Abstract
The waters of rivers Swat and Kabul are the main water source for domestic and irrigation purposes in the northwestern part of Pakistan. However, this water has been contaminated due to human activities. This study aimed to analyze the water of these rivers for occurrence of antibiotic resistance genes among Gram-negative bacteria. Samples were collected from 10 different locations of these rivers. The samples were processed for the isolation of Gram-negative bacteria. Isolated bacteria were checked against 12 different antibiotics for susceptibility. The isolates were also analyzed for the presence of seven antibiotic resistance genes. A total of 50 bacterial isolates were recovered that belonged to five different bacterial genera, that is, Escherichia coli, Klebsiella oxytoca, Pseudomonas aeruginosa, Raoultella terrigena (Klebsiella terrigena), and Pseudomonas fluorescens. Antibiotic resistance pattern was cefixime 72%, cephalothin 72%, ampicillin 68%, nalidixic acid 68%, kanamycin 54%, streptomycin 42%, sulfamethoxazole–trimethoprim 28%, chloramphenicol 28%, meropenem 8%, gentamicin 8%, amikacin 2%, and tobramycin 2%. The prevalence of bla-TEM gene was 72% (n = 36), aadA gene 34% (n = 17), sul gene 32% (n = 16), bla-CTXM gene 12% (n = 6), int gene 66% (n = 33), and int1 gene 6% (n = 3). This information highlights the need for controlling and monitoring the release of domestic wastes to rivers.
Introduction
Water pollution is one of the main issues affecting people throughout the world. Consumption of contaminated and inadequately treated water is the major cause of waterborne illnesses and multiple outbreaks of infectious diseases. Pakistan is faced with problems of water quality and quantity due to poor water resources management strategies. It is reported that approximately 40% deaths in Pakistan occur due to waterborne diseases (Azizullah et al., 2011). Waterborne diarrhea is the cause of 60–80% deaths of children in Pakistan (Sardar et al., 2015). The situation is quite alarming in some cities of Pakistan where drinking water is not safe for human consumption because of contamination with bacterial and viral pathogens. Contamination of river water mainly through human activities has been reported that could lead to the development and spread of infections causing bacteria carrying resistant genes (Servais and Passerat, 2009; Watkinson et al., 2007). Water pollution frequently occurs due to release of domestic wastes into water bodies (Strohschön et al. 2013). Domestic waste could be a source of antibiotic-resistant bacteria, and water contaminated with such bacteria could play an important role in spreading infections (Ham et al., 2012).
Indiscriminate use of antibiotics is responsible for environmental contamination. In aquatic environment, antimicrobials are released through anthropogenic activities leading to development of resistant bacteria. High prevalence of antibiotic resistance genes has been shown to correlate with antibiotic contamination of water (Chen et al., 2013). Different studies have revealed the occurrence of antibiotic-resistant bacteria in aquatic environment with the capability of transferring resistance characteristics to other bacteria (Akinbowale et al., 2006; Cheng et al., 2014; Ham et al., 2012). Pathogenic bacteria released into river water possess various resistance genes that are part of both genomic DNA and transferable genetic particles like plasmids, integrons, and transposons and could spread among bacteria of water and soil environments (Alonso et al., 2001; Livermore, 2012; Schlipköter and Flahault, 2010).
Rivers Kabul and Swat in district Charsadda, Pakistan, consist of a complex network of interweaved rivers. The waters of the two rivers are used by local people for domestic and irrigation purposes. There are complaints of people living near these rivers in district Charsadda regarding human and livestock diseases due to polluted water. Fish population has also been affected due to water pollution in the area. Hence, this study aimed to analyze the waters of these rivers for the existence of antibiotic resistance genes possessed by Gram-negative bacteria occupying these water environments.
Materials and Methods
Sample collection and bacterial isolation
Samples were collected from the waters of rivers Kabul and Swat at 10 different areas of district Charsadda, Pakistan, according to standard procedures (Parija, 2006). The samples were labeled according to the sampling location and immediately transferred to the laboratory and processed for isolation of Gram-negative bacteria. Samples were serially diluted in saline solution. These dilutions were used to inoculate MacConkey agar. After inoculation, plates were incubated at 37°C for 18–24 h. The plates were then checked for bacterial growth. Single colonies were then picked and inoculated onto fresh MacConkey agar to get pure cultures. Isolated bacteria were identified using morphological and biochemical tests like Gram staining, oxidase test, catalase test, citrate test, indole test, urease test, motility test, triple sugar iron test, methyl red and Voges–Proskauer test, nitrate test, and DNase test (Parija, 2006).
