MATERIALS AND METHODS

Ethical approval

This study was conducted in accordance with the ethical principles and research guidelines established by Kalasin University, Thailand. The study protocol was reviewed and approved by the Research and Development Institute Committee of Kalasin University. All procedures complied with applicable national regulations and institutional standards for microbiological and environmental research. Laboratory procedures involving opportunistic pathogenic bacteria, including Klebsiella pneumoniae, were performed in accordance with institutional biosafety regulations and standard containment practices for handling potentially pathogenic microorganisms. No experimental manipulations or invasive procedures involving live animals were performed during this study, as only environmental water samples were collected from commercial aquaculture ponds.

Study period and location

This study was conducted from July 2024 to June 2025 at the Kalasin University Excellent Laboratory for Agricultural and Food Product Standard Testing Center, Faculty of Agricultural Technology, Kalasin University, Thailand.

Study design

A cross-sectional observational study was designed to investigate the occurrence of ARB and clinically important ARGs in freshwater prawn aquaculture ponds. Water samples were collected quarterly during both rainy and dry seasons to account for seasonal variation. Bacterial isolates were recovered using selective culture techniques and subsequently identified by 16S rRNA gene sequencing. Antimicrobial susceptibility profiles were determined using the agar disk diffusion method, and PCR assays were employed to detect clinically relevant β-lactamase genes. Descriptive and inferential statistical analyses were performed to evaluate resistance patterns and the prevalence of MDR bacteria.

Study area description

Water samples were collected from earthen aquaculture ponds culturing giant freshwater prawn (_Macrobrachium _rosenbergii) across three districts in Kalasin Province, Northeastern Thailand (Figure 1).

Figure 1: Geographic locations and spatial distribution of giant freshwater prawn (Macrobrachiumrosenbergii) aquaculture ponds sampled in Mueang, Yang Talat, and Huai Mek districts of Kalasin Province, Northeastern Thailand. Geographic coordinates of the sampling sites are shown to provide geographical and epidemiological context for the study area.

A total of nine earthen aquaculture ponds (three ponds per district) were included in this study. Water sampling was conducted quarterly between July 2024 and June 2025 to capture both rainy and dry seasons. One water sample was collected from each pond at each sampling event, yielding a total of 36 water samples over the study period. Pond selection was based on active cultivation of _M. _rosenbergii and the farmer's willingness to participate in the study.

Organic load was estimated qualitatively based on turbidity and sediment accumulation observed during sampling. No major disease outbreaks were reported during the sampling period, although occasional reductions in water quality and prawn survival were noted by farmers. Ponds were selected using convenience sampling based on accessibility, active freshwater prawn cultivation, and farmers’ willingness to participate in the study.

Basic environmental parameters, including water temperature (28–33°C), pH (7.2–8.1), and dissolved oxygen levels (4.5–6.8 mg/L), were monitored during sampling using portable field meters. Average pond depth ranged from approximately 1.2 to 1.8 m.

Sample collection and bacterial isolation

Water samples were collected aseptically from earthen aquaculture ponds culturing _M. _rosenbergii across three districts in Kalasin Province, Northeastern Thailand. Samples were collected into sterile 1-L polyethylene bottles and transported on ice to the laboratory for immediate processing within 6 h to preserve microbial viability [12].

To selectively recover environmental ARB associated with local freshwater prawn farming systems, a multi-medium isolation strategy supplemented with 50 µg/mL ampicillin (AMP) was employed. This selective approach was designed to enhance the detection of resistant bacterial populations that may be exposed to antimicrobial selective pressure in earthen aquaculture ponds.

Serial 10-fold dilutions of each water sample were prepared using sterile 0.85% normal saline. A 1-mL aliquot of each dilution (10⁻¹–10⁻²) was plated onto selective media containing 50 µg/mL AMP to isolate potential ARB. AMP was selected as the primary screening antibiotic because β-lactam antibiotics are among the most commonly used antimicrobial agents in freshwater aquaculture systems and are frequently associated with the emergence of MDR environmental bacteria. In addition, AMP supplementation has been widely used to selectively recover Gram-negative ARB, particularly Aeromonas spp. and members of the family Enterobacteriaceae, from aquatic environments.

The media used included Luria-Bertani agar (HiMedia Laboratories Pvt. Ltd., Mumbai, India) for general bacterial growth, blood agar base (HiMedia) for detecting hemolytic activity, Streptococcus selection agar (HiMedia), Salmonella-Shigella agar (HiMedia), and thiosulfate-citrate-bile-sucrose agar (HiMedia) for Vibrio spp. Following incubation, CFU counts on AMP-supplemented media ranged from approximately 1.2 × 10² to 4.8 × 10⁴ CFU/mL depending on the sampling site and season. Approximately 10–20 colonies displaying distinct morphological characteristics were screened from each sample before selecting representative isolates for further molecular characterization.

