Emergence of bla_NDM-1 Harboring Klebsiella pneumoniae ST29 and ST11 in Veterinary Settings and Waste of Pakistan

Introduction

The modern intensive integrated livestock production systems require regular antibiotics use at farms to maintain animal health and production. Globally, it is suggested that antibiotic consumption is double in animals compared to humans. The regular and imprudent use of antibiotics in modern veterinary practices is associated with the emergence of different multidrug-resistant (MDR) bacteria.^1,2 These MDR pathogens of animal origin may be disseminated to humans via the wider environment including food products, sewage and agricultural system.^1–3

Klebsiella pneumoniae is the most important member of the Klebsiella genus of Enterobacteriaceae. It can transfer to humans through contaminated products of like poultry products, beef, fish, milk, etc.^4,5 Additionally, the prevalence of carbapenem-resistant K. pneumoniae causing community and hospital-acquired infections is increasing significantly throughout the world, especially in developing countries.^6,7 Carbapenems are a class of beta-lactam antibiotics, are regularly prescribed as a last choice in the treatment of Gram-negative bacteria infection but the emergence of carbapenem-resistant Enterobacteriaceae is a critical clinical problem in the world.^8,9

The key mechanisms for carbapenem resistance include alteration in porins function or expression, efflux pumps, and acquirement of enzymes capable of antibiotic hydrolyzing called carbapenemases. These enzymes are b-lactamases of Ambler class A, class D, and class B (Metallo-β-lactamases; MBLs).^1,10,12–14 Currently, a total of 5 Carbapenemases are key public health concerns, among them, New Delhi Metallo- β-lactamase-1 (NDM-1) has attracted the most attention because bla_NDM-1 has diverse antibiotic resistance activity.¹⁵ The bla_NDM-1 was the first time reported in K. pneumoniae isolated from an Indian patient in Sweden in 2008.⁸ Since then a lot of studies from various regions of the world have reported the occurrence of the bla_NDM-1.^8,9 Despite all these reports so far, data about the detection of bla_NDM-1 in veterinary and animal settings is still lacking.

Pakistan is a leading livestock populated country in the world. In the last decade, intensive livestock farming increased on a large scale in Pakistan. Therefore, livestock farmers are adapting the modern intense farming protocols in which antibiotics are used as food supplements routinely. This unrestrained antibiotic application in livestock feed is a leading factor that steers the evolution of multidrug-resistant (MDR) pathogens. On the other hand, monitoring practices of antibiotics usage in veterinary and farm settings is inadequate.¹⁶ Recently we have reported the first bla_KPC harboring K. pneumoniae ST258 from Pakistan and we also have reported the occurrence of ESBL-producing K. pneumoniae in hospital settings and waste from Pakistan.^3,17 Here, for the first time from Pakistan, we demonstrated the prevalence of bla_NDM-1 harboring carbapenem-resistant K. pneumoniae in veterinary settings and waste from Pakistan.

Materials and Methods

Ethical Approval and Study Settings and Sample Collection

The study was approved by the Ethical Review Board (ERB), Government College University, Faisalabad Pakistan (letter No. 4162 dated 23–11-2017). Sample collection and processing was done at the Department of Microbiology Government College University, Faisalabad Pakistan while, molecular investigations were conducted at the Paul G. Allen School for Global Animal Health, Washington State University, Washington, USA, and Biomedical Research Center, Northwest Minzu University, Lanzhou China. In this study, we applied a nonprobability sampling technique to collect the samples (n= 138). The detail about samples distribution is given in Table 1. Standard aseptic conditions and microbiological procedures were adopted for sample collection and transportation.

Table 1 Description the Sample Distribution and Their Positive Results

Isolation and Identification of Klebsiella pneumoniae

In brief, abattoir samples and fecal swabs were dipped into 1mL of PBS and then streaked on the surface of nutrient media; however, milk and wastewater samples were directly streaked on nutrient agar plates and were incubated at 37°C for 24 hours. Subsequently, Mackonkey’s agar and HiChrom Klebsiella Selective agar (Himedia^®) were used as selective media for the isolation of K. pneumoniae. API 20E Kit and VITEK identification system (bioMérieux, France) was applied for the biochemical identification of the isolates.

