Phenotypic & genotypic study of antimicrobial profile of bacteria isolates from environmental samples.

Background & objectives: The resistance to antibiotics in pathogenic bacteria has increased at an alarming rate in recent years due to the indiscriminate use of antibiotics in healthcare, livestock and aquaculture. In this context, it is necessary to monitor the antibiotic resistance patterns of bacteria isolated from the environmental samples. This study was conducted to determine the phenotypic and genotypic profile of antimicrobial resistance in Gram-negative bacteria isolated from environmental samples.
Methods: Two hundred and fifty samples were collected from different sources, viz. fish and fishery products (99), livestock wastes (81) and aquaculture systems (70), in and around Mangaluru, India. Isolation, identification and antimicrobial profiling were carried out as per standard protocols. The isolates were screened for the presence of resistance genes using PCR.
Results: A total of 519 Gram-negative bacteria comprising Escherichia coli (116), Salmonella spp. (14), Vibrio spp. (258), Pseudomonas spp. (56), Citrobacter spp. (26) and Proteus spp. (49) were isolated and characterized from 250 samples obtained from different sources. A total of 12 antibiotics were checked for their effectiveness against the isolates. While 31.6 per cent of the isolates were sensitive to all the antibiotics used, 68.4 per cent of the isolates showed resistance to at least one of the antibiotics used. One-third of the isolates showed multidrug resistance. Maximum resistance was observed for ampicillin (43.4%), followed by nitrofurantoin (20.8%). Least resistance was seen for carbapenems and chloramphenicol. PCR profiling of the resistant isolates confirmed the presence of resistance genes corresponding to their antibiotic profile. Interpretation & conclusions: This study results showed high rate of occurrence of antimicrobial resistance and their determinants in Gram-negative bacteria isolated from different environmental sources.

The development and spread of antibiotic resistance among bacteria affecting human health are most challenging problems of the modern world. The genetic determinants responsible for conferring resistance are often carried in self-transmissible mobile elements such as conjugative plasmids, gene cassettes in integrons and transposons, which are transferred between bacterial species, resulting in transmission of resistance to other species1.

The development of resistance to a specific antimicrobial compound is influenced by several environmental factors and it has been reported that the organisms isolated from environment with high faecal contamination can easily acquire resistance to the common antimicrobial drugs2. In addition, the non-therapeutic use of antimicrobials in animal husbandry, aquaculture, agriculture, poultry and piggery has increased the incidence of multidrug resistance forms in the environment34. The presence of antibiotic resistance in Gram-negative bacteria has been well documented including their occurrence in aquaculture5 and livestock6. Among the Gram-negative bacteria, the prevalence of antimicrobial resistance has been reported to be high in Escherichia coli7, Salmonella spp.8, Vibrio spp.9, Pseudomonas spp.10, Citrobacter spp.11 and Proteus spp.12. It is important to understand and monitor the antibiotic resistance patterns of human pathogenic bacteria persisting in the environment. Hence, the main objective of this study was to investigate the phenotypic and genotypic profile of antimicrobial resistance in Gram-negative bacteria isolated from environmental samples.

Material & Methods

A total of 250 samples from different sources, viz. 99 from fish and fishery products comprising 61 of fish/shellfish and 38 of oyster/clam/molluscs; 81 from livestock wastes comprising poultry (23), piggery (14) and cattle wastes (44) samples and 70 from aquaculture systems comprising fish farm water (40) and pond sediment (30), in and around Mangaluru, India, were aseptically collected fortnightly during the study period (2011-2014). The samples were subjected to isolation of associated bacteria by culture-based conventional methods13. Typical colonies from the selective plates were sub-cultured onto Luria Bertani (L-B) agar (HiMedia Laboratories Pvt. Ltd., Mumbai) and identified using standard biochemical tests, viz. Gram staining, motility, cytochrome oxidase, catalase, oxidation fermentation test, urease and triple sugar iron agar1415. E. coli, Salmonella spp. and Vibrio spp. were further confirmed by single-step PCR using species-specific primers (Table I)161718. Identified isolates were preserved in 30 per cent glycerol L-B broth and stored at −80°C for further studies.

