Prevalence and Abundance of Beta-Lactam Resistance Genes in Hospital Wastewater and Enterobacterales Wastewater Isolates

Antimicrobial resistance may develop in nature including in hospital wastewater through horizontal genetic transfer. Few studies were conducted on the antimicrobial resistance genes in hospital wastewater and wastewater isolates in Indonesia. The prevalence and abundance of beta-lactam resistance genes in hospital wastewater and Enterobacterales wastewater isolates were investigated. Twelve wastewater samples were collected from an influent wastewater treatment plant. Escherichia coli and Klebsiella pneumoniae were isolated from the wastewater samples by culture-based methods. DNA was extracted from wastewater samples and the isolates. Nineteen beta-lactam resistance genes were tested by a high throughput qRT-PCR method. blaGES and blaTEM were the most abundant genes detected in hospital wastewater and Escherichia coli, respectively (p < 0.001). The relative abundance of blaCMY_2, blaCTX-M5, blaCTX-M8, blaGES, blaNDM, and blaSHV11 in Klebsiella pneumoniae was higher than in the wastewater and Escherichia coli (p < 0.001; p = 0.006; p = 0.012; p < 0.001; p = 0.005; p < 0.001). Klebsiella pneumoniae might be associated with resistance to piperacillin/tazobactam, ceftriaxone, and cefepime (p < 0.001; p = 0.001; p < 0.001). In conclusion, ESBL genes showed higher abundance than carbapenemase genes in hospital wastewater samples. The ESBL-producing bacteria that were predominantly found in hospital wastewater may originate from clinical specimens. The culture-independent antibiotic resistance monitoring system might be developed as an early warning system for the increasing beta-lactam resistance level in clinical settings.