Molecular identification was done by amplifying and sequencing the 16S rDNA using specific primers through PCR. The reaction started with the initial melting temperature at 95°C for 3 min, a total of 35 PCR cycles were carried out with following scheme: denaturation at 95°C for 30 secs, annealing at 50°C for 25 secs, extension at 72°C for 65 sec and final extension was done at 72°C for 5 min. Afterward, agarose (ThermoFisher Scientific, USA) gel electrophoresis was performed and results were interpreted in the gel documentation system (Bio-Rad, USA).

Antibiotic Susceptibility Testing

The disc-diffusion assay was used to decipher the antibiotic resistance pattern of the isolates As described in CLSI 2018.¹⁸ Various groups of antibiotics along with their concentration used in the antibiogram analysis of K. pneumoniae include Penicillins (ampicillin 10µg and piperacillin 100µg), Cephalosporins (cefuroxime 30µg, cefixime 5µg, ceftriaxone 30µg, and cefepime 30µg), Carbapenems (meropenem 10µg), Fluoroquinolones (ciprofloxacin 5µg), Tetracycline (tetracycline 30µg and minocycline 30µg), Sulfa drugs (trimethoprim-sulfamethoxazole 1.25/23.75µg), Polymyxins (colistin 10µg), and tigecycline 15µg. All the tests were performed in duplicate and a control strain of K. pneumoniae (ATCC^® 13,883™) was used for the assay.

For further confirmation, the minimum inhibitory concentration (MIC) of the above-listed antibiotics against the 13 CRKP isolates was determined using broth micro-dilution assay as described previously.^12,13 Briefly, fresh isolates of K. pneumoniae were used to obtain 0.5 McFarland standards. Dilutions of the antibiotics were made with the concentrations range 0.06 μg/mL to 256 μg/mL. Inoculum from the wells showing the turbidity was streaked on nutrient agar plates for the confirmation of bacterial growth. The MIC of the isolates showing growth in the well having the antibiotic concentration of 256 μg/mL were considered as ≥256 μg/mL Results were interpreted according to the CLSI recommendations.

Phenotypic Detection of Carbapenem-Resistant K. pneumoniae

The Double Disc Synergy Test (DDST) was employed for the phenotypic characterization of ESBL producing K. pneumonia as described previously.¹⁸ Briefly, a 30 μg containing cefotaxime disc alone and a cefotaxime disc in combination with clavulanic acid, having the concentration of 30:10 μg, respectively were placed at a distance of 20 mm, and the difference between discs in terms of zone of inhibition was observed, the difference of ≥5 mm was considered positive for ESBL production.

Modified Hodge Test (MHT) was employed for phenotypic confirmation of carbapenem resistance. A 0.5 McFarland standard dilution of ATCC 25,922 (E. coli) was prepared in broth; a lawn was made using ten-fold dilution. In the center, ertapenem (10 µg) disk was placed. From the edge of the disk, the isolate was streaked up to the edge of the plate. Plates were incubated at 37°C for 24 hours. According to CLSI guidelines, MHT Positive ATCC1705 (K. pneumoniae) and MHT Negative ATCC1706 (K. pneumoniae) were used as control. Besides, MHT positive isolates were confirmed through the CarbNP test as recommended by CLSI 2018.

Molecular Detection of Antibiotic Resistance Genes (ARGs)

Freshly grown isolates of K. pneumoniae were subjected to DNA extraction through the DNA Extraction kit (Qiagen, Hilden, Germany). Quantification of extracted genomic DNA was carried out by using Thermo Scientific™ Nanodrop 2000, USA and DNA ≥60ng/μL concentration were considered further for the experiment. The extracted DNA of all isolates were subjected to PCR for the amplification of ARGs listed in Table 2 by using specific primers (IDT™, USA). The composition of the reaction mixture was; 5 µL of template DNA, 2 µL 100 pM forward and reverse primers, 10 µL of PCR Master Mix (2X DreamTaq, Thermo-Scientific™). The final volume was adjusted to 25 µL with sterile dH2O/Nuclease free water (Ambion-AM9932). The annealing temperature of each gene was adjusted according to primer’s Tm described in Table 2. Subsequently, PCR products were analyzed on 1.2% agarose gel electrophoresis and examined under the gel documentation system (BioRad, USA).

