Introduction

Sea turtles have an evolutionary history of more than 200 million years; they are distributed in all the tropical and subtropical oceans, both in the neritic and oceanic zones (Abdelrhman et al. 2016). This group includes green turtles (Chelonia mydas) that can be found in the seas of 80 countries, including Mexico (Bass et al. 2006, Sarti-Martínez et al. 2012). Nowadays, various human activities such as illegal and incidental fishing, tourism, depredation, have contributed to a higher than 50% decrease in the population’s global decrease in less than three generations. This population reduction has led the International Union for Conservation of Nature (Seminoff 2004) to list green turtles as endangered (Sarti-Martínez et al. 2012). In addition to human activity and predation, infectious diseases are another important factor that have contributed to the decrease of the population of sea turtles (Alfaro et al. 2006; Wallace et al. 2010). A related aspect is the loss of immune balance and the development of different pathologies in captive turtles (Alfaro et al. 2006; Muñoz et al. 2013). Ulcerative stomatitis is a disease that affects turtles; it is characterized by an inflammatory process associated with oral microbiota (Mehler and Bennett 2003). The disease can involve inflammation of the oral cavity, the respiratory tract or the gastrointestinal tract; if not treated, it can lead to septicemia, pneumonia and even death (Mehler and Bennett 2003). During captivity, factors such as diet, overcrowding and the conditions of the facilities can stress the animals, which can lead to immunosuppression. As a consequence, the alterations in the microbiota result in the development of pathologies such as ulcerative stomatitis (Mehler and Bennett 2003; Chuen-Im et al. 2010). There are few studies on this disease in captive individuals. Glazebrook et al. (1993) report the presence of the disease in 58.6% of captive green turtles and C. caretta turtles in the neonatal and juvenile stages. The same study reported that when the condition was associated with obstructive rhinitis and pneumonia, the mortality rate was 70%. A longitudinal study conducted for three consecutive years in a sea turtle conservation center found that the prevalence of the disease was 45.8% (Chuen-Im et al. 2010). No studies on this subject have been carried out in Mexico; thus, the purpose of this work was to identify which bacteria are more frequently associated with ulcerative stomatitis in captive green turtles in Mexico.

Materials and methods

This study was conducted in juvenile green turtles at Xcaret Ecological Park, located in Playa del Carmen, Quintana Roo, Mexico (86° 55′ W, 21° 20′ N). The park has a warm and subtropical climate with an average annual temperature of 27 °C and an average humidity of 74%. The approximate age of the individuals was 4 to 5 months. This population is part of the conservation and release program of Xcaret, under which the turtles are released when they reach 15 months of age. The individuals were kept in cement tanks of 4.57 m in length; 4.34 m in width and 1.05 m deep, each holding 2000 L of water, with continuous water flow of 2.8 l/min. The average temperature (26.38 °C), salinity (36 ppm) and pH (8.2) were monitored. Twice a day, the turtles under study were provided with food containing 35% protein, 5% fiber and 3.5% fat (Sea Turtle Food, Silver Cup, Salt Lake City, Utah 84,157, USA); the amount of food provided was 3% of the live weight of each turtle.

Isolation and identification of bacteria

Oral mucosa samples were taken with sterile swabs from 20 clinically healthy turtles and ten animals with clinical symptoms suggestive of ulcerative stomatitis. After taking the samples, the swabs were placed in Stuart transport medium and kept refrigerated at 4 °C until processed in the laboratory. Samples were seeded by streaking directly onto blood agar and McConkey media and incubated overnight in microaerobiosis at 37 °C. The isolated colonies were analyzed by Gram stain; Gram negative and positive bacteria were tested for oxidase and catalase, respectively. The selected bacteria were maintained on “A” agar (0.7% nutrient medium, 0.1% yeast extract, 0.2% glycerol, 200 mM of K2HPO4 and 9 mM of KH2PO4) for up to 2 months prior to performing biochemical identification with the automated Vitek system (BioMérieux). GNI and GPI cards (BioMerieux, France) were used for gram-negative bacilli and gram-positive cocci, respectively, according to the following procedure: each of the isolated colonies was seeded by closed streaking on trypticase soy agar medium and incubated at 37 °C for 18–24 h. A bacterial suspension was prepared from the culture (at 0.5 in the McFarland standard) using sterile MilliQ water; this suspension was then used to inoculate the identification cards of the automated Vitek system.

Statistical analysis

As statistical analysis a Fisher’s exact test (Level of significance p < 0.05) was performed to compare growth frequency of bacterial genera isolated between the group of turtles with ulcerative stomatitis and the control group.

Results

Bacteria isolation and identification

Eighty-three colonies were isolated from healthy turtles, and 82 colonies from turtles with ulcerative stomatitis. Bacterial colonies were maintained in Agar “A” at room temperature for 2 months where we recovered 56 colonies (67%) from healthy turtles and 51 colonies (62%) from turtles with ulcerative stomatitis. Bacteria identified from healthy turtles included Enterococcus faecium in 11 animals (55%), Kocuris kristinae in four animals (20%) and Serratia marcescens in three animals (15%). Other bacteria, such as Enterococcus gallinarum, Citrobacter braakii, Pseudomonas aeruginosa, Ochrobactrum anthropi and Micrococcus luteus were identified in one individual, not necessarily the same (Table 1).

