI was digested by HinfI (Takara, Japan). The grouping was based on the electrophoresis results of the six fragments’ enzymatic digestion. The PCR products were sent to Sangon Biotech for Sanger sequencing, and the fragments were assembled using the CAP3 program in SnapGene software (version 5.2). One isolate from each type was sequenced and annotated using the RAST (Rapid Annotation using Subsystems Technology) annotation website to determine the bla_KPC region.

Statistical Analysis

GraphPad Prism software (version 8) was used for statistical analysis. Fisher’s exact test was performed to analyze statistical significance. Only p < 0.05 was considered statistically significant.

Results

Prevalence of Carbapenem-Resistant Aeromonas

16 Aeromonas caviae and 17 Aeromonas veronii isolates from HWW were collected in this study. The percentage of carbapenem resistant Aeromonas was 26.83%. The percentage of other carbapenem resistant organisms detected have been listed in Table S4.

Antimicrobial Susceptibility Testing

16 Aeromonas caviae and 17 Aeromonas veronii isolates from HWW were collected in this study. Aeromonas caviae demonstrated high resistance to FOX (87.5%), CRO (100.0%), FEP (100.0%), ATM (100.0%), ETP (100.0%) and IPM (100.0%). The resistance to TZP (37.5%), AN (25.0%) and LVX (12.5%) was less than 40.0% in Aeromonas caviae. Meanwhile, Aeromonas veronii exhibited high resistance to CRO (100.0%), FEP (94.1%), ATM (88.2%), ETP (100.0%), IPM (94.1%), while showing low resistance to TZP (35.3%), AN (35.3%), CIP (29.4%) and SXT (35.3%). The Results are listed in Table 1.

Table 1 Antimicrobial Susceptibility Analysis of Aeromonas caviae and Aeromonas veronii

Drugs	Aeromonas caviae (n=16)			Aeromonas veronii (n=17)			p-value
	Susceptible	Intermediate	Resistant	Susceptible	Intermediate	Resistant
TZP	-	62.5%	37.5%	5.9%	58.8%	35.3%	>0.05
FOX	12.5%	-	87.5%	5.9%	29.4%	64.7%	>0.05
CRO	-	-	100.0%	-	-	100.0%	-
FEP	-	-	100.0%	5.9%	-	94.1%	>0.05
ATM	-	-	100.0%	11.8%	-	88.2%	>0.05
ETP	-	-	100.0%	-	-	100.0%	-
IPM	-	-	100.0%	5.9%	-	94.1%	>0.05
AN	68.7%	6.3%	25.0%	41.2%	23.5%	35.3%	>0.05
CN	56.2%	-	43.8%	23.5%	17.6%	58.8%	>0.05
CIP	12.5%	31.3%	56.3%	47.1%	23.5%	29.4%	>0.05
LVX	56.2%	31.3%	12.5%	88.2%	11.8%	-	>0.05
SXT	43.7%	-	56.3%	64.7%	-	35.3%	>0.05

Carbapenemase Genes Profiles

All 33 Aeromonas isolates in our study contained carbapenemase genes which included bla_KPC and cphA (Table 2). 16 Aeromonas caviae isolates carried bla_KPC-2. Among the 17 Aeromonas veronii isolates, 1 isolate carried cphA, 4 isolates carried bla_KPC-2, 4 isolates carried both cphA and bla_KPC-24, 8 isolates carried cphA and bla_KPC-2 simultaneously.

Table 2 The Frequency of Carbapenemase Genes Among Aeromonas caviae and Aeromonas veronii

Genes	Total population; n(%)	Aeromonas caviae; n(%)	Aeromonas veronii n(%)	p
blaKPC	32 (97.0)	16 (100.0)	16 (94.1)	>0.05
cphA	13 (39.4)	–	13 (76.5)	<0.05

ERIC-PCR

The fingerprints of Aeromonas isolates consisted of various amplification bands, ranging in size from 100 to 5000 bp. The similarity coefficient of Aeromonas caviae ranged from 0.34 to 0.82 and that of Aeromonas veronii ranged from 0.19 to 0.94 (Figure 1). When the similarity coefficient was 0.70, Aeromonas caviae were divided into 13 clusters and Aeromonas veronii were divided into 11 clusters. The results of ERIC-PCR in this study divided Aeromonas into various clusters, confirming the high genetic diversity in Aeromonas.

Figure 1 ERIC-PCR analysis of Aeromonas caviae (a) and Aeromonas veronii (b).

