Mycelial Fungal Diversity Associated with the Leatherback Sea Turtle (Dermochelys coriacea) Nests from Western Puerto Rico
Abstract
This work describes the mycelial fungal diversity associated with leatherback sea turtle nests and eggs from Mayagüez-Añasco Bay Coast (MABC), Puerto Rico. Comparisons are made of conditions previous to leatherback nesting season, during leatherback nesting season, and during nest hatching season. Prior to Dermochelys coriacea nesting season, the fungal community along the MABC showed a normal distribution (p = 0.098) by One-Way ANOVA. We found that Aspergillus was the most frequent genus (0.15), followed by Cladosporium (0.09) and Curvularia (0.08). At the time of oviposition, Penicillium was the most frequent isolate (0.15), followed by Cladosporium (0.11), Aspergillus (0.11), and Fusarium (0.07). No fungi were isolated from nesting leatherback's ovipositor samples. During hatching season, fungal diversity was evaluated from the sand of hatched nests and from failed eggs. Fusarium solani was the most frequent isolate (0.57) from hatched nest sand and was the only species isolated from the interior of failed eggs. A strong positive correlation was obtained between fungal abundance and the number of failed eggs in the nests (r = 0.853, p < 0.001). This was the first attempt to study fungal diversity associated with D. coriacea nests and eggs in Puerto Rico.
Puerto Rico's shorelines are frequently visited by 4 of the 8 globally distributed species of sea turtles. Three of these, the hawksbill turtle (Eretmochelys imbricata Linnaeus), the green turtle (Chelonia mydas Linnaeus), and the loggerhead turtle (Caretta caretta Linnaeus), have hard shells. The leatherback (Dermochelys coriacea Vandelli), unlike the others, lacks a bony shell. The leatherback is the one with the highest nesting occurrence in Puerto Rico but has been listed as an endangered species attributable to habitat contamination, erosion, and predation. Other factors contributing to risk include the loss of eggs attributable to high relative humidity in the nests, relatively low temperatures, and nest infection and colonization by microorganisms such as bacteria, protozoa, and fungi (Eckert and Eckert 1990).
Because the leatherback is an endangered species, effective management of the species requires an understanding of the factors that negatively affect its reproduction and survival. Despite the fact that the leatherback sea turtle has the highest nesting frequency in Puerto Rico, no previous studies have described the diversity and abundance of mycelial fungi that could be affecting their nests. Hatching success is limited by either direct or indirect disturbances such as human intrusion, environmental conditions, fungal and bacterial infections, and predation (Eckert and Eckert 1990; Bjorndal 1995). Some fungi may produce lytic enzymes that could degrade the eggshell (Phillot and Parmenter 2001b, 2006). These researchers conclude that the fungi found in the nests thrive from the nutrients provided by damaged eggs. From this source, mycelial networks extended to viable eggs, covering them completely and consequently causing their loss.
Other possible ways in which the mycelium might cause debilitation and developmental retardation of embryos have been proposed: 1) gas exchange impediment attributable to pore obstruction of the eggshell; 2) fungal spore transfer from the allantois to the embryonic tissue; and 3) calcium depletion (Solomon and Baird 1980). It has been demonstrated that it is not necessary for the mycelium to cover the entire egg to cause damage, but it is enough to cover the egg's north pole, where the exchange takes place, to cause the egg loss (Phillot and Parmenter 2001a). It has been suggested that egg ultrastructure may provide a surface for fungal development. There is evidence of fungi directly associated with leatherbacks eggs from Malaysia (Chan and Solomon 1989), but Phillot and Parmenter (2006) concluded that the egg ultrastructure does not contribute directly to fungal infection in turtles from Australia. They found that the size of the spaces between the inorganic and organic matrices of the eggs does not allow spore diffusion or the penetration of hyphae (Phillot and Parmenter 2006).
Leatherback egg composition has been studied in an effort to provide more information about possible causes of egg loss. The eggshell is composed of an inner organic membrane and an outer inorganic layer of calcium carbonate deposited as aragonite (Chan and Solomon 1989). The main component found in eggs is albumin, comprising nearly two-thirds of the egg and approximately 80% of the mass of every clutch (Wallace et al. 2006). Albumin has antifungal properties, similar to the mucus secreted by the turtle during oviposition (Phillot and Parmenter 2006). Regardless, Fusarium oxysporum, Fusarium solani, and Pseudallescheria boydii have been regularly isolated from the exterior of failed eggs and from embryonic tissue (Phillot et al. 2001, 2004). These species produce enzymes that degrade inorganic and organic components of the sea turtle eggshells and, therefore, penetrate and colonize the egg (Phillot 2004). Fusarium solani has been demonstrated to be pathogenic and responsible for mass mortality of loggerhead sea turtles eggs (Sarmiento-Ramírez et al. 2010). This fungus, along with F. oxysporum, has also been documented from eggs and stillbirths of hawksbill turtles from Brazil (Neves et al. 2015). Another species documented as an emerging pathogen from hatched and unhatched hawksbill eggs from Ecuador is Fusarium falciforme (Sarmiento-Ramírez et al. 2014). All of these Fusarium species are members of the F. solani species complex (FSSC), where some are well-known plant and animal pathogens (Short et al. 2015; Coleman 2016).