Plates were incubated at 37°C for 18–24 h. Colonies exhibiting distinct morphological characteristics were selected and purified through repeated subculturing. Colony morphology, including shape, margin, elevation, and pigmentation, was recorded. Gram staining was performed to classify bacterial groups [13].

Retrospective interviews with pond operators indicated occasional use of oxytetracycline and amoxicillin-based formulations for disease prevention and management. However, complete dosage records and treatment frequencies were inconsistently documented and therefore could not be quantitatively analyzed.

Molecular identification of bacteria

DNA was extracted from pure bacterial isolates using a standard boiling method. The 16S rRNA gene region was subsequently amplified using universal primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′), as previously described [14]. PCR reactions were carried out in a total volume of 25 µL using Taq DNA polymerase (Vivantis Technologies Sdn. Bhd., Selangor, Malaysia) and optimized annealing temperatures.

PCR amplification was performed in a thermal cycler under the following conditions: initial denaturation at 95°C for 5 min, followed by 35 amplification cycles consisting of denaturation at 95°C for 30 s, annealing at 55–60°C for 30 s depending on primer specificity, and extension at 72°C for 1 min, followed by a final extension at 72°C for 10 min. Positive control strains carrying known resistance genes and nuclease-free water as a negative control were included in each PCR assay.

PCR products were visualized on 1% agarose gels stained with ethidium bromide and subsequently submitted for sequencing using the Macrogen sequencing service (Macrogen Inc., Seoul, South Korea). Resulting sequences were analyzed using the Basic Local Alignment Search Tool against the GenBank database of the National Center for Biotechnology Information (NCBI) using a sequence identity threshold of ≥99% [15]. Phylogenetic trees were constructed using Clustal Omega alignment and the neighbor-joining method [16].

Antimicrobial susceptibility testing

Antimicrobial susceptibility testing was performed using the agar disk diffusion method on Mueller-Hinton agar according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) [17]. Inocula were adjusted to the 0.5 McFarland standard and evenly spread on agar plates using sterile cotton swabs.

Commercial antibiotic disks (Oxoid, Thermo Fisher Scientific, Basingstoke, United Kingdom) were used for susceptibility testing and included AMP (10 µg), vancomycin (VAN, 30 µg), azithromycin (AZM, 15 µg), streptomycin (STR, 25 µg), rifampicin (RIF, 5 µg), and chloramphenicol (CHL, 30 µg).

Plates were incubated at 37°C for 24 h, and inhibition zones were measured in millimeters. Results were interpreted as resistant (R ≤15 mm), intermediate (I = 16–20 mm), or susceptible (S ≥21 mm) according to CLSI breakpoints.

All antimicrobial susceptibility tests were performed in triplicate. Escherichia coli American Type Culture Collection (ATCC) 25922 and Staphylococcus aureus ATCC 25923 were used as quality control strains in accordance with CLSI M100, 31st edition [17].

Detection of ARGs

Plasmid DNA was extracted from selected antibiotic-resistant isolates using the GF-1 Plasmid DNA Extraction Kit (Vivantis Technologies Sdn. Bhd., Selangor, Malaysia) according to the manufacturer's instructions. Extracted plasmid DNA was quantified using a NanoDrop™ spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Targeted screening for clinically relevant carbapenemase and β-lactamase genes in environmental isolates was conducted to assess the potential dissemination of AMR determinants within freshwater aquaculture ecosystems within a One Health framework.

PCR assays were conducted to detect six resistance genes associated with β-lactamase production, namely blaTEM_, blaSHV, blaOXA, blaKPC-2, blaNDM-1,_ and blaIMP, using specific primers and previously published protocols [18, 19]. PCR products were resolved on 1.5% agarose gels, stained with ethidium bromide, and visualized under ultraviolet illumination.

The present study focused primarily on plasmid-associated resistance determinants. Chromosomal DNA-associated resistance genes were not specifically investigated.

Statistical analysis

Descriptive statistics were used to summarize bacterial prevalence and AMR profiles. Differences in resistance frequencies among bacterial species and sampling districts were evaluated using Fisher's exact test or chi-square analysis where appropriate. Statistical significance was defined as p < 0.05. The 95% confidence interval (CI) for MDR prevalence was calculated using the Wilson method.