MLST

16 Aeromonas caviae isolates from HWW showed a high level of genetic diversity. 12 STs were identified in our study, including ST2170, ST2171, ST2379, ST2172, ST2173, ST2380, ST2174, ST2178, ST2179, ST2180, ST2181 and ST747. Additionally, 11 STs were novel founded (except ST747). ST2174 (3/16, 18.8%) was the most prevalent in Aeromonas caviae. Among 17 Aeromonas veronii isolates, 4 novel STs were identified, including ST2381, ST2175, ST2176 and ST2177. ST2176 (10/17, 58.8%) was the most prevalent ST in Aeromonas veronii. The evolutionary trees (Figure 2) could more directly illustrate the diversity among Aeromonas caviae and Aeromonas veronii. Notably, with the identification of 15 novel STs, this study also importantly contributed to the diversity in the MLST database of Aeromonas.

Figure 2 Phylogenetic tree analysis of Aeromonas caviae (a) and Aeromonas veronii (b).

Conjugation, Transformation and Plasmid Replicon Types Analysis

None of 32 bla_KPC-positive Aeromonas isolates were able to transfer bla_KPC into E.coli EC600 successfully. A total of 18 bla_KPC-positive Aeromonas (56.3%) isolates could transfer bla_KPC into E.coli DH5α through chemical transformation successfully, which included 5 Aeromonas caviae and 13 Aeromonas veronii isolates. All of the replicon types of mobilizable bla_KPC-carrying plasmids were IncP-6.

Plasmid Stability Assay and Fitness Assessment

The plasmid stability assay showed bla_KPC-IncP-6 plasmids could be stably maintained in E.coli DH5α for 30 days without any loss. But the bla_KPC-IncFII plasmids were not as perfect as bla_KPC-IncP-6 plasmids, bla_KPC-IncFII-31 began to lose stability from day 12 and plasmid bla_KPC-IncFII-1008 began to lose stability from day 18. However, plasmid bla_KPC-IncFII still showed a high level of stability, the rate of bla_KPC remaining above 96.0%. The results of plasmid stability were shown in Figure 3.

Figure 3 Results of plasmid stability for bla_KPC-IncP-6 and bla_KPC-IncFII in E.coli DH5α.

The fitness assessment was shown by in vitro growth competition experiment. The relative fitness costs of bla_KPC-IncP-6 plasmids were 1.024, 1.018 and 0.981. The relative fitness costs of bla_KPC-IncFII plasmids were 0.962, 1.016 and 1.039. There was 1 isolate (33.3%) with IncP-6 plasmid and 1 isolate (33.3%) with IncFII plasmid that increased the fitness cost of the host in our study. The results of fitness assessment were listed in Table 3.

Table 3 Different Types of bla_KPC-Carrying Plasmids and Their Fitness Cost Results

Plasmid	Inc group	Host	Relative Adaptive Fitness
p3a64	IncP-6	E.coli DH5α	1.024
p15a82	IncP-6	E.coli DH5α	1.018
p25a85	IncP-6	E.coli DH5α	0.981
p12	IncFII	E.coli DH5α	0.962
p31	IncFII	E.coli DH5α	1.016
p1008	IncFII	E.coli DH5α	1.039

bla_KPC Genetic Region Analysis

In template plasmid p10265-KPC, from the genetic content we found that it had undergone two main evolutionary events from ΔISKpn6-bla_KPC-2-ISKpn27: (i) the insertion of Δbla_TEM−1 between ISKpn27 and Tn3, and (ii) the insertion of ISApu1-ORF-ISApu2 into the Tn3 tnpA gene, disrupting the fragment into two separate parts.²² The genetic region of bla_KPC might undergoing the events of gene acquisition and lost in our study. Figure 4 showed these types of bla_KPC genetic regions.

Figure 4 bla_KPC regions from IncP-6 plasmids and comparison with the related plasmids.

Type A1 has the same genetic region with the reference sequence p10265-KPC, as ΔISKpn6-bla_KPC-2-Δbla_TEM-1-ISKpn27-tnpR-ΔtnpA-ISApu2-ORF-ISApu1-ΔtnpA. Among type A2, there was a transposase element ISKmi1, which was inserted between the ORF and ISApu1. There was a transposase element ISAs3 inserted between ISApu1 and ΔtnpA in type A3. The loss of Δbla_TEM-1 was found in type A4 and A5. Deletion of downstream fragments of tnpR occurred in type B. Type C received bla_KPC-24 instead of bla_KPC-2 and experienced loss of the downstream fragments of ΔtnpA.

