Sea turtles have been classified as endangered species in Appendix I of the Convention of International Trade in Endangered Species (CITES) since 1982. In recent years, the dramatically decreased number of sea turtles in nature has been causing concern in several countries. As a result, several conservation campaigns have been established, attempting to increase the number of animals through a variety of strategies. In Thailand, the Royal Thai Navy successfully conducted an early intervention program for sea turtle conservation. Two species of sea turtles, green turtles (Chelonia mydas) and hawksbill turtles (Eretmochelys imbricata), were included in the program, which collected eggs from the nests and incubating them in places safe from egg harvest. After hatching, juvenile sea turtles are raised in captivity until about 4 months of age and then released into the sea. Approximately 2000–3000 captive-raised juvenile green turtles and hawksbill turtles have been released annually. Yet despite this success in rearing large numbers of turtles, the conservation program has encountered disease problems, with the prevalence of sick turtles increasing every year.

There are few studies documenting diseases in free-living or captive sea turtles (Wiles and Rand 1987; Glazebrook and Campbell 1990a, b; Raidal et al. 1998; Work and Balazs 1998; Torrent et al. 2002; Greer et al. 2003; Orós et al. 2004, 2005), and reports of diseases in hatchling and juvenile sea turtles are rare (Glazebrook et al. 1993). Generally, the diseases observed in sea turtles are caused by parasites and micro-organisms, including fungi, viruses, and bacteria. Compared with all reported diseases, bacterial infection seems to be one of the major causes of disease in sea turtles with high rates of morbidity and mortality (Orós et al. 2004). Several studies have suggested that bacterial infection is associated with secondary diseases of both free-living and captive sea turtles that display other diseases such as fibropapillomatosis or parasitic infection (Aguirre et al. 1994; Raidal et al. 1998; Work et al. 2003). However, a study of disease in hatchling and juvenile green turtles and loggerhead turtles (Caretta caretta) in Australia reported only bacterial disease (Glazebrook et al. 1993). Specifically, 3 bacterial diseases (ulcerative stomatitis, obstructive rhinitis, and pneumonia) were found to be responsible for mortality rates of up to 70% in farmed and oceanarium-reared juvenile green turtles and loggerhead turtles (Glazebrook et al. 1993).

The purpose of head-starting juvenile sea turtles is to raise the sea turtle hatchlings and release them when their size is large enough to reduce predation, usually at the age of between 3 and 5 months. However, aspects such as diet, number of turtles in captivity, material, and design of captive environments are important to consider because these factors can affect turtle health and susceptibility to disease. Furthermore, assessment of health and disease of captive turtles should be regularly performed. The aims of this study were to survey and characterize microbes associated within juvenile green turtles and hawksbill turtles from the Sea Turtle Conservation Centre of Thailand (STCCT). The results obtained will provide microbiological information important for understanding disease mechanisms and improving sea turtle conservation and recovery plans.

Materials and Methods

From June 2005 to August 2007, a total of 53 turtles from the STCCT early intervention program (30 juvenile green turtles and 23 juvenile hawksbill turtles of 2–3 months of age) were collected immediately after death. The curved carapace length and width were measured on each carcass. Gross necropsies examining for the presence of lesions or abnormal symptoms both externally and internally were then performed, and samples were immediately prepared for microbiological and histological investigations.

For transmission electron microscopy (TEM) investigation, tissues from infected organs were dissected and fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer pH 7.2. After incubating for 24 h, the samples were postfixed in 1% osmium tetraoxide at 4°C for 1–2 h. All specimens were then washed 3 times with 0.1 M phosphate buffer pH 7.2 followed by dehydration through a graded acetone series, prior to being embedded in spurr resin. The samples were stained with 0.5% uranyl acetate and lead citrate and observed under TEM (JEOL 1230) operating at 80 kV.

After being washed with 0.9% normal saline buffer, the infected tissues were fixed in Bouin's fluid for 48 h. The samples were dehydrated with a graded ethanol series followed by 50%, 70%, 80%, 90%, and absolute ethanol, processed and embedded in paraffin. The specimens were sectioned at 6-µm thickness before being stained with haematoxylin and eosin (H&E). The stained tissue was histologically observed using a light microscope.