Table 2 Table Showing the Primer Sequence, Amplification Conditions and Distribution of Different Antibiotic-Resistance Determinants Among K. pneumoniae Isolates

Multi Locus Sequence Typing

The MLST was done according to the Pasteur MLST scheme (https://bigsdb.pasteur.fr/klebsiella/klebsiella.html) by amplification of seven housekeeping genes of K. pneumoniae isolates. The primers for gapA, infB, mdh, pgi, phoE, rpoB and tonB were manufactured from Integrated DNA Technologies Inc. (California, USA). MLST PCR cycle conditions were as follow: denaturation at 94°C for 2 minutes, followed by 35 cycles of initial denaturation at 94°C for 20 seconds, annealing for all genes at 50° expect for gapA (60°) and tonB (45°) for 30 seconds and extension at 72°C for 30 seconds and a final extension at 72°C for 5 minutes in PCR Thermal Cycler (Bio-Rad Inc., USA).¹⁹

Conjugation Assay

Plasmid transferability was determined by performing conjugation assay using E. coli (ThermoFisher™, USA) as recipient cells.²⁰ Initially, LB media and LB broth (HIMEDIA^®, India) were prepared, autoclaved, and pour into two different sets of plates and culture tubes. In the first set of plates and tubes, only LB media and LB broth were poured without adding anything and labeled. In other sets of culture tubes, LB broth was supplemented with DAP (diaminopimelic acid, 0.3 mM), ceftriaxone (30 µg/mL) and meropenem (10 µg/mL) and labeled. A fresh K. pneumoniae suspension was prepared in LB broth having the antibiotics by incubating over nightly. Fresh recipients E. coli suspension was prepared in plain LB broth by incubating over nightly. In a tube, 50+50 µL of both cells were mixed by repeated pipetting to make a ratio of 1:1 and 100 µL of mixed cell suspension was poured on LB agar plates containing meropenem (10 µg/mL) and labeled as “conjugation plate”, kept at 37 °C for 24 hours. Moreover, PBS suspension of the bacterial colony was used for further confirmation of the transconjugants via PCR using specific primers against the bla_NDM-1 gene.

Statistical Analysis

The differences in the percentage/proportions of K. pneumonia bacteria isolated from different sources were tested by using “prop. Test” function in R software. This test is used for testing the null that the proportions in each group are the same (R Core Team, 2017).

Results

Distribution of K. pneumoniae According to the Respective Sample Source

The findings of the study revealed that the overall prevalence of confirmed carbapenem-resistant K. pneumoniae was 9.4% (13/138). The highest prevalence of 10.7% (7/65) was recorded from abattoir/wastewater samples followed by 10% (4/40) prevalence from milk and fecal samples While 6 % (2/33) prevalence was found in veterinary sludge samples (Table 1). Statistically, the results from Pearson’s chi-squared test statistic (using prop. test in R software) showed that the difference between proportions or percentage of occurrence of K. pneumoniae isolated from seven different sources (shown in Table 1) is non-significant (p = 0.96, Chi-squared statistics = 1.35, df = 7).

Resistance Profile of the Isolates

For the 13 carbapenem-resistant isolates, the disc diffusion assay showed that all the isolates 100% (13/13) of K. pneumoniae were resistant to all the generations of cephalosporins as well as to meropenem. The resistance to ciprofloxacin, tetracyclines, and trimethoprim-sulfamethoxazole were 76.9% and 46.1% and 38.46% respectively as shown in Table 3.