METHODS
Data were collected during 3 periods. The first samples were taken between August and December 2008. During this period, mycelial fungal diversity along the Mayagüez-Añasco Bay Coast (MABC) was studied by traditional culturing techniques to establish pre-leatherback nesting conditions. The second sampling period extended from March to June 2009 (leatherback reproductive season). These samples provided information about mycelial fungal diversity associated with the leatherback nests and the ovipositor of nesting females during oviposition. The last sets of samples were taken from the sand and eggs from hatched and failed nests after incubation period, between 60 and 80 d after oviposition.
Assessment of the Fungal Community at the MABC Prior to Leatherback Nesting Season
Three study plots (i.e., Caño La Puente, Tres Hermanos, and Jardines del Caribe) were established at the MABC using data from the last 2 reproductive seasons of the leatherback turtles provided by the DNER (M. Justiniano, pers. comm., 2008). Each plot was subdivided in 3 subplots, where sand cores were collected from 3 different parts (high tide mark, center of the backshore, and hindmost of the backshore) at 3 different depths using a disinfected probe. Sand cores (∼0.6 m in depth) were separated in three fractions: top (T), middle (M), bottom (B) according to its position relative to the nest. Individual sand fractions were placed in sterile plastic bags and processed within 24 hrs.
Dilution plate method was performed for recuperation of fungal colonies from the sand fractions (Davet and Rouxel 2000). A volume of 0.5 ml from each dilution was spread on the surface of Glucose–Yeast Agar (GYA) and Marine Agar (MA) with 100 μg/ml chloramphenicol to inhibit bacterial growth in 100 × 15-mm Petri plates. Plates were incubated at 25°C ± 2°C in duplicate for a period of 5 to 30 d. Fungi were characterized based on morphological features to the level of species when possible.
Statistical analysis of the data collected was performed by 1-way ANOVA using SigmaStat 3.5 software. Plots were analyzed first independently and then as single plot. Shannon's and Simpson's diversity indexes were calculated using BioDiversity Professional (version 2) software (McAleece et al. 1997). Fungal diversity similarity among plots was studied by Bray-Curtis Cluster Analysis.
Assessment of Fungal Community Associated with Leatherback Nests During Nesting Season at the MABC
During this period, samples were taken at the time of oviposition from 16 nesting females. Once the turtle finished digging the nest, a sample of sand from the bottom was collected on a sterile plastic bag using a sterile spoon. A sample from the ovipositor was collected using sterile transport swabs by inserting the swab into the nesting female cloaca before oviposition started. Sand samples were processed as previously described. Swabs containing ovipositor samples were streaked over solidified Marine Agar (MA) plates and incubated for 5 to 30 d at 25°C ± 2°C.
Assessment of Fungal Community Associated with Leatherback Hatched Nests and Failed Eggs at the MABC
After incubation period, hatched nests were opened and samples of the sand associated with the eggs were collected from the bottom of the egg chamber on a sterile plastic bag using a sterile spoon. Hatched and failed eggs were counted to determine hatching success of the nests. Failed eggs were collected in sterile bags and taken to the laboratory for processing following the dilution method.
Eggs were surface-washed with 100 ml of sterile phosphate buffer 1× and agitated for 15 min to remove any fungal propagules. Dilutions from the resulting wash solution were performed from 10−1 to 10−4, and 100 μl of each dilution and stock solution were spread over solidified MA medium on 100 × 15-mm Petri plates in duplicate. After the surface-wash protocol, eggs were superficially disinfected by submerging each one in hydrogen peroxide (H2O2) for 1 min, then in 95% ethanol for 1 min, and followed by 2 washes in sterile distilled water for 1 and 5 min, respectively. This method was developed taking under consideration H2O2 and ethanol disinfecting success on hen eggs (Shane and Faust 1996). Eggs were cut open carefully with a sterile scalpel. Egg content was separated based on thickness, resulting in 2 subsamples: soft content (SC) and thick content (TC). A 10−1 dilution of each sample was prepared by adding 10 ml of the sample to 90 ml of sterile phosphate buffer and agitating for 10 min. Successive extractions of 1 ml from this suspension and then from the following were performed until 10−4 dilutions were obtained. A volume of 100 μl from each dilution was spread over solidified MA medium on 100 × 15-mm Petri plates in duplicate. All plates were incubated at 25°C ± 2°C for 7 to 14 d.