Antibiotic susceptibility
Twelve different antibiotics were used for antibiotic susceptibility. These antibiotics included amikacin (AMK, 10 µg), tobramycin (TOB, 10 µg), cefixime (CFM, 5 µg), meropenem (MEM, 10 µg), ampicillin (AMP, 10 µg), chloramphenicol (CAM, 30 µg), kanamycin (KAN, 30 µg), nalidixic acid (NX, 30 µg), cephalothin (CET, 30 µg), gentamicin (GEN, 10 µg), sulfamethoxazole–trimethoprim (TMP-SMX, 25 µg), and streptomycin (STR, 10 µg). Inoculum concentration in saline solution was adjusted by comparing with 0.5 McFarland standards. The sample was inoculated onto Mueller–Hinton agar plates. Antibiotic discs were then added to the plates and the plates kept at 37°C for 16–24 h.
Resistance genes analysis
Bacterial DNA was isolated using standard protocols (Wilson, 1987). Fresh bacterial cultures on nutrient agar medium were grown. Nutrient broth culture (5 mL) was grown at 37°C for 16–24 h. This culture (1.5 mL) was added to a microcentrifuge tube and the sample centrifuged at 1300 rpm for 2 min. Bacterial pellet was resuspended in lysis buffer (570 µL Tris-EDTA buffer, 30 µL 10% sodium dodecyl sulphate, and 3 µL of 20 mg/mL proteinase K). The sample was kept at 37°C for 1 h. Next, 100 µL NaCl (5 M) was added to it and the sample mixed again. Next, 80 µL CTAB/NaCl solution was mixed with the sample and the sample was kept at 65°C for 10 min. In the next step, an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) was mixed with the sample and the sample was centrifuged at 1300 rpm for 6 min. The supernatant was transferred to another tube, and an equal volume of chloroform:isoamyl alcohol was again mixed with the sample. The sample was centrifuged at 1300 rpm for 6 min. The supernatant was transferred to a fresh tube, and 600 µL isopropanol was mixed with the sample. DNA was precipitated by inverting the tube. It was followed by centrifugation at 1300 rpm for 3 min. The supernatant was discarded, and 70% ethanol was added to the tube. DNA pellet was washed by inverting the tube. The sample was then centrifuged at 1300 rpm for 6 min. The supernatant was discarded and the DNA pellet air dried. The DNA pellet was then dissolved in 100 µL research grade water. The DNA quality was analyzed through gel electrophoresis.
The isolated DNA was used as template for detection of intI (integron), Int (integrase), Sul (sulfonamide), aadA (aminoglycoside), bla-TEM (beta lactam), Bla-CTXM (cefotaxime), and Bla-NDM1 (carbapenemase) genes using previously published primers (Farshadzadeh et al., 2014; Peerayeh et al., 2014; Shanthi et al., 2014; Van et al., 2008; Zhao et al., 2001). The sequences of the primers used were intI (forward primer: 5ʹGGCATCCAAGCACAAGC3ʹ, reverse primer: 5ʹAAGCAGACTTGACTGAT3ʹ), Int (forward primer: 5ʹ CCTCCCGCACGATGATC3ʹ, reverse primer: 5ʹ TCCACGCATCGTCAGGC-3′ʹ), Sul (forward primer: 5ʹ TTCGGCATTCTGAATCTCAC-3′ʹ, reverse primer: 5ʹ ATGATCTAACCCTCGGTCTC-3′ʹ), aadA (forward primer: 5ʹ TGATTTGCTGGTTACGGTGAC-3′ʹ, reverse primer: 5ʹ CGCTATGTTCTCTTGCTTTTG-3′ʹ), bla-TEM (forward primer: 5ʹ GAGTATTCAACATTTTCGT-3′ʹ, reverse primer: 5ʹ ACCAATGCTTAATCAGTGA-3′ʹ), Bla-CTXM (forward primer: 5ʹ CGCTTTGCGATGTGCAG-3′ʹ, reverse primer: 5ʹ ACCGCGATATCGTTGGT-3′ʹ), and Bla-NDM1 (forward primer: 5ʹ GGGCAGTCGCTTCCAACGGT-3′ʹ, reverse primer: 5ʹ GTAGTGCTCAGTGTCGGCAT-3′ʹ).