Table I

Primers used for the confirmation of suspected Escherichia coli, Salmonella enterica and Vibrio spp. isolates

Isolate	Gene	Sequence 5’- 3’	Size (bp)	Tm (°C)	Reference
Escherichia coli	uidA	AAAACGGCAAGAAAAAGCAG	146	60	16
		ACGCGTGGTTACAGTCTTGCG
Salmonella	invA	GTGAAATTATCGCCACGTTCGGGCAA	284	64	17
		TCATCGCACCGTCAAAGGAACC
	hns	TACCAAAGCTAAACGCGCAGCT	156	60	18
		TGATCAGGAAATCTTCCAGTTGC
Vibrio	16S rRNA	GTAAATTATCGCCACGTTCGG	321	55	16
		AACGGCAAGAAAAAGCAGTG

Tm, melting/annealing temperature of primers

Antibiotic susceptibility test: Antibiotic susceptibility tests were performed for all the isolates using the disc diffusion method described by Bauer et al19 as per the Clinical and Laboratory Standards Institute guidelines20. Twelve common antibiotics (in μg) namely nalidixic acid (30), tetracycline (30), co-trimoxazole (25), ciprofloxacin (5), chloramphenicol (30), ampicillin (10), gentamicin (10), nitrofurantoin (300), imipenem (10), meropenem (10mc), cefotaxime (30) and piperacillin-tazobactam (100/10) (HiMedia) were used for antibiotic profiling. ATCC 25922 E. coli culture was used as a standard quality control strain for AST for Enterobacteriaceae groups.

Detection of antibiotic resistance determinants by PCR: The isolates showing resistance to particular antibiotics were selected and screened for the presence of antibiotic resistance determinants by PCR using the primers listed in Table II21222324. Genomic DNA was extracted from the bacterial culture by cetyl-trimethyl ammonium bromide method25. Using genomic DNA as a template, PCR was carried out in 30 μl reaction mixture containing 10× buffer (100 mM Tris-HCl, p H 8.3, 20 mM MgCl2, 500 mM KCl, 0.1% gelatin), 200 mM of dNTPs (dATP, dTTP, dGTP and dCTP), 10 pmol each of forward and reverse primers and 1.0 unit of Taq DNA polymerase enzyme (HiMedia) in a MJ-Research Thermo Cycler (PTC-200, Bio-Rad, USA). The PCR conditions included initial denaturation at 94°C for 5 min, followed by 35 cycles with each cycle consisting of denaturation at 94°C for 60 sec, annealing for 60 sec at an optimized temperature depending on the primer set used and extension at 72°C for 30 sec. The final extension was set at 72°C for 10 min.

Table II

Primers used for detection of different antibiotic resistance genes

Antimicrobials	Resistance genes	Forward and reverse primer 5’- 3’	Size (bp)	Present in samples	Reference
Tetracycline	tetA	TTGGCATTCTGCATTCACTC	494	Yes	21
		GTATAGCTTGCCGGAAGTCG
	tetB	CAGTGCTGTTGTTGTCATTAA	571	Yes
		GCTTGGAATACTGAGTGTTAA
	tetC	CTTGAGAGCCTTCAACCCAG	418	Yes
		ATGGTCGTCATCTACCTGCC
	tetD	GCAAACCATTACGGCATTCT	546	Yes
		GATAAGCTGCGCGGTAAAAA
	tetE	TATTAACGGGCTGGCATTTC	544	Yes
		AGCTGTCAGGTGGGTCAAAC
	tetG	GCTCGGTGGTATCTCTGCTC	550	Yes
		CAAAGCCCCTTGCTTGTTAC
	tetL	CATTTGGTCTTATTGGATCG	488	No	22
		ATTACACTTCCGATTTCGG
	tetM	GTTAAATAGTGTTCTTGGAG	657	Yes
		CTAAGATATGGCTCTAACAA
	tetS	TGGAACGCCAGAGAGGTATT	660	Yes
		ACATAGACAAGCCGTTGACC
Sulphonamides	sul I	TTTCCTGACCCTGCGCTCTAT	425	Yes	21
		GTGCGGACGTAGTCAGCGCCA
	sul II	CCTGTTTCGTCCGACACAGA	435	Yes
		GAAGCGCAGCCGCAATTCAT
	sul III	ATGAGCAAGATTTTTGGAATCGTAA	792	Yes
		CTAACCTAGGGCTTTGGATATTT
Chloramphenicol	cat1	AACCAGACCGTTCAGCTGGAT	549	Yes
		CCTGCCACTCATCGCAGTAC
	cat2	AACGGCATGATGAACCTGAA	547	Yes
		ATCCCAATGGCATCGTAAAG
	cat3	ATCGGCATCGGTTACCATGT	531	No
		ATCCCCTTCTTGCTGATATT
	cmlA	GGCCTCGCTCTTACGTCATC	662	Yes
		GCGACACCAATACCCACTAGC
	cmlB	ACTCGGCATGGACATGTACT	840	No
		ACGGACTGCGGAATCCATAG
	floR	ATGACCACCACACGCCCCG	1,213	No
		AGACGACTGGCGACTTCTCG
Quinolones	qnrA	ATTTCTCACGCCAGGATTTG	516	Yes	23
		GATCGGCAAAGGTTAGGTCA
	qnrB	GATCGTGAAAGCCAGAAAGG	469	Yes
		ACGATGCCTGGTAGTTGTCC
	qnrS	ACGACATTCGTCAACTGCAA	417	Yes
		TAAATTGGCACCCTGTAGGC
Ampicillin	blaTEM	CTCACCCAGAAACGCTGGTG	569	Yes	24
		ATCCGCCTCCATCCAGTCTA
Cefotaxime	blaCTX-M	ACGTTAAACACCGCCATTCC	356	Yes
		TCGGTGACGATTTTAGCCGC

