Staphylococcus aureus and methicillin-resistant Staphylococcus aureus in juvenile green turtle (Chelonia mydas) carcasses, rearing seawater, feed and their antibiotic resistances

Sample collection

The experiments were conducted during years 2022–2024. All samples were kindly provided by the Sea Turtle Conservation Center of Thailand (STCCT), Sattahip, Chonburi Province. The STCCT, operated under the Royal Thai Navy, has established an early intervention program for sea turtle conservation (Chuen-Im et al., 2010). According to the program, green turtle eggs will be collected from nests to incubate them in places safe from humans and animals. After hatching, juvenile turtles will be raised in captivity and released back to the sea at the age of about 4 months. Ten juvenile green turtles, raised at the STCCT were collected immediately after death and stored on ice during transportation to the laboratory. Gross necropsies were performed by examining the internal and external abnormal symptoms or the presence of lesions prior to microbiological examination. Two seawater sources were used in the study, i.e., rearing water and water supply for juvenile turtle rearing. Rearing seawater samples were collected from cement tanks that were used to keep the collected juvenile turtle samples. The water supply for turtle rearing was incoming coastal seawater at a distance of about 10–20 m offshore. Water sample collection was carried out as previously described (Chuen-Im et al., 2019, 2021). The samples were kept on ice during transportation to the laboratory. Fish fillet samples were collected five times separately and used in microbiological investigation. The fish meat was kept in a sterile plastic bag and placed on ice during transportation. This study was approved by the Silpakorn University Animal Ethics Committee (MHESI 8603.16/4291_2Nov2563 and MHESI 8603.16/0300_18Jan2567) and the Department of Fisheries, Ministry of Agriculture and Cooperatives (the Approval number 3/2566_17Nov2566).

Isolation and identification of Staphylococcus aureus

Isolation of Staphylococcus aureus from the water samples was carried out by spreading 0.1 mL of each sample on baird parker agar (BPA). For juvenile turtle carcasses, internal organs, including the heart, kidney, liver, and contents of the stomach and the small intestine, were used for bacterial examination. Samples were prepared using the aseptic technique. After dissection, the organs were weighed and washed several times with sterile 0.85% (w/v) normal saline buffer before homogenization using a pestle and mortar. The ground tissue was resuspended in 0.85% normal saline buffer at a ratio of 1:10 (w/v) to prepare a 10^–1 dilution. The 10^–1 dilution was further diluted into 10^–2 to 10^–3 before being spread (0.1 mL) on BPA. Similar to the juvenile turtle sample preparation, approximately 10–15 g of fish fillet was homogenized using a pestle and mortar. The ground meat was mixed thoroughly before 1 g sample was resuspended in 0.9 mL 0.85% normal saline buffer to make a 10^–1 dilution. Using the tenfold serial dilution technique, 10^–2 and 10^–3 dilutions were prepared prior to the spreading on BPA. All plates were incubated at 37 °C for 24–48 h. Gray-black colonies were isolated from BPA and then streaked onto nutrient agar (NA) for further pure culture. S. aureus was identified using the following biochemical tests: motility, growth on NaCl, decarboxylase test, hemolysis, oxidase, catalase, urease, oxidation/fermentation, carbohydrate fermentation (glucose, lactose, mannitol, and sucrose), coagulase test, and Gram-staining and bacterial shape observation. All media were purchased from HiMedia Laboratories (Mumbai, India).

Antibiotic susceptibility

The antibiotic susceptibility of S. aureus was determined through the disc diffusion method (Clinical and Laboratory Standards Institute (CLSI), 2015; Chuen-Im et al., 2021). Briefly, bacteria were cultured in nutrient broth at 37 °C overnight and then diluted to a concentration of 10⁸ CFU/mL. The bacterial suspension was swabbed on Mueller Hinton Agar (HiMedia Laboratories, Mumbai, India) and used for antibiotic resistance examination. Discs of 11 antibiotics were used in the study, including beta-lactam (penicillin (10 IU), ampicillin (10 μg), cefazolin (30 μg), and cefoxitin (30 μg); aminoglycosides: streptomycin (10 μg), kanamycin (30 μg), gentamicin (10 μg), amikacin (10 μg), and tobramycin (10 μg); and others: tetracycline (30 μg) and chloramphenicol (30 μg) (HiMedia Laboratories, Mumbai, India). Staphylococcus aureus (ATCC 25923) was used as the drug susceptibility control. All plates were incubated at 37 °C for 24 h. The inhibition zone was determined by measuring the diameter of the clear zone, including that of the disc, and recorded in millimeters. Multidrug-resistant bacteria were defined as bacteria resistant to at least one of the three drug classes.