Target genes were amplified using FIREPol master mix (Solis BioDyne, Cat. No. 04-12-00125). PCR reaction consisted of 20 µL final volume having 4 µL master mix (5×), 1 µL template DNA, 1 µL primer (0.5 µM) both forward and reverse, and 13 µL water. Reaction conditions for amplification were used according to previous reports (Farshadzadeh et al., 2014; Peerayeh et al., 2014; Shanthi et al., 2014; Van et al., 2008; Zhao et al., 2001). Amplified products were detected through gel electrophoresis.
Statistical analysis
Correlations between phenotypic resistance to antibiotics and occurrence of antibiotic resistance genes were compared using Pearson′s correlation coefficient. The analysis was carried out through JASP computer software (Version 0.19.0).
Results
Types of isolated bacteria
A total of 50 bacterial isolates were recovered that belonged to five different bacterial genera, that is, Escherichia coli, Klebsiella oxytoca, Pseudomonas aeruginosa, Raoultella terrigena (Klebsiella terrigena), and Pseudomonas fluorescens. The most common isolated bacteria were P. aeruginosa (n = 19) followed by E. coli (n = 14), K. oxytoca (n = 12), P. fluorescens (n = 4), and R. terrigena (n = 1).
Antibiotic sensitivity
Resistance pattern against the 12 antibiotics was CFM 72% (n = 36), CET 72% (n = 36), AMP 68% (n = 34), NX 68% (n = 34), KAN 54% (n = 27), STR 42% (n = 21), TMP-SMX 28% (n = 14), CAM 28% (n = 14), MEM 8% (n = 4), GEN 8% (n = 4), AMK 2% (n = 1), and TOB 2% (n = 1). Multiple antibiotic resistance was observed in 40 isolates (Table 1). One isolate of P. aeruginosa (V2) was resistant to nine different antibiotics. Four isolates were resistant to eight different antibiotics. These included one isolate of K. oxytoca (K1D) and three isolates of P. aeruginosa (V5, V6, and V9). Three isolates were resistant to seven different antibiotics. These included one isolate of E. coli (K5E) and two isolates of P. aeruginosa (V4 and V11). Ten isolates were resistant to six different antibiotics. These included four isolates of E. coli (K1A, KIG, K2I, and K6I), one isolate of R. terrigena (K3I), three isolates of P. aeruginosa (K4B, V8, and V10), and two isolates of K. oxytoca (K5F, K6D). Eight isolates were resistant to five different antibiotics. These included two isolates of E. coli (K2A and K3J), three isolates of K. oxytoca (K4D, K4E, and K6B), and three isolates of P. aeruginosa (K4H, K6A, and V3).
Antibiotic Resistance Status of Isolated Gram-Negative Bacteria
AMK, amikacin; AMP, ampicillin; CAM, chloramphenicol; CET, cephalothin; CFM, cefixime; GEN, gentamicin; KAN, kanamycin; MEM, meropenem; NX, nalidixic acid; STR, streptomycin; TMP-SMX, sulfamethoxazole–trimethoprim; TOB, tobramycin.
Occurrence of drug resistance genes
Among the 50 isolates, 36 (72%) were positive for bla-TEM gene (Fig. 1A, Table 2). Among these, 12 E. coli isolates (K1A, K1G, K1I, K2A, K2G, K2I, K3B, K3F, K3J, K5E, K5I, and K6J), 11 K. oxytoca isolates (K1B, K1D, K3H, K4D, K4E, K5F, K6B, K6D, K6E, K6G, and K6I), 12 P. aeruginosa isolates (K6A, K1F, K4B, K4H, K5A, K8B, V4, V5, V6, V8, V10, and V11), and 1 R. terrigena isolate (K3I) were positive for bla-TEM gene. The occurrence of aadA gene was confirmed in 17 (34%) isolates (Fig. 1B, Table 2). Among these, 6 K. oxytoca isolates (K1B, K4D, K6B, K6E, K6G, and K6I), 7 P. aeruginosa isolates (K1F, K4B, K8B, K9G, V8, V11, and G7), and 4 E. coli isolates (K1H, K2G, K3B, and K5I) possessed aadA gene. The sul gene was detected in 16 (32%) isolates (Fig. 1C, Table 2). These included 5 isolates of K. oxytoca (K1B, K6B, K6D, K6E, and K6G), 5 isolates of E. coli (K1I, K2A, K2G, K2I, and K3B), 4 isolates of P. aeruginosa (V4, V5, V8, and G7), and 2 isolates of P. fluorescens (G4 and G9). Six (12%) isolates had bla-CTXM gene (Fig. 1D, Table 2). These included 3 isolates of E. coli (K1G, K1H, and K5I), 1 isolate of K. oxytoca (K6E), and 2 P. aeruginosa (V6 and V10). All bacterial isolates were negative for bla-NDM1 gene (Fig. 1E, Table 2). Thirty-three (66%) isolates had int gene (Fig. 1F, Table 2). These included 11 E. coli isolates (K1A, K1G, K1H, K1I, K2A, K2G, K3B, K3F, K5E, K5I, and K6J), 8 K. oxytoca isolates (K1B, K1D, K3H, K4E, K5F, K6B, K6E, and K6G), 12 P. aeruginosa isolates (K1F, K4B, K4H, K5A, K5G, K6A, K8B, K9G, V2, V5, V8, and G7), 1 R. terrigena isolate (K3I), and 1 P. fluorescens isolate (G4). Class 1 integron (int1 gene) was observed in 3 (6%) isolates (Fig. 1G, Table 2). The size of variable region of Class 1 integron ranged from 400 to 1200 bp. Class 1 integron size was 400 bp in K. oxytoca isolate K3H, 800 bp in E. coli isolate K1G, and 1200 bp in K. oxytoca isolate K6G.