Sequencing of antibiotic-resistant determinants: The amplified PCR products (antibiotic-resistant determinants) were purified using PCR Purification Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol and were outsourced for capillary sequencing. The obtained sequences were analyzed by BLAST () and antibiotic resistance genes database for their homology with the database sequences and the confirmed sequences were submitted to the GenBank.

Results

Five hundred and nineteen bacterial strains were isolated and identified to the genus level by conventional methods using a battery of biochemical tests. The details of the isolates identified from the different sources are given in Table III. Molecular confirmation of E. coli, Vibrio spp. and Salmonella spp. was done by PCR using primers namely uidA for E. coli, invA and hns for Salmonella and 16S rRNA for Vibrio (Fig. 1).

Table III

Total number of Gram-negative bacterial strains isolated from different non-human sources

Source	Number of samples	Escherichia coli	Vibrio spp.	Salmonella spp.	Pseudomonas spp.	Citrobacter spp.	Proteus spp.	Total
Fish and fishery products (fish/shrimps and clams/oysters/mussels)	99	45	169	8	18	6	13	259
Livestock wastes (poultry, piggery and cattle farm)	81	61	7	6	17	10	18	119
Environmental samples (farm water and pond sediment)	70	10	82	-	21	10	18	141
Total	250	116	258	14	56	26	49	519

Fig. 1

Gel-electrophoresis of PCR amplified products of Escherichia coli and Salmonella isolates. Lane1: 100 bp DNA ladder; lane 2: uidA gene-positive control (E. coli); lane 3: uidA gene-negative control (E. coli); lanes 4 & 5: uidA gene-positive E. coli isolates; lane 6: positive control for invA gene (Salmonella); lane 7: negative control for invA; lanes 8 & 9: invA gene-positive Salmonella isolates; lane 10: positive control for hns gene (Salmonella); lane 11: negative control for hns gene; lanes 12-13: hns gene-positive Salmonella spp. isolates.

Antimicrobial susceptibility test: Among the 519 isolates tested for antimicrobial susceptibility, only 31.6 per cent (164 isolates) showed susceptibility to all the antibiotics tested. While 68.4 per cent (355 isolates) were resistant to at least one of the antibiotics, 33.7 per cent (175 isolates) displayed multidrug resistance (resistance to more than one antibiotic). As shown in Fig. 2, maximum resistance was observed for ampicillin (225 isolates, 43.4%) followed by nitrofurantoin (108 isolates, 20.8%), nalidixic acid (71 isolates, 13.7%), cefotaxime (62 isolates, 11.9%), tetracycline (58 isolates, 11.2%), co-trimoxazole (38 isolates, 7.3%), ciprofloxacin (30 isolates, 5.8%), gentamicin and piperacillin/tazobactam (28 isolates, 5.4%), imipenem (18 isolates, 3.5%), meropenem (16 isolates, 3.1%) and chloramphenicol (15 isolates, 2.9%).

Fig. 2

Antibiotic resistance pattern of all isolates against 12 different antibiotics used in the study. NA, nalidixic acid; TE, tetracycline; COT, co-trimoxazole; CIP, ciprofloxacin; C, chloramphenicol; AMP, ampicillin; GEN, gentamicin; NIT, nitrofurantoin; MRP, meropenem; CTX, cefotaxime; PIT, piperacillin/tazobactum.