The identification of MRSA and methicillin-susceptible S. aureus (MSSA) was performed in accordance with the Clinical and Laboratory Standards Institute (CLSI) (2013, 2021). Isolates that showed resistance to 30 μg cefoxitin disc (clear zone ≤21 mm) were reported as MRSA, and MSSA (clear zone >21 mm) was reported when an isolate showed susceptibility to cefoxitin.

Amplification of mecA and femA genes via polymerase chain reaction (PCR)

S. aureus DNA was extracted using a modified protocol from the work of Taechowisan, Mungchukeatsakul & Phutdhawong (2018). Briefly, a pellet of 1.5 mL overnight-cultured bacteria was resuspended in 600 μL Tris-ethylenediaminetetraacetic acid buffer, added with 3 μL 10% sodium dodecyl sulfate and 3 μL 20 mg/mL proteinase K, and incubated at 65 °C for 30 min. The mixture was then mixed with 600 μL phenol/chloroform/isoamyl alcohol (25:24:1) and centrifuged at 12,000 rpm for 5 min. The supernatant was transferred into a new tube, and an equal amount of absolute ethanol was added to precipitate DNA. The DNA pellet was obtained through centrifugation at 12,000 rpm for 5 min before rinsing with 1 mL 70% ethanol and air-drying. Finally, DNA was dissolved with TE buffer, and the concentration was adjusted to 100 ng/mL before storage at −20 °C until required.

To investigate the presence of mecA, which encodes for an alternative penicillin-binding protein PBP2a, and femA, which encodes for a protein precursor in peptidoglycan biosynthesis in S. aureus, in the DNA sample, we set up a mixture containing 1X polymerase chain reaction (PCR) buffer with MgCl₂, 0.5 μL 10 mM dNTP, 0.5 μL of 10 pg forward and reverse primers each, and 1 μL 100 ng/mL DNA template, adjusted with sterile dH₂O to 25 μL. The PCR conditions were initially denatured at 95 °C, 5 min before being subjected to 35 cycles of denaturation at 94 °C for 30 s, annealing at 52 °C, 30 s, extension at 72 °C for 1 min, and final extension at 72 °C for 5 min before holding at 4 °C. The PCR product was analyzed via ethidium bromide-stained 1% agarose gel electrophoresis. Table 1 shows the sequences of the gene-specific primers used in this study.

Statistical analysis

To investigate the significant differences in the antibiotic resistance of S. aureus from the three sample sources, we analyzed the hypothesis test results using R stat for the Friedman test and PMCMRplus packages for the post hoc tests (Pohlert, 2024). Significant differences in antibiotic resistances of S. aureus from seawater, fish fillets, and juvenile turtle carcasses were reported when p-value < 0.01.

Detection of Staphylococcus aureus in seawater supply and juvenile green turtle rearing seawater and their antibiotic resistance

To investigate S. aureus in coastal seawater, which were used as input to the sea turtle containers, and rearing water, we used BPA to screen bacteria from water samples. A total of 240 colonies were randomly selected and streaked on NA to obtain a pure culture. Then, all isolates were identified using biochemical tests. From the total of 240 isolates, the highest identified bacterial number was that of Staphylococcus spp. (171 isolates; 71%) followed by those of Staphylococcus aureus (44 isolates; 18%) and Micrococcus spp. (21 isolates; 9%). The other two bacterial species found were Bacillus spp. and Gram-negative bacteria Proteus sp. (three isolates and one isolate, respectively). All 44 S. aureus isolates were from rearing water, and no S. aureus was isolated from coastal sea water (water supply). Results of the coagulase test indicated that all S. aureus isolates were coagulase positive.

The 44 S. aureus isolates were then examined for their antibiotic resistance to 11 drugs through the disc diffusion approach. From the results, the most frequent resistant antibiotic was penicillin in β-lactam class (23%) (Table 2), followed by chloramphenicol (14%), cefazolin (9%), and finally amikacin (2%). We found two isolates (SA211 and SA321) that were resistant to cefoxitin (Figs. 1B, 1C), which suggests that they may be MRSA compared with MSSA (isolate SAP210 in Fig. 1A).

In this study, 11 drugs belonging to four classes (β-lactam, aminoglycosides, protein synthesis inhibitor tetracycline and chloramphenicol) were used. Most isolates were sensitive to all tested antibiotics (25 isolates; 57%) (Fig. 2). Ten isolates (23%) were intermediate or resistant to one drug, whereas nine isolates were intermediate- and/or resistant to >1 drug. The highest antibiotic resistance number, 7 out of 11 drugs, was observed in isolates SAP215 and SA321. The multidrug-resistant SAP215 isolate was resistant to all aminoglycoside drugs tested (streptomycin, gentamycin, kanamycin, tobramycin, and amikacin) and two protein synthesis inhibitors; tetracycline and chloramphenicol. MRSA isolate SA321 also showed the highest antibiotic-resistant number (seven drugs). In addition to cefoxitin, this isolate was resistant to six other drugs, which belonged to the β-lactam (penicillin, ampicillin, and cefazolin) and aminoglycoside (gentamicin, kanamycin, and tobramycin) classes. The third-highest resistant antibiotic number (four drugs) was detected in another MRSA isolate, SA211, which was resistant to three drugs in the β-lactam class (penicillin, cefazolin, and cefoxitin) and tetracycline.