Occurrence of Target Genes Among Isolated Bacteria
Phenotypic resistance association with occurrence of antibiotic resistance genes
Association between phenotypic antibiotic resistance and occurrence of resistance-related genes is shown in Table 3. No significant correlation was observed between phenotypic resistance and existence of resistance genes. Surprisingly, a strong correlation between kanamycin resistance and occurrence of bla-TEM gene was observed (p < 0.01).
Pearson’s Correlations (r) of Antimicrobial Resistance and Resistance Genes
p < 0.01.
AMK, amikacin; AMP, ampicillin; CAM, chloramphenicol; CET, cephalothin; CFM, cefixime; GEN, gentamicin; KAN, kanamycin; MEM, meropenem; NX, nalidixic acid; STR, streptomycin; TMP-SMX, sulfamethoxazole–trimethoprim; TOB, tobramycin.
Discussion
The current study evaluated the presence of resistance genes in Gram-negative bacteria inhabiting rivers Swat and Kabul for the first time. The waters of the two rivers were found to be contaminated with E. coli (28%), K. oxytoca (24%), P. aeruginosa (38%), R. terrigena (Klebsiella terrigena) (2%), and P. fluorescens (8%). A study of aquatic environments in Congo, India, and Switzerland reported 39% prevalence of P. aeruginosa like the current findings (Devarajan et al., 2017). Another report from Brazil reported 50% prevalence of carbapenem-resistant P. aeruginosa in river waters (Turano et al., 2016). Antibiotic-resistant P. aeruginosa were also reported from river water and wastewater treatment plants in Portugal and France (Quinteira and Peixe, 2006; Slekovec et al., 2012). Mackowiak et al. (2018) reported 100% prevalence of E. coli in various river water samples from an urban river in Germany.
In the current study, 40 isolates showed multiple antibiotic resistance. A study of the river Rhine, Germany, reported 9% resistance against TMP-SMX in coliform bacteria and 6% multidrug resistance (Stange et al., 2016). A study from Spain reported up to 94.7% resistance to AMP and 65.5% resistance to STR from different sites of a river (Sidrach-Cardona et al., 2014). Walia et al. (2016) reported 25–40% chloramphenicol resistance and 30–50% gentamicin resistance in bacterial isolates from various locations of Clinton River Water, USA. Prevalence of resistant bacteria in rivers Swat and Kabul could be the result of the release of domestic and industrial wastes and other human activities alongside the two rivers. This interference could also be a contributing factor toward the development of multidrug-resistant bacteria that could rapidly evolve in the presence of selection pressure imposed by the release of contaminants in river water (Ram et al., 2008).
In the current studies, a high prevalence (72%) of bla-TEM gene was observed. A previous study of River Rhine, Germany, reported 14.4% prevalence of bla-TEM in isolated E. coli (Stange et al., 2016). A study of Nigerian rivers reported 21% prevalence of bla-TEM among E. coli isolates recovered from the rivers (Titilawo et al., 2015). Beta lactams are the commonly used antimicrobials that have broad-spectrum activities and low toxicities. Apart from the release of beta lactam resistance bacteria into the river through anthropogenic activities, high prevalence of bla-TEM in the current study could also be due to naturally existing beta lactam-resistant bacteria that might have evolved in response to long-term exposure to trace amounts of beta lactam antibiotics as suggested previously (Esiobu et al., 2002).