For microbiological analysis, the tissues with macroscopic signs of lesions were cleansed with 0.9% normal saline buffer and dried with sterile gauze before cross-cutting at the lesion area. Then bacterial samples were collected by carefully swabbing on the cross-cut surface and streaking on nutrient agar (NA), potato dextrose agar (PDA), baird parker agar base, eosin methylene blue (EMB) agar, and thiosulphate citrate bile salt agar (TCBS). Both NA and PDA were supplemented with 1% or 3% NaCl (marine agar). All cultures were incubated at 25°C aerobically and examined after 24 h. Each distinctive colony, based on colony characteristics and growth pattern, was isolated and treated as a separated organism. Categorization of microorganism was performed using Gram reaction and morphology of the bacterial colonies. Pure cultured colonies were identified by standard bacterial taxonomy procedures described in the eighth edition of Bergey's Manual of Determinative Bacteriology (Buchanan and Gibbons 1974), such as carbohydrate use, glucose fermentation, glucose O/F medium, IMViC test, gelatin use, triple sugar iron (TSI) and hydrogen sulfide production (H₂S), urea, catalase test, oxidase reaction, motility test, salt requirement and tolerance, hemolytic properties, antibiotic sensitivity, lactose fermentation.

Results

Postmortem examination of turtles revealed gross lesions at carapace, fore- and hind flippers, tail and anus, oral cavity, stomach, intestine, kidney, and liver (Table 1). When considering the prevalence of lesions in each year between 2005 and 2007, only 2 diseases were found in the turtles in 2005. The primary lesion was ulcerative stomatitis (65%), whereas necrotizing hepatitis was the second most prevalent lesion observed (35%). For 2006, in addition to the diseases observed in 2005, complex diseases were also found such as traumatic ulcerative dermatitis, gastrectasis caused by fluid gas, enteritis, impacted intestine, and ulcerative shell disease. In 2007, the prevalence of primary lesions found in the animals changed, with a lower incidence of ulcerative stomatitis (32.1%) and increased hepatitis (46.4%) (Table 2). From a total of 72 lesions observed in 53 turtles, the oral cavity was the region with the most frequently observed lesions (45.8%) (Table 2) containing yellow caseous material adhering to mucosal surfaces (Fig. 1a). The second most frequently observed lesion was necrotizing hepatitis (31.9%). Most livers showed the symptoms of green or grey multifocal necrosis, whereas the surrounding tissue appeared normal with a red color. In some cases, the whole organ had turned yellow and hard, and the part of the intestinal tract adjacent to the affected liver showed similar changes (Fig. 1b). In another case, the intestinal tract was full of gas with yellow mucoid fluid (Fig. 1c). Similar symptoms were also observed in the stomachs. Skin and appendages were the next organs that showed most frequently observed lesions (13.9%). No macroscopic parasites were visible.

View Figure

• Download as Powerpoint
• Download Figure

Table 1 Lesions or diseases observed in 53 juvenile sea turtles: Chelonia mydas and Eretmochelys imbricata.

View Fullscreen

Table 2 Prevalence of diseases from the total of 72 lesions observed in 30 juvenile green turtles and 23 juvenile hawksbill turtles.

View Fullscreen

Next, we examined the lesion tissues using TEM and histological technique. The results showed that the tissues were destroyed in association with numerous bacteria where the size of bacteria ranged from 0.5–2.0 µm. No hyphae of fungi were observed. Bacterial presence was not found in the nonlesioned tissues of juvenile sea turtles. Ultrastructural examination of carapace and oral cavity tissue sections revealed a severe diffusion of bacteria in both the acute abscesses and purulent ulcerative glossitis (Fig. 2a, b). In Figure 2b, the nuclei of affected cells appeared normal despite the presence of intracytoplasmic bacterial infection. The lesion tissues were infected with numerous rod-shaped, curved rod-shaped, and coccus-shaped bacteria inside the cytoplasm of cells (Fig. 2a–c). Additionally, pseudohyphae, a characteristic of Candida albicans, were seen in oral cavity lesions, co-infecting with bacteria (Fig. 2d). For the appendages (tail and anus), tissue showed signs of destruction with a number of cells containing intracytoplasmic bacteria. The size of bacteria ranged from 0.5–2 µm (Fig. 3a). For the case of hepatitis (Fig. 3b), the tissue appeared to be heavily degenerate (on the left) with clumps of bacteria in the cell cytoplasm (arrows). The infected cell had disorganization of intracytoplasmic structure with many spaces; the cell boundary was also difficult to identify (Fig. 3b). Figure 3c is an H&E-stained liver section showing signs of necrosis and inflammation, such as granulocyte infiltration (arrows), tissue destruction, multinucleated giant cell (star). Moreover, distribution of bacterial clumps was observed in the sections (data not shown). In Figure 3d, clumps of bacteria were also found in the surrounding tissue of lesions. These bacteria were within membrane-bound vesicles in the cytoplasm (arrows).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Microorganisms were swabbed from carapaces, skin lesions (tails and anuses), oral cavities, kidneys, and livers, and cultured yielding bacterial colonies. On a few occasions, yeast was isolated from skin and hind flipper and oral cavity lesions. When cultured on rice infusion agar, pseudohyphae, similar to that seen in Figure 2d, were obtained, implying that these yeasts might be C. albicans. No growth of fungi was observed from any sample. Gram staining of all isolates demonstrated that these bacteria were either gram-positive or -negative bacteria. Similar to that observed by TEM, these bacteria were either coccus- or rod-shaped bacteria. Table 3 lists the prevalence of bacteria isolated from different lesions of juvenile green turtles and hawksbill turtles. Vibrio alginolyticus was the most frequently isolated bacteria type (26.8%), following by beta-haemolytic Staphylococcus spp. (19.6%). Also, several opportunistic and potentially pathogenic bacteria were cultured from the lesions, which included Citrobacter freundii (10.7%), Micrococcus spp. (10.7%), Edwardsiella spp. (8.9%), Aeromonas hydrophila (7.1%), Aureobacterium spp. (5.4%), Vibrio parahaemolyticus (3.6%), Streptococcus group C (3.6%), and Corynebacterium spp. (3.6%). Additionally, a number of gram-positive and -negative bacteria were also found that could not be identified. In addition to being the most frequently observed lesion, ulcerative stomatitis appeared to contain the highest diversity of bacteria (8 of 10 strains identified; Table 3). The most and second-most commonly isolated bacteria were beta-haemolytic Staphylococcus spp. and V. alginolyticus, respectively. The second highest diversity of lesion-associated bacteria was found in necrotizing hepatitis where V. alginolyticus was the most commonly isolated bacterium.