Introduction
Antimicrobial resistance is an important public health problem worldwide issued by the World Health Organization [1]. The high prevalence of antimicrobial resistance leads to the high mortality and healthcare cost either in developed countries or low-middle income countries [2,3]. Previous studies reported that more than 2.8 million antimicrobial-resistant infections and 29,000 mortality cases occur each year in the United States [2,4].
Antibiotic resistance is high among Gram negative Enterobacterales nosocomial pathogens Escherichia coli and Klebsiella pneumoniae [5]. A national surveillance on antimicrobial resistance in Indonesia reported that the third generation cephalosporins resistant-Escherichia coli and Klebisella pneumoniae were the most WHO priority pathogens encountered in clinical specimens [6]. It might be associated with the high use of ceftriaxone as the empirical antibiotic therapy in Indonesian hospitals [7,8].
Beta-lactam, as broad-spectrum antibiotics, are the most commonly prescribed antibiotics in clinical settings. Thus, the extensive use of these antibiotics applies selective pressure towards human microbiota and pathogens, increasing the risk of developing resistant strains and limitation of antibiotic therapy [9].
There are several mechanisms of beta-lactam resistance such as inactivation by betalactamase production, decreased penetration to target sites, changes in penicillin-binding protein of target sites, and drug efflux through specific pumping mechanisms [10]. The beta-lactamase enzymes are encoded by several resistance genes both chromosomal DNA and plasmid DNA [11]. Spontaneous mutation and horizontal gene transfer play a role in inducing beta-lactamase production on chromosomal DNA and plasmid DNA, respectively [11].
Horizontal gene transfer greatly contributes to the rapid spread of antibiotic resistance [12]. The mechanism of horizontal gene transfer provides a wide range of opportunities for bacteria that exist in the same ecosystem to express a resistance toward a certain group of antibiotics. This rapid spread of antibiotic resistant genes allows much larger epidemic of antibiotic resistant bacteria to exist in future settings [13]. The activity of such gene transfer remains unrecognized in low-and middle-income countries allowing the development of antibiotic resistant bacteria in those ecosystems yet. Nevertheless, the importance of such detection could not be underpinned, making ways of detecting such gene transfer activity in certain high-risk environment with various easier detection methods is of topmost importance [14].
Hospital wastewater is regarded as a hotspot for antimicrobial resistance, allowing antibiotic resistance genes to be transferred horizontally between pathogens and commensal bacteria [5,15,16]. The abundance of ARGs in pathogens, particularly those causing healthcare associated infections, such as Enterobacterales isolated from hospital wastewater is unclear. Escherichia coli and Klebsiella pneumoniae are major nosocomial pathogens that deserve to be tested for its abundance of antibiotic resistance genes.
To present the aforementioned data regarding the current conditions on horizontal gene transfer on high-risk environment settings such as hospital wastewater would prove to provide a great benefit to monitor the possibility of potential outbreak. Especially in developing countries, the availability of this particular data would increase the awareness of in-depth monitoring on antibiotic resistance activity.
Majlander J. et al., 2021, reported a novel wastewater-based monitoring of antibiotic resistance including beta-lactam antibiotic resistance, which allows comparison of betalactam antibiotic resistance profile over time. In addition, the correlation between bla KPC and Klebsiella pneumoniae in wastewater samples was reported. Such methods could reflect both clinical activity and the relationship between antibiotic use and the relative abundance of genes encoding antibiotic resistance in hospital wastewater. Therefore, the cultureindependent antibiotic resistance monitoring system would provide real-time data of the increasing beta-lactam resistance level [17].
Urgency of the availability of such monitoring method in low-and middle-income countries need to be presented in an assured and massive manner to provide sufficient data in real world setting towards future environmental policy [14]. The one health approach in hospital environments, either wards or wastewater, is expected to support antibiotic resistance control in hospital settings. To our knowledge, study on antimicrobial resistance in hospital wastewater in Indonesia is scarce [18,19]. The analysis of antibiotic resistance genes in nosocomial pathogens obtained from hospital wastewater has not yet been performed. This study aimed to measure the beta-lactam resistance genes through two different approaches, including directly from hospital wastewater and from two nosocomial pathogens (Escherichia coli and Klebsiella pneumoniae) isolated from the hospital wastewater, in a referral hospital in Indonesia. The correlation of beta-lactam resistance genes according to the samples collected each week, and the difference between the relative abundance of the beta-lactam resistance genes in hospital wastewater and nosocomial pathogens were evaluated in this study.

Setting and Sample Collection
The study was carried out in the Dr. Saiful Anwar Hospital in Malang, Indonesia, which is a 700-bed referral hospital. Wastewater samples were collected from an influent aerobic wastewater treatment plant containing untreated wastewater flowing from a mixture of wards, laundry, and kitchen. Sampling was performed twice a week randomly for six weeks from October 2021 until November 2021 (weeks 40-45); in total, we collected 12 hospital wastewater samples. Samples were collected using clean equipment by trained sampling personnel with personnel protective equipment. Grab samples of one liter of wastewater were collected in sterile bottles and kept cold for the 30 min it took to reach the lab.

Wastewater Filtration
The wastewater filtration method was carried out as previously described [17]. As soon as the wastewater sample arrived at the laboratory, 100 mL wastewater sample was concentrated directly into autoclaved poly-ethersulfone (PES) hydrophilic membranes 0.2 m using a Nalgene Rapid-Flow disposable filter unit (Thermo Fisher Scientific, MA, USA). After being transferred to PowerWater bead tubes, the filters were stored at −20 • C until the DNA was extracted.