In the current study, prevalence of aadA gene was 34%. Some studies of Lake Geneva, Switzerland, have reported high occurrence of aadA since the beginning of the 20th century (Devarajan et al., 2015; Devarajan et al., 2017), which suggests that the gene could be present in natural bacterial population before the introduction of antibiotics (Demaneche et al., 2008). Aminoglycosides disturb the function of ribosomes leading to blocking of protein synthesis. The use of aminoglycosides has been very low because of toxicity in terms of kidney damage and injury to auditory nerves (Demaneche et al., 2008) that further suggests the development of aminoglycosides resistance in the absence of antibiotics use.
The prevalence of sul1 gene was 32% in the current study that is higher than a previous report (8% prevalence) from Nigeria (Titilawo et al., 2015). Pham et al. reported 94.1% prevalence of sul1 in water effluents of channels connecting shrimp farms to a river (Pham et al., 2018). Sulfonamides competitively inhibit folic acid synthesis (Levy and Marshall, 2004). These antibiotics have relatively low toxicity that has led to indiscriminate use of these antibiotics globally and contributed toward high level of resistance in the environment (Lateef, 2004). Additionally, sul1 has been detected in non-polluted water environments and frequently associated with class 1 integrons suggesting its horizontal transfer among bacteria and possible release into the environment (Pham et al., 2018; Hoa et al., 2008).
In this study, prevalence of bla-CTXM gene was 12%. A study from Brazil reported 92.7% prevalence of bla-CTXM in clinical and domestic effluents and river water samples (Conte et al., 2017). Vivant et al. (2016) reported 84% prevalence of bla-CTXM in a wetland in France. Environmental pollution with Gram-negative bacteria from domestic wastes could contribute toward the spread of bla-CTXM in river water.
Class 1 integron (int, Int1 genes) were also detected in the current study. A study from Switzerland reported 8% prevalence of int1 in Gram-negative bacteria isolated from water samples (Carnelli et al., 2017). Another study reported 6.9−16.7% prevalence of the gene from different sites of Zimny Potok River, Poland (Makowska et al., 2016). Class 1 integrons had 10% prevalence in bacteria recovered from River Rhine, Germany (Stange et al., 2016). A Chinese study reported 41% occurrence of class 1 integrons in E. coli recovered from Minjiang River that was attributed to the indiscriminate use of antibiotics in human therapy and livestock, fish, and poultry farming in the area (Chen et al., 2011).
To find the association of antibiotic resistance with distribution of resistance genes, Pearson’s correlation analysis was performed. We found no association between antibiotic resistance and the corresponding resistance genes. Similar results have been reported by Liu et al. (2021). The existence of multiple antibiotic resistance mechanisms and occurrence of diverse resistance genes in bacteria conferring resistance to a single type of antibiotic could be responsible for the lack of correlations in the current studies. Further studies could be designed to investigate a large number of resistance genes in the isolated bacteria to better understand phenotypic resistance and genotypic factors. It is also likely that environmental selection pressure contributes toward diversity and spread of antibiotic resistance among bacteria inhabiting river water (Natarajan et al., 2018). Further studies could be designed to find associations between environmental factors and resistance patterns. Surprisingly, a strong correlation between KAN resistance and occurrence of bla-TEM gene was noted in the current study. The co-occurrence of the two genes on the same mobile genetic elements like transposons and plasmids could account for this correlation that has also been reported previously (Altayb et al., 2022).
Conclusion
In conclusion, presence of multiple antibiotic-resistant bacteria and resistance genes in the water of rivers Swat and Kabul is a serious health risk. These findings highlight the need for proper sanitation strategies and safe water provision in the area. The indiscriminate use of antibiotics and other chemicals for human and animals should be avoided. Modern methods of wastewater treatment should be adopted to control release of resistant bacteria into river water. Proper monitoring of sewage water, animal waste, and wastewater disposal is needed to restrict the spread of resistant bacteria and antibiotic resistance genes through water.
Footnotes
Acknowledgment
The authors thank the directress of Centre of Biotechnology and Microbiology, University of Peshawar, for her support to accomplish the study.
Authors’ Contributions
Both the authors contributed substantially to the design, experimental work, preparation, and revision of the article. The article was approved by both the authors for submission. R.S. conducted the experiments and data analysis and contributed to article writing. K.A. contributed to the design, data analysis, article writing, and proof reading.
Disclosure Statement
The authors have no conflict of interest to declare.
Funding Information
The research was not funded.