Table 3 Identified bacteria isolated from lesions in juvenile sea turtles no.  =  number of examined lesions, N  =  number of time isolated, n  =  the total number of identified isolates.

View Fullscreen

Discussion

In this study, a total of 53 turtles, including 30 juvenile green turtles and 23 juvenile hawksbill turtles, were necropsied immediately after death. Electron microscopic and microbiological study provided strong evidence of bacterial infection in these turtles. One type of yeast was isolated from the oral cavity and appendage lesions, occurrence of pseudohyphae seen on both H&E-stained section and rice infusion agar implying that they are C. albicans, one frequently isolated yeast species from clinical specimens. Candida albicans infection of sea turtles has been previously documented by Orós et al. (2003). Candidiasis associated with intestinal lesions caused by foreign body, a fishing line, was found in a loggerhead sea turtle (C. caretta), which implies that water quality may be an important factor in the disease incidence. There were no signs of fungal or parasitic infections. It was found that all 53 turtles had been affected by bacterial infection in different organs, including skin and appendages, shells, oral cavity, liver, and kidney. Survey of diseases in both captive and free-living sea turtles has been conducted in various locations such as in Australia (Raidal et al. 1998; Glazebrook and Campbell 1990a, b; Glazebrook et al. 1993), Hawaii (Work and Balazs 1998), and Canary Islands (Orós et al. 2004, 2005). In those reports, bacterial infections have been documented as an important case of disease in sea turtles and, in many cases, associated with high rates of morbidity and mortality (Orós et al. 2004). Our data are in agreement with this.

Ulcerative stomatitis was the most frequently found lesion in the oral cavity (45.8%) (Table 2), making it the most prevalent cause of disease of both captive and free-living sea turtles (Glazebrook and Campbell 1990a, b; Orós et al. 2004). Moreover, microbiological observation indicates that this organ exhibited the highest diversity of bacterial infection. Perhaps, the oral cavity is the organ having heaviest exposure to the environment, leading to high chance of bacterial infection. The second most frequently found lesion was necrotizing hepatitis (31.9%). The high prevalence of such lesions observed in this study is similar to findings in Canary Islands sea turtles (Orós et al. 2004). This suggests that necrotizing hepatitis is another important disease caused by bacteria in sea turtles. Ulcerative skin lesions including shell diseases (13.9%) were the most frequently found lesions after necrotizing hepatitis. These lesions are commonly found in farmed turtles which mainly is caused by biting (Glazebrook and Campbell 1990a; T. Chuen-Im, pers. obs.). We found that the number of such lesions correlated with the number of sea turtles in the basins and subsequent crowding. In other words, the high density of sea turtles in the basins resulted in the biting behavior, leading to bacterial infection in the lesions.