Discussion

Aeromonas is a widespread organism in aquatic environment and carry large numbers of CRGs, it might play a significant role in the transmission of CRGs. Previous studies showed that cphA and bla_KPC are prevalent in carbapenem resistant Aeromonas.^15,23 Notably, the most prevalent carbapenemase gene in Aeromonas from wastewater was the plasmid-associated bla_KPC.¹⁵ In this study, 97.0% Aeromonas isolates carried bla_KPC indicating the potential transmission risk of bla_KPC. The high prevalence of cphA and bla_KPC suggests that carbapenem resistant Aeromonas may play a role in the transmission of carbapenem resistance. Although Aeromonas were not the predominant species, the high prevalence of CRGs and strong capacity to acquire various mobile genetic elements demonstrate the contribution of carbapenem resistant Aeromonas to CRGs’ dissemination. Given that cphA is located on bacterial chromosomes, plasmid-associated bla_KPC-2 might exhibit stronger dissemination capabilities.²⁴ The high prevalence and transmission risk of bla_KPC indicate that Aeromonas might serve as a potential reservoir for bla_KPC transmission.

In this study, we found that bla_KPC-carrying plasmids cannot be transferred through conjugation in Aeromonas, but can be transferred through transformation. All of the replicon types of mobilizable bla_KPC-carrying plasmids were IncP-6. The failed conjugation might be attributed to the lack of necessary genetic elements for conjugation in IncP-6 plasmids (such as the traA and traB operons).²⁵ Meanwhile, Aeromonas is equipped with the type VI secretion system which might be implicated in transformation.¹³ Previous studies found that bla_KPC cannot be transferred through conjugation in Aeromonas with or without the presence of self-transmissible plasmid.^25,26 Therefore, transformation might be an important mechanism for the transmission of bla_KPC in Aeromonas.

To further investigate the transmission risk of bla_KPC, we analyzed the stability of bla_KPC-IncP-6 plasmids by plasmid stability and in vitro competition test. The bla_KPC-IncP-6 plasmids had a strong stability and low fitness cost in this study. The reason for the high stability of bla_KPC-IncP-6 plasmids might be due to them carrying some genetic elements (such as the ParA/B/C operon), which can maintain their existence in the host.²⁰ A previous study found that bla_KPC-IncP-6 plasmids were broad host range plasmids and present in diverse environmental bacterial species of opportunistic, indicating that it might be a suitable vector of bla_KPC in HWW.⁹ The plasmid persistence depends on plasmid stability and their effect on the host fitness.²⁷ The high stability of IncP-6 plasmids is beneficial to the persistence of plasmids in host and the low fitness cost of IncP-6 plasmids could be an important advantage of carbapenem-resistant bacteria propagation. Therefore, the high stability of IncP-6 plasmids might promote the transmission of CRGs.

The wide spread of CRGs has posed a major public health threat. Previous research has suggested some solutions to control the spread of carbapenem resistance in clinic, such as antimicrobial restriction systems.²⁸ However, HWW, an important source of CRGs, had received increasing attention. HWW connects human activities with the ecological environment, promoting the spread of clinically important CRGs from clinical bacteria to environmental bacteria.³ Compared to other wastewater environments, HWW may have a stronger promoting effect on ARGs’ dispersion. Therefore, more emphasis should be placed on the contribution of HWW on the CRGs’ dissemination and an efficient method of controlling the CRGs’ dissemination is also urgently needed to relieve the current crisis of antimicrobial resistance.

There is a limitation to our study: whole genome sequencing was not applied in our study. Therefore, we have no idea about the genetic environment and molecular evolutionary relationships of carbapenem resistant Aeromonas in HWW, which might limit the exploration of contribution of Aeromonas in transmission of bla_KPC. In our further study, the phylogenomic analysis of carbapenem resistant Aeromonas will be further explored. Despite these limitations, this study deciphered the molecular epidemiological characteristics of carbapenem resistant Aeromonas and the potential transmission mechanism of bla_KPC in Aeromonas, helping to gain a more in-depth insight into the contribution of Aeromonas to CRGs’ dissemination and enhance public awareness about resistance-related risks.

Conclusions

Carbapenem resistant Aeromonas in HWW showed high genetic diversity. Aeromonas mainly carried bla_KPC, which exhibited structural diversity. Transformation might be an important mechanism for the transmission of bla_KPC in Aeromonas. Aeromonas might serve as reservoirs for bla_KPC and promote its spread in HWW.

Author Contributions

All authors contributed to data analysis, drafting or revising the article, have agreed on the journal to which the article will be submitted, gave final approval of the version to be published, and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Data Sharing Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Additional information

Funding

This work was supported by Fujian Research and Training Grants for Young and Middle-aged Leaders in Healthcare.