Table 3 Distribution of MICs of Carbapenem-Resistant Klebsiella pneumoniae (KP) Isolates Belonging to Various Multilocus Sequences Typing (MLST)

Distribution of ARGs

After phenotypic characterization, all the carbapenem-resistant K. pneumoniae isolates were subjected to PCR for the detection of various ARGs. Among various ESBL genes bla_TEM was detected in all (100%; 13/13) the isolates followed by bla_SHV (92.30%; 12/13) and bla_CTXM-1 (84.6%; 11/13). Among various MBL genes bla_NDM-1 was detected in 92.30% (12/13) isolates followed by bla_OXA-48 84.61% (11/13), while bla_KPC was detected in 01 (7.69%) isolate only. Whereas, bla_IMP, bla_VIM, and bla_GIM were not detected in any isolate of the study. Other detected ARGs include, qnrA 0%, qnrB 53.84% (7/13) and qnrS 46.1% (6/13), tetA 30.7% (4/13), tetB 23% (3/13), sul1 and sul2 30.7% (4/13). Details of the ARGs distribution in K. pneumoniae isolates along their respective PCR product size were given in Table 2. The distribution of different Beta-lactamases genes among carbapenem resistant K. pneumoniae isolates is shown in Table 4.

Table 4 Co-Existence of Different β-Lactamases and Metallo-β-Lactamases Encoding Genes Among Carbapenem-Resistant Klebsiella pneumoniae (KP) Isolates

Confirmation of Transconjugants

Bacterial colonies that appeared on LB agar plates supplemented with meropenem (10 µg/mL) were considered as transconjugants. Additionally, bla_NDM-1 was detected in all the transconjugants subjected to colony PCR. All the transconjugants displayed a strong resistance pattern against meropenem, as all the transconjugants showed a MIC of 128 µg/mL, which is significantly higher than the interpretative breakpoint given by CLSI.

Prevalent Sequence Types of K. pneumoniae in Veterinary Sample Sources

MLST results revealed that overall, three different STs of K. pneumoniae were detected in the study which includes ST 29, ST11 and ST 258. Among these STs, ST 29 was found in 53.84% (7/13) isolates followed by ST11 which was found in 38.46% (5/13) isolates whereas, single isolate (7.69%; 1/13) of the study showed the allelic profile of ST258. It was found that ST29 was the most prevalent ST in abattoir wastewater, as all the isolates 100% (3/3) from abattoir wastewater displayed the allelic profile of ST29. While both isolates 100% (2/2) from fecal samples belong to ST11. The single isolate of K. pneumoniae which showed the allelic profile of ST258 was isolated from poultry sludge (Table 3).

Discussion

During the last few years, gram-negative bacteria displayed a significant increase in the resistance against beta-lactam antibiotics, because of different plasmid-mediated ESBL genes, present in the Enterobacteriaceae family especially in E. coli and K. pneumoniae. The emergence of resistant K. pneumoniae due to the abuse of antibiotics in livestock farms and veterinary settings is a serious public as well as livestock health concern; it may ultimately be disseminated to the human via various environmental niches.²¹

Pakistan is an agriculture-based country, 70% of people directly or indirectly are involved with agriculture particularly livestock farming. Intense modern livestock farming practices force them to use antibiotics as feed additives, which ultimately associated with the development and dissemination of antibiotic-resistant superbugs. In this context, the prevalence of carbapenem-resistant K. pneumoniae (CRKP) has been well reported from various regions of the world.^3,7 However, data regarding the prevalence of CRKP from Pakistan is very limited. Here, in the present study, we demonstrated the prevalence of carbapenem-resistant K. pneumoniae in veterinary settings and waste from Pakistan.

Findings revealed that overall 9.4% (13/138) prevalence of CRKP was observed in various samples from veterinary settings and waste. The prevalence was higher in wastewater samples compared to the other samples source (Table 1). Yet no such study has reported the prevalence of CRKP in veterinary settings and waste from Pakistan. However, the prevalence of CRKP has been well documented in Veterinary settings and waste in the world.^4,22 Livestock farms have also been recognized as a reservoir of ARGs and their dissemination in Jiangsu Province, China.²³ Similarly, some studies from India also reported that animal milk and meat, etc. are the potential source of ESBL producing K. pneumoniae.^24,25 In this study, we also have isolated CRKP from milk and slaughterhouse waste. The possible explanation of this high prevalence is that antibiotics are aberrantly used in veterinary practice here in Pakistan and as a growth promoter in food animals, which ultimately steer the evolution of antibiotic-resistant bacterial strains. Several studies have documented that food-producing animals are the possible source or reservoir for the dissemination of resistant bacterial strains or ARGs to humans.^26,27 While, another possible reason for the high prevalence of CRKP in milk may be due to mastitis, as bovine mastitis is a substantial factor for the growth and colonization of bacterial pathogens including K. pneumoniae.²⁶ In the past, a study reported the detection of ESBL producing K. pneumoniae in the milk of healthy cows.²⁸