Ultrastructural, histological, and microbiological examinations revealed bacterial presence in all observed lesions. The results shown in Tables 1 and 3 indicate that some of the turtles had systemic lesions. This may be caused by septicaemia where bacteria entered the circulation through the primary lesions. Identification of bacteria isolated from lesion material showed a wide range of gram-negative and gram-positive bacteria, including C. freundii, A. hydrophila, V. alginolyticus, V. parahaemolyticus, Micrococcus spp., Staphylococcus spp. (beta-haemolytic), Streptococcus group C, Edwardsiella spp., Corynebacterium spp., and Aureobacterium spp. These bacteria have been reported as opportunistic invaders and normal flora in either free-living or captive green turtles (Buchanan and Gibbons 1974; Glazebrook and Campbell 1990a; George 1997; Santoro et al. 2006).

The most common type of isolated bacteria was V. alginolyticus, found in the lesions in the oral cavity, liver, intestine, and stomach. From our preliminary data, this bacterium was commonly cultured from sea water collected from the basin, indicating that it is part of the normal flora in that environment or could be present as a surface or gastrointestinal commensal. Vibrio alginolyticus has been reported as a common commensal in cloacal or nasal flora of healthy green turtles, with and without fibropapillomatosis (Aguirre et al. 1994; Santoro et al. 2006). However, some reports found that it was commonly found in association with various lesion types in farmed or free-living green turtles in Australia (Glazebrook and Campbell 1990a,b), loggerhead turtles stranded in the Canary Islands, Spain (Orós et al. 2005), or in the Bermuda Aquarium (Wiles and Rand 1987), and bacteraemia in free-ranging green turtles with fibropapillomatosis (Work et al. 2003). Thus, this bacteria type may be one potentially important pathogen for the turtles in the Thai conservation center.

Micrococcus spp. and Staphylococcus spp. are known as normal flora in marine environments. Also, they were noted as normal flora in skin of farmed green turtle (Glazebrook and Campbell 1990a) and nasal and cloacal passages of nesting green turtles (Santoro et al. 2006).

The bacterium C. freundii has been reported as an opportunistic environmental organism in free-living green turtles with spirorchid cardiovascular fluke infection in Australia (Raidal et al. 1998) and as a primary pathogen in freshwater turtles as well as implicated in shell diseases. In the Raidal et al. (1998) study, the infection of C. freundii was found in liver, lung, and kidney of the examined turtles. They concluded that C. freundii infection is a secondary disease, which causes systemic illness and death in parasitically infected sea turtles. The bacterium was also found in Hawaiian green turtles both with and without fibropapillomas (Aguirre et al. 1994). However, no significance of this bacterium in causing disease was noted in that report.

Several reports on sea turtle diseases have documented that bacterial infections are secondary diseases of both free-living and captive sea turtles, associated with other diseases, such as fibropapillomatosis and parasitic infection, which possibly result from immunosuppression in the turtles (Aguirre et al. 1994; Raidal et al. 1998; Work et al. 2003). Subsequently, bacterial infection may be the cause of illness and death (Raidal et al. 1998). Unlike previous reports, no evidence of fungal, viral, and parasitic infection was observed in juvenile green turtles and hawksbill turtles in this study. It might be possible that the cause of death in these juvenile sea turtles was mainly from bacterial infection. Like other juvenile animal species, the juvenile turtles more than adults may be susceptible to bacterial infection partly because development of the cell-mediated defense mechanisms is still in process. As a result, they have a low ability to fight the infections, leading to sickness and subsequently death. Moreover, the immune system of the hosts also may be inhibitory affected by stressing conditions such as density of animals in captivity or inadequate farm conditions, making the animals more susceptible to infection from either opportunistic bacteria or pathogens. There are other factors required for explanation of the pathogenesis in the affected animals. Currently, evaluation of the water resource and the water in the habitat is performed including quantitative bacteriology and antibiotic resistance patterns of the isolated bacteria.

The results from microbiological isolation demonstrated that mixed bacterial infections were present in single lesions in examined organs. One explanation is that various bacteria may function synergistically to cause disease (Gorge 1997). Some of the isolated bacteria reported herein are known as normal flora or opportunistic invaders; some are potentially pathogenic and have ability to cause disease or death in animals. However, there are a few bacteria that are the most common pathogens frequently isolated from diseased sea turtles (e.g., Escherichia coli, Pseudomonas spp.; George 1997) that were not recovered in this study. This emphasizes the significance of establishing normal flora associated with sea turtles in each region. This fundamental information is essential for the conservation of protected animals in terms of prevention of potential disease and rehabilitation of sick animals. Also, it contributes to a better understanding in the management of this endangered animal not only in Thailand but also in other regions.