The drug resistance patterns of the isolates according to the source of collection are presented in Fig. 3. Among isolates obtained from fishery products and environment samples, 47.5 per cent showed resistance to ampicillin and 18 per cent to nitrofurantoin, suggesting maximum resistance to these two antibiotics. Among isolates from livestock wastes, maximum resistance was observed for nitrofurantoin (33.6% of the isolates) followed by ampicillin (28.6% of the isolates) and tetracycline (22.7% of the isolates).

Fig. 3

Source-wise representation of antibiotic resistance pattern of the isolates to 12 antibiotics. Abbreviations are as given in Fig. 2.

Detection of antibiotic resistance genes: The antibiotic-resistant determinants associated with resistance were detected by PCR. Of the 58 isolates showing resistance to tetracycline, 45 (77.6%) harboured one or more than one tetracycline-resistant genes (tetA, tetB, tetC, tetD, tetE, tetG, tetM and tetS). Remaining 13 isolates did not carry any of the tet genes tested even though these were phenotypically resistant. Among 38 isolates resistant to co-trimoxazole, 12 (31.6%) harboured at least one of the sul genes (sul1, sul2, sul3) and the 26 isolates did not harbour any of the tested genes (Fig. 4). Two hundred and twenty five ampicillin-resistant isolates were tested for the presence of blaTEM, the gene responsible for the resistance. However, only eight isolates (3.6%) showed the presence of this gene. Of the 62 cefotaxime-resistant isolates, only five (8%) had the blaCTX-M gene conferring the resistance trait. Among the 15 isolates resistant to chloramphenicol, five (33.3%) carried one of the resistance genes (cat1, cat2 and cmlA), but none of these showed the presence of cat3, cmlB and floR (Fig. 5). Of the 71 nalidixic acid-resistant isolates, nine (12.7%) carried either qnrA, qnrB or qnrS.

Fig. 4

Detection of PCR amplified products of antibiotic resistance genes. Lane 1: 100 bp DNA ladder; lane 2: tetA (492 bp); lane 4: tetB (571 bp); lane 6: tetE (544 bp); lane 8: sul1 (425 bp); lane 10: sul2 (435 bp); lane 12: sul3 (792 bp); lane 14: 500 bp DNA ladder and lanes 3, 5, 7, 9, 11, 13: negative controls.

Fig. 5

Gel-electrophoresis of PCR amplified products of chloramphenicol-resistant isolates. Lane 1: 100 bp DNA ladder; lane 2: cat1 gene-positive control; lane 3: cat1 gene-negative control; lanes 4-5: cat1 gene-positive isolates; lane 6: positive control for cat2 gene; lane 7: negative control for cat2 gene; lanes 8-9: cat2 gene-positive isolates; lane 10: positive control for cmlA gene; lane 11: negative control for cmlA gene; lane 12-13: cmlA gene-positive isolates.

Further, sequencing of the PCR products of antibiotic resistance genes from representative isolates revealed 98 per cent identity with the existing antibiotic resistance gene sequences in the database. The GenBank accession numbers of these sequences are given in Table IV.

Table IV

List of antibiotic resistant gene sequences submitted to GenBank with accession number

Title	Accession number
Escherichia coli isolate T1 tetracycline resistance protein class A (tetA) gene, partial cds. (EC12)	KF240812.1
E. coli strain T2 tetracycline resistance protein class B (tetB) gene, partial cds. (EC23)	KF240811.1
Salmonella sp. T1a tetracycline resistance proteins class A (tetA) gene, partial cds. (S131)	KF240813.1
Salmonella sp. S1a sulphonamide resistance protein (sul1) gene, partial cds. (S131)	KF240817.1
E. coli strain S2 sulphonamide resistance protein (sul2) gene, partial cds. (EC11)	KF240815.1
E. coli strain S3 sulphonamide resistance protein (sul3) gene, partial cds. (EC10)	KF240814.1
E. coli strain S1 sulphonamide resistance protein (sul1) gene, partial cds. (EC10)	KF240816.1

Discussion

The results of antibiotic susceptibility test revealed that the number of isolates showing resistance to one or more antibiotics was on the rise, suggesting the high occurrence of antibiotic resistance in animal, wastewater, soil and other natural environments. Regardless of their source of isolation, resistance to three antibiotics, namely ampicillin, nitrofurantoin and tetracycline, was most frequently observed. Although ampicillin resistance was highest among all the antibiotics used.