Cultivation and Identification of Target Nosocomial Pathogens
We screened Escherichia coli and Klebsiella pneumoniae as target nosocomial pathogens from hospital wastewater samples. Fifty milliliters of hospital wastewater from sample bottle were transferred into conical centrifuge tube. Then, the samples were concentrated by centrifugation with 3000 rpm speed for 10 min to generate intact bacterial pellets.
A loop of bacterial pellets was inoculated to Eosin Methylene Blue (EMB) agar (Oxoid, Basingstoke, UK) and MacConkey agar (Oxoid, Basingstoke, UK), at 37 • C overnight to get cultural characteristics of Escherichia coli and Klebsiella pneumoniae colonies, respectively. Colonies showing metallic sheen in EMB agar and mucoid lactose fermenter colonies in MacConkey agar were suspected. In case the colonies were not isolated, the subculture was carried out for obtaining pure culture.

DNA Isolation
DNA isolation was carried out through two approaches. First, the DNA was directly isolated from the PES membrane in PowerWater bead tubes using the DNeasy PowerWater Kit (Qiagen, Venlo, The Netherlands) based on the manual instructions. It represented the DNA encountered in the hospital wastewater. Second, the DNA was extracted from the targeted nosocomial pathogens isolated from the hospital wastewater sample using DNeasy Blood and Tissue Kit (Qiagen, Venlo, The Netherlands). DNA was spectrophotometrically measured for both quantity and quality using NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA). Prior to delivery to Resistomap, Finland, DNA was kept at −20 • C.
The 72 primers sets were quantified using a SmartChip Quantitative PCR (HT-qPCR) system, a SmartChip Real-Time PCR system (Takara Bio, Mountain View, CA, USA). The SmartChip Real-Time PCR has 5184 reaction wells in a 100 nL volume that contain 1X SmartChip TB Green Gene Expression Master Mix (Takara Bio, Japan), nuclease-free PCRgrade water, 300 nM of each primer, and a 2 ng/L DNA template mixture. A 10 min denaturation at 95 • C was followed by 40 cycles of 30 s at 95 • C and 30 s at 60 • C. Each primer set was subjected to melting curve analysis. As the detection limit, a Ct value of 27 was chosen. Amplicons with non-specific melting curves and multiple peaks were ruled out [20].

Statistical Analysis and Visualization Using ResistApp
As previously described [17], data from the High-Throughput SmartChip quantitative PCR system was processed and analyzed using a digital platform, ResistApp (Resistomap, Finland). SPSS 26.0 was used to compare the relative abundance of beta-lactam antibiotic resistance genes from hospital wastewater to those obtained from nosocomial pathogens encountered in hospital wastewater. p < 0.05 was considered significant.

Monitoring of Beta-Lactam Resistance Genes in Hospital Wastewater and Nosocomial Pathogens by Time
Monitoring 19 beta-lactam resistance genes in the hospital wastewater was carried out for six weeks in this study. Relative abundance of beta-lactam resistance genes detected over time in hospital wastewater samples, Escherichia coli, and Klebsiella pneumoniae were presented in Table 1. Table 1. Gene abundance of beta lactamase resistance genes relative to 16S rRNA in hospital wastewater samples and pathogens (copies/16s rRNA gene copies).

Klebsiella pneumoniae
Week 40 Week 41 Week 42 Week 43 Week 44 Week 45 Except for two genes, bla IMI and bla SME , all targeted beta-lactam resistance genes were found in hospital wastewater over time. We discovered 12 beta-lactam resistance genes that were always present with varying relative abundance in the hospital wastewater after six weeks of monitoring, out of the 17 beta-lactam resistance genes detected over time. Several wastewater samples lacked five carbapenem resistance genes: bla KPC , bla KPC2 , bla OXA48 , bla OXA51 , and bla VIM . Within six weeks of monitoring, however, no beta-lactam resistance genes were always detected in Escherichia coli and Klebsiella pneumoniae isolated from hospital wastewater. Escherichia coli and Klebsiella pneumoniae revealed only eight and seven beta-lactam resistance genes, respectively ( Table 1). The relative abundance of beta-lactam resistance genes either carbapenem resistance genes or extended spectrum beta-lactamase (ESBL) genes was not significantly different by the time of sampling (Kruskal-Wallis analysis; p = 0.142).