All 44 isolates that were biochemically identified as S.aureus were further investigated for the presence of peptidoglycan biosynthesis gene femA, and the penicillin-binding protein PBP2a gene, mecA, in their genome through PCR with specific-gene primers. Figure 1D shows the amplified DNA fragments obtained from two MRSA isolates SA211 and SA321. The results of 1% agarose gel electrophoresis showed PCR products with sizes of 450 and 293 bp obtained from both isolates, which indicates the presence of femA and mecA, respectively, in their genome but not in the case of Staphylococcus sp. (Fig. 1D; lane 5). From the 44 isolates, 43 and three gave PCR products from femA- and for mecA-specific primes, respectively.

Observation of S. aureus and MRSA in the feed of juvenile green turtles and their antibiotic resistance

As no S. aureus isolate was found in the sea water input to the rearing containers, we questioned whether the source of the bacteria detected in rearing water was the fish fillet used as feed. To investigate this issue, we used the fish fillet kindly provided by STCCT in isolation and identification. A total of 23 isolates were obtained from fish fillet in five samplings. All isolates were then subjected to disc diffusion and PCR experiments to determine the presence of MRSA. Table 3 shows the isolate numbers of cefoxitin-resistant or -susceptible bacteria and/or the detection of the mecA gene. From the 23 S. aureus isolates, 3 MRSA isolates were cefoxitin resistant and mecA gene positive, and 3 MSSA isolates were cefoxitin sensitive and mecA negative. However, 17 isolates that were positive in mecA-amplified PCR showed susceptibility to cefoxitin, which implies the presence of the mecA gene in their genome.

We further investigated the antibiotic resistance of these isolates to 10 other drugs. As displayed in Table 2, the S. aureus isolated from fish fillet showed resistance to five out of 11 drugs, which can be ranked in descending order as follows: chloramphenicol (39%), penicillin (30%), tetracycline (22%), cefoxitin (13%), and ampicillin (9%). From 23 isolates, five isolates (21%) were susceptible to all antibiotics, whereas the remaining isolates were mainly resistant to one drug and two drugs (eight isolates, 35% each). Finally, two isolates (9%) were resistant to three antibiotics. For MRSA, three isolates were resistant to cefoxitin (13%), and all isolates were resistant to penicillin. No multidrug resistance was observed.

Investigation of S. aureus and MRSA in juvenile green turtle carcasses and their antibiotic resistance

To assess the health risk of juvenile green turtles from S. aureus infection, we investigated the presence of S. aureus and MRSA in juvenile turtle carcasses. Postmortem examination revealed that the primary lesion was ulcerative shell and traumatic ulcerative dermatitis (50%), mostly at the skins and appendages, whereas two out of 10 turtles (samples no. 5 and 7) showed no sign of lesions or abnormal symptoms (Fig. 3, Table 4). Afterward, the soft tissue organs of 10 turtles, which were the heart, liver, kidney, stomach, and small intestine, were used in bacterial isolation and identification. Black colonies with an opaque halo on BPA were further isolated for pure culture before identification using a biochemical test. S. aureus was isolated from seven out of 10 samples, which results in a total of 40 isolates (Table 4).

All turtle-isolated S. aureus were further examined for antibiotic susceptibility to 11 drugs. The results showed that five turtles (50%), which all exhibited lesions or abnormal symptoms, had MRSA (Table 4). When considering the percentage of isolates resistant to each antibiotic (Table 2), most were resistant to penicillin (62%), followed by those resistant to cefoxitin (32%) and then streptomycin (27%; 11 isolates).

Next, we compared the antibiotic resistance profiles of S. aureus from juvenile turtle carcasses with the isolates from rearing water and feed. Statistical analysis demonstrated a significant difference in the antibiotic resistance of S. aureus from the three sample sources (p-value = 0.003905). As provided in Table 5, for isolates with resistance to only one antibiotic, resistance to penicillin or chloramphenicol was observed in all sample sources. Most of the isolates from rearing water and fish fillet were resistant to one or two antibiotics, whereas almost half of the turtle isolates were resistant to >3 antibiotics, with one isolate showing resistance to all antibiotics tested. Moreover, six multidrug-resistant bacterial isolates were detected in juvenile turtle carcasses, and only one isolate was found in reared seawater. Of these, five isolates were MRSA (5 out of 7 isolates; 71%) and isolated from juvenile turtle carcasses. This finding indicates a severe situation of antibiotic resistance in S. aureus infection and a high risk of the health status of juvenile turtles at STCCT.