Table 1 Genus and species of bacteria isolated from healthy turtles and turtles with ulcerative stomatitis
Full size table

For turtles with ulcerative stomatitis, we identified Staphylococcus lentus in six individuals (60%); Citrobacter braakii in five (50%), and Enterococcus faecium in four (40%), the last one having the highest frequency of isolation among the two groups of individuals of the study, without significant difference in Fisher’s test (p 0.74). Additionally, Klebsiella pneumoniae and Pseudomona aeruginosa were identified in two turtles, while Enterococcus gallinarum, Citrobacter youngae, Serratia marcescens and Micrococcus luteus in only one individual (Table 1). Bacteria such as S. lentus, C. youngae and K. pneumoniae were isolated only from turtles with stomatitis (Table 1).

The analysis species identified in both groups of turtles showed that E. faecium, S. lentus, and C. braakii isolated in the group of turtles with stomatitis were associated with other microorganisms isolated in low numbers (Table 2). In contrast, in the group of healthy turtles the species identified tended to be unique and showed no association with other bacteria. The mean prevalence of bacteria positive animals in the stomatitis group was twice that of healthy. S. lentus, C. youngae and K. pneumonia were unique to stomatis animals whereas O. anthropic was unique to healthy.

Table 2 Association between bacterial genera in turtles with ulcerative stomatitis
Full size table

The comparison between sick and healthy animals showed that only S. lentus (60%) showed significant differences (p 0.0041), and C. braakii (50%) although with a high frequency in the sick turtles did not show statistical differences (p 0.06) between both groups of turtles (Table 1).

Discussion

There appeared to be a shift in turtles with stomatitis from a flora dominated by Enterococcus faecium, S. marcescens, and K. kristinae to one dominated by S. lentu and C. braaky. Thus, the source of the infectious agent may be the extant microbiota of the turtles (Ferronato et al. 2009; Chuen-Im et al. 2010). To be able to determine which microorganisms is part of the microbiota of green turtles, the present study included a higher number of healthy turtles (20) than of sick animals (10). In this regard, only E. faecium was identified with high frequency in both groups of turtles (Table 1). Although it was not present in all the animals, Fisher’s exact test showed that the frequency of the E. faecium isolate in both groups of animals was not significantly different (p 0.74), which suggests that this microorganism regularly colonize these animals. However, the observations suggest that the etiology of the condition could be related to both S. lentus and C. braakii, because when an association analysis between the more common isolated microorganisms was performed, the association S. lentus and C. braakii showed significative statistical differences (Table 2).

Other studies have been reported the isolation of Staphylococcus spp., Citrobacter spp. and Enterococcus spp. in cloacal samples of young turtles (Nowakiewicz et al. 2015) and of S. lentus, E. faecium and C. freundii in the oral microbiota of river turtles in Brazil (Ferronato et al. 2009). There are also reports of the isolation of Citrobacter spp. and Staphylococcus spp. in sea turtles (C. caretta, C. mydas, and D. coriacea) with infectious forms of fibrotic gastritis, enteritis and granulomatous hepatitis (Oros et al. 2004). The isolation of S. lentus has been reported in other species of captive turtles (suborder Cryptodira) with conjunctivitis (Di Ianni et al. 2015). Enterococcus spp. has been found in association with respiratory, joint and skin infections that developed during the rehabilitation period of L. kempii turtles kept in temporary captivity (Innis et al. 2014); as well as in fibropapillomatosis lesions in green turtles, and in bladder, liver, lung, and muscle lesions in C. caretta (Aguirre et al. 1994; Fichi et al. 2016; de Morais et al. 2011).

The reason why only some animals get sick may be related to alterations in their immune system. However, biotic and abiotic factors such as temperature, the presence of microorganisms and contamination can influence the immunity of captive animals under stress (Zimmerman et al. 2010; Muñoz et al. 2013). Studies in newborns and infant show that in the early stages of life, the immune system has immunological regulators that seem to favor the development of the individual and its colonization by microbiota; this can be related to the higher susceptibility to infections at these stages of life (Zhang et al. 2017). This behavior could also take place in turtles, and could be part of the explanation of why different authors find higher susceptibility to infectious diseases in young turtles (Santoro et al. 2006; Chuen-Im et al. 2010).

The results obtained allow us to conclude that E. faecium is part of the microbiota of green turtles and that S. lentus and C. braakii are potentially responsible for the ulcerative stomatitis symptoms observed because these bacteria dominated the oral flora of sick animals. Understanding the mechanistic factors of mucosal immunity that permit such bacterial shifts, the development of ulcers, and the actual role of these bacteria in the genesis of these ulcers might be fruitful future avenues of study. Such knowledge would aid captive management of this endangered species by reducing the burdent of ulcerative stomatitis which continues to be significant problem in captive green turtles (Muñoz et al. 2013).