Comparison of the Relative Abundance of Beta-Lactam Resistance Genes in Hospital Wastewater, Escherichia coli, and Klebsiella pneumoniae
The relative abundance of each beta-lactam resistance genes in hospital wastewater, Escherichia coli, and Klebsiella pneumoniae within six weeks of monitoring was compared. One way ANOVA analysis showed that the relative abundance of bla GES was significantly higher compared to other beta-lactam resistance genes found in the wastewater samples (p < 0.001). Furthermore, bla TEM was significantly most abundant among beta-lactam resistance genes in Escherichia coli (p < 0.001) ( Table 2). Table 3 shows that among ESBL genes, bla GES reached the highest abundance only in hospital wastewater (1 × 10 −2 copies/16s rRNA gene copies), whereas bla TEM was the most abundant in Escherichia coli (4 × 10 −2 copies/16s rRNA genes copies) and Klebsiella pneumoniae (1 × 10 −4 copies/16s rRNA genes copies). The median of relative abundance of each beta-lactam resistance genes in hospital wastewater, Escherichia coli, and Klebsiella pneumoniae during the study period was analyzed by Kruskal-Wallis. The relative abundance of all beta-lactam resistance genes was significantly different among hospital wastewater samples, Escherichia coli, and Klebsiella pneumoniae except for bla OXA48 and bla TEM (Table 3).

Antibiotic Susceptibility Profile of Escherichia coli and Klebsiella pneumoniae Isolated from Hospital Wastewater
We found seven Escherichia coli and 10 Klebsiella pneumoniae isolates in 12 hospital wastewater samples. The antibiotic susceptibility test was carried out prior to further ARGs analysis. The antibiotic susceptibility test showed multidrug resistant Klebsiella pneumoniae isolated from hospital wastewater, which were resistant to all antibiotics tested except meropenem. All isolates were susceptible to meropenem. Statistical analysis showed that Klebsiella pneumoniae might be associated with the resistance to piperacillin/tazobactam (p < 0.001), ceftriaxone (p = 0.001), and cefepime (p < 0.001) ( Table 4).  The colors indicate the gene abundances relative to 16s rRNA gene; ww = wastewater; Ec = Escherichia coli; Kp = Klebsiella pneumoniae; ND = not determined; a p < 0.001; b p < 0.001; c p = 0.153. Table 3. Abundance of beta-lactam resistance genes in hospital wastewater and nosocomial pathogens isolated from hospital wastewater (copies/16s rRNA gene copies).

Genes
.001 N = number of samples; LCI = lower confidence interval; UCI = upper confidence interval; * median of relative abundance was analyzed by Kruskal-Wallis.