It has been shown that isolates from fish and shrimp farms have widespread resistance to nitrofurantoin26. In this study, approximately 20 per cent of the isolates showed resistance to nitrofurantoin. Similarly, 18 per cent of the isolates obtained from fish farms and related products showed resistance to this antibiotic, and the results were in agreement with that of an earlier study27. The resistance to nitrofurantoin could be either due to the inhibition of nitrofurantoin reductase or due to the nucleotide changes (mutation) in nfsA and nfsB encoding oxygen-insensitive nitro reductase. Nitrofurantoin and its metabolites have zero tolerance in fishery products, and yet the high occurrence of resistance to nitrofurantoin observed in the study indicating the use of this antibiotic in aquaculture and other environments.

Resistance to tetracycline has been reported frequently from environmental samples28. In the present study, 11.2 per cent of the isolates exhibited phenotypic resistance to tetracycline, which was consistent with earlier studies that reported tetracycline resistance in 15-56 per cent of the isolates28. Resistance to tetracycline is more frequently observed in poultry isolates because it is widely used as growth promoters in poultry production29. More than 60 per cent of poultry isolates showed resistance to tetracycline in this study.

Skockova et al30 had earlier reported that tetA and tetB were the most common genes responsible for resistance to tetracycline. In our study, 77 per cent of the phenotypically resistant isolates showed the presence of one or more tet genes. Although the remaining 23 per cent did not carry any of the tet genes tested in the study despite being phenotypically resistant, the resistance in such isolates could be due to other mechanisms such as enzymatic inactivation or target modification31.

In this study, 7.3 per cent of the total isolates showed moderate resistance to co-trimoxazole. The sulphonamide resistance genes (sul1, sul2 and sul3) are known to be associated with class 1 integrons responsible for capturing and excision of genes during site-specific recombination events32, which further facilitates the emergence of multidrug resistance among bacterial pathogens. Similar to the pattern seen for other antibiotics, the phenotypic and genotypic association was inconsistent for co-trimoxazole with only 31.6 per cent of the phenotypically confirmed isolates showing the presence of one of the three sul genes. The absence of sul in the remaining isolates suggests the possibility of some new genes or resistance mechanism to confer resistance to co-trimoxazole.

The antibiotic resistance pattern obtained for chloramphenicol in this study was low profile i.e.,2.9 per cent. The results of genotypic characterization showed inconsistency with 20 per cent carrying both type 1 and type 2 cat genes and 13.3 per cent harbouring cmlA. However, none of the isolates were positive for type 3 cat, cmlB and floR. The absence of resistance genes was observed in 33.3 per cent of the phenotypically resistant isolates, suggesting the possibility of other mechanism(s) such as overexpression of efflux pumps, mutations or modifications in the target sites or decreased outer membrane permeability contributing to chloramphenicol resistance3334.

The resistance to β-lactam group of antibiotics such as ampicillin, cefotaxime, imipenem, meropenem and piperacillin/tazobactam was also analyzed in this study. Except for ampicillin, which showed maximum resistance, the percentage of resistance was relatively less with 3.5 and 3.1 per cent of isolates showing resistance to imipenem and meropenem, respectively. In our study, 20 per cent of the isolates showed resistance to nalidixic acid and five per cent towards ciprofloxacin. The usual cause of resistance to quinolones is due to point mutation/mutations in the quinolone resistance determining regions (QRDR) or due to presence of active efflux or outer membrane permeability. In addition, the plasmid-mediated quinolone resistance (PMQR) has also been reported. In the present study, nalidixic acid- and ciprofloxacin-resistant isolates carried qnrA, qnrB and qnrS indicating that quinolone resistance was acquired through plasmid-mediated determinant. Although point mutations in the QRDR are the main reason for resistance to quinolone/fluoroquinolones, the occurrence of PMQR genes cannot be neglected since these play a major role in the transmission of resistance among bacterial isolates35. The bacterial isolates showed similar pattern of resistance to cell wall synthesis inhibitors such as meropenem and imipenem. Although the susceptibility of the majority of the isolates to carbapenems was encouraging, the small percentage of resistance observed should be viewed seriously.

In conclusion, the results of this study on the pattern of resistance to antimicrobials in environmental samples highlighted the importance of continuous vigilance on the distribution of multidrug-resistant human pathogens in the environment. High rate of multidrug resistance among bacterial isolates from environmental samples suggested the indiscriminate use of antimicrobials in various sectors of animal husbandry. Judicious use of antibiotics, use of alternative bio-control approaches or development of pathogen-specific antimicrobial agents could help in combating antimicrobial resistance.