Different ARGs (Table 2) have been detected in the isolates of the study. Particularly, a total of 12 (92.3%) isolates were positive for the bla_NDM-1 gene and 11 (84.6%) isolates were positive the bla_OXA-48. The occurrence of the bla_NDM-1 gene in food-producing animals has already been reported form different parts of the world²⁹ but no such report is published from Pakistan. Additionally, several studies stated that usually bla_NDM-1 gene is located on mobile a genetic element that carry few additional ARGs as well which helps in the dissemination of bla_NDM-1.³⁰ The Indian subcontinent is recognized, as a reservoir of bla_NDM-1 harboring Enterobacteriaceae. Moreover, many studies have also demonstrated the distribution of carbapenem-resistant Enterobacteriaceae in livestock from all across the globe.³¹ On the other hand, the incidence of bla_NDM-1 harboring K. pneumoniae has only been well documented in clinical and hospital settings from Pakistan.^32,33 It is very tough to compare our results with local findings, as according to our best knowledge yet no study has reported the prevalence of bla_NDM-1 harboring K. pneumoniae in veterinary settings and waste from Pakistan.

All the CRKP isolates were subjected to conjugation assay to check the transferability of the bla_NDM-1 gene. As expected, it was observed that bla_NDM-1 was transferable to the donor E. coli cells, indicating that irrespective of the ST, all the bla_NDM-1 was located on conjugative plasmids and a significant source of horizontal gene transfer (HGT). The range of conjugation efficiency was 1.0 x 10⁻⁶ and 3.5 x 10⁻⁶ per transconjugant which also suggested the probability of HGT. Different sample niches especially veterinary sludge and wastewater may have high bacterial density and there would be a significant chance for horizontal gene transfer of bla_NDM-1 to various bacterial strains present in these sources. The same recommendations have been made in the past that resistance genes may be disseminated through bacterial strains via mobile genetic elements and plasmids. Additionally, detailed molecular studies have been reported that K. pneumoniae strains may disseminate different ARGs particularly bla_NDM-1 and bla_OXA-48 via HGT.^6,34

To know the sequence types of the isolates MLST was performed. During MLST three discrete STs of K. pneumoniae were observed which include ST29 (46.1%), ST11 (38.5%) and ST 258 (7.7%). In this study, ST29 in a leading sequence type which is quite interesting. The same results have been reported from Saudi Arabia, where they found that ST29 was the predominant ST of carbapenem-resistant K. pneumoniae in Riyadh.³⁵ Incidence of K. pneumoniae ST29 in veterinary products has been reported from various regions, a study conducted in Ghana reported the occurrence of ESBL producing K. pneumoniae ST29 in local and imported poultry meat.^36,37 Several studies have described the transmission of ESBL harboring Enterobacteriaceae from food animals and their products to humans. Detection of similar clones of ESBL- harboring E. coli and K. pneumoniae in animals and humans provides indirect proof of such cross species-transmission.³⁸ In this study, we have reported ST11 from veterinary settings of Pakistan for the very first time, before that ST11 has only been documented in clinical and hospital settings in Pakistan.³³ Whereas, ST11 has been recognized as the most prevalent ST in Asia particularly in China and Taiwan.^37,39 Detection of K. pneumoniae ST258 in the present study is also worrisome, as ST258 is one of the most detected STs of K. pneumoniae in the world.^40,41 Few factors contribute substantially to the global expansion of ST258 are virulence genes, type IV secretion system, type IV pilus, and type-III restriction-modification system.⁴²

Conclusion

To best of our knowledge, this is the first study from Pakistan which demonstrated that veterinary settings and waste are the potential sources of carbapenem-resistant K. pneumoniae. In Pakistan, sanitation facilities are inadequate, which is a considerable risk factor associated with the dissemination of CRKP to the community. Strong surveillance and monitoring policy are required to estimate the exact burden of this public and livestock health menace which would be beneficial to curtail this health concern.