Discussion
To our knowledge, this is the first study that analyzed the presence and abundance of antimicrobial resistance genes (ARGs) in hospital wastewater samples compared to nosocomial pathogens isolated from hospital wastewater samples in Indonesia. In addition, this study specifically analyzed the various beta-lactam resistance genes in hospital wastewater samples and nosocomial pathogens isolated from the hospital wastewater.
Through our analysis, extended spectrum beta-lactamase (ESBL) genes and carbapenem resistance genes were present in hospital wastewater samples over time. bla GES was the most abundant beta-lactam resistance genes found in hospital wastewater (p < 0.001). In concordance with the hospital wastewater samples, we detected two Klebsiella pneumoniae isolates bearing high abundance of bla GES . bla GES is a transferable gene located in plasmid encoding carbapenem resistant to Pseudomonas aeruginosa and Klebsiella pneumoniae [21]. Although the bla GES was detected over time in wastewater samples, the abundance of bla GES was lower than in Klebsiella pneumoniae isolates. Wastewater treatment processes might affect the presence of antibiotic resistance genes in wastewater samples [22]. Further study is required to investigate the correlation between bla GES in hospital wastewater and Klebsiella pneumoniae isolated from the hospital wastewater. Furthermore, the clinical activity may be represented by ARGs profile in hospital wastewater.
Antibiotic susceptibility profile showed higher resistant to piperacillin/tazobactam, ceftriaxone, and cefepime among Klebsiella pneumoniae than Escherichia coli isolated from hospital wastewater (p < 0.001, p = 0.001, and p < 0.001). The high resistance of Klebsiella pneumoniae in the hospital wastewater might be associated with the high prevalence of third generation cephalosporins clinical Klebsiella pneumoniae isolates as reported by a national surveillance on antimicrobial resistance in Indonesia [5]. The nosocomial pathogens encountered in wastewater samples may originate from clinical isolates. Therefore, hospital wastewater has higher risk of ARGs dissemination through horizontal gene transfer such as transduction, transformation, and conjugation [22,23].
The present study showed predominant ESBL genes including bla TEM , bla CTX-M5 , and bla CTX-M8 , which were encountered in Escherichia coli and Klebsiella pneumoniae isolated from wastewater samples. bla TEM was the most abundant among the ESBL genes in this study. This result is in concordance with the high prevalence of ESBL-producing Escherichia coli and Klebsiella pneumoniae among clinical cultures reported in the previous studies [24][25][26]. It is suggested that Escherichia coli and Klebsiella pneumoniae in the wastewater samples originated from the clinical specimens. Similar to bla GES , bla TEM is a transferable gene located in a plasmid encoding ESBL enzymes; therefore, they spread easily among different bacteria.
Carbapenemase genes, including bla CARB , bla KPC , bla OXA48 , bla OXA51 , and bla VIM were detected in hospital wastewater but not in Escherichia coli and Klebsiella pneumoniae isolated from the wastewater samples. It is aligned with the antibiotic susceptibility profile, presenting no meropenem resistant among Escherichia coli and Klebsiella pneumoniae. In accordance with the previous study, there was no association between the relative abundance of antibiotic resistance genes in wastewater samples to those in the clinical isolates [27,28].
In this study, bla IMI and bla SME genes were not detected either in hospital wastewater samples or in Escherichia coli and Klebsiella pneumoniae isolates over time. bla IMI and bla SME are carbapenemase genes encoding IMI and SME enzymes located in the chromosome of Enterobacter cloacae and Serratia marcescens, respectively [29]. Therefore, the bla IMI and bla SME genes are restricted in Enterobacter and Serratia genus due to the less transferable genes [21,30].

Conclusions
We reported the dynamics of the abundance of beta-lactam resistance genes in hospital wastewater and nosocomial pathogens in six weeks of monitoring. Higher abundances of bla GES in hospital wastewater and bla TEM in Escherichia coli and Klebsiella pneumoniae isolated from the hospital wastewater were detected than other beta-lactam resistance genes. Further investigation is required to evaluate the correlation between the ESBL genes in hospital wastewater and the prevalence of ESBL-producing Escherichia coli and Klebsiella pneumoniae obtained from clinical cultures. Therefore, a potential outbreak of ESBL-producing Escherichia coli and Klebsiella pneumoniae could be detected by hospital wastewater-based monitoring systems using culture independent methods. This study had certain limitations. First, we used grab samples instead of composite samples leading to less representative ARGs data. Second, we screened two nosocomial pathogens including Escherichia coli and Klebsiella pneumoniae using conventional culture methods; therefore, the presence of other pathogens was not detected. Third, wastewater samples were collected from an influent wastewater treatment plant containing untreated wastewater flowing from a mixture of wards, laundry, and kitchen that might be influenced by detergent or disinfectant antimicrobial activities. Fourth, the present study was a pilot of monitoring ARGs in hospital wastewater in Indonesia; therefore, we started with a small-scale study. Further investigation with more hospitals involved is recommended.

Informed Consent Statement: Not applicable.
Data Availability Statement: The data from this study are available upon request from the corresponding author.