DNA Extraction, PCR Amplification, and Sequencing of Internal Transcribed Spacer (ITS) Region from Fungal Pure Cultures
For molecular characterization, DNA extraction was performed by growing the fungi on Potato Dextrose Broth in 1.5-ml tubes for 7 d at 25°C. Mycelia was then removed from tubes and processed with the Fast DNA® Spin Kit for Soil (MP Biomedicals, USA). DNA fragments containing ITS1 and ITS2, including 5.8S, were amplified and sequenced with primer pair ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) (White et al. 1990). An initial cycle of denaturalization at 95°C for 3 min was followed by 35 cycles of denaturalization at 95°C for 30 sec, annealing at 56°C for 1 min, and extension at 72°C for 2.5 min. A final step of extension at 72°C for 10 min was done. Samples were sent for sequencing to High-Throughput Genomics Unit (HTGU), Department of Genome Sciences, University of Washington. Sequences were edited with Sequencher 4.10 (©Gene Codes Corporation). Nucleotide BLASTN searches were used to compare sequenced obtained against the sequences in the National Center of Biotechnology Information (NCBI) nucleotide databases.
Phylogenetic Analysis of Fusarium solani Isolates from Leatherback Failed Eggs
The program BioEdit version 7.0.9.0 (Hall 1999) was used for alignment of the ITS sequences of F. solani isolates from leatherback failed eggs and selected sequences obtained from the NCBI nucleotide databases. For the external group, a Fusarium staphyleae (Nectria atrofusca) (AF178423) was selected based on a previous phylogenetic analysis of the genus Fusarium (O'Donnell et al. 2008). MrBayes version 3.2.6 (Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003) was used for phylogenetic analysis. Phylogeny was computed using the Kimura 2-parameter method selected by jModeltTest2 (Guindon and Gascuel 2003; Darriba et al. 2012) applying the Bayesian Information Criterion (BIC). Posterior probabilities were computed with 3 million generations.
RESULTS
Assessment of the Fungal Community at the MABC Sites Prior to Leatherback Nesting Season
A total of 85 morphotypes were isolated from sand previous to leatherback nesting season. The most common genera isolated were Aspergillus, Cladosporium, and Curvularia with relative frequencies of 0.15, 0.09, and 0.08, respectively, followed by Penicillium (0.06), Trichoderma (0.06), and Fusarium (0.05). There were no significant differences in fungal diversity among sand cores between plots (p = 0.563), and the fungal population showed a normal distribution (p = 0.098). Based on Shannon's Diversity Index (H′) and Simpson's Diversity (D), Tres Hermanos transect showed the highest mycelial fungal diversity (H′ = 0.852 and D = 0.176), whereas Caño La Puente showed the lowest diversity with H′ = 0.689 and D = 0.384. Similarity among transects' fungal diversity determined by Bray-Curtis Cluster Analysis showed that Caño La Puente and Tres Hermanos transects had a low similarity of 35%, whereas these 2 were 8% similar to Jardines del Caribe. These differences in fungal diversity might be explained in terms of the geographical barrier that the coastal currents that distribute freshwater and sediments north and south of the Río Grande de Añasco river mouth may represent, which separates Jardines del Caribe transect from the other 2. Furthermore, Tres Hermanos and Caño La Puente have a higher human impact than Jardines del Caribe because the first 2 are located at the Tres Hermanos public beach.
Assessment of Fungal Community Associated with Leatherback Nest During Nesting Season at the MABC
A total of 27 morphotypes were isolated during this period. No fungi were isolated from nesting females' ovipositor. Fungi identified include Penicillium (0.15), Cladosporium (0.11), Aspergillus (0.11), and Fusarium (0.07). Identification of all of the other 15 morphotypes was not possible by traditional culturing techniques because no reproductive structures were produced.
Assessment of Fungal Community Associated with Leatherback Hatched Nests and Failed Eggs at the MABC
A total of 39 morphotypes, representing at least 5 genera, were isolated from sand of hatched nests and failed eggs of D. coriacea at the MABC. Most of the fungi isolated were identified within Fusarium (0.57), Aspergillus (0.13), Penicillium (0.10), Cladosporium (0.08), and Scedosporium (0.05) genera. All 5 genera were isolated from sand samples and from the surface of failed eggs. Only F. solani was isolated from the interior of failed eggs. Correlation between fungal colony-forming units (CFU) from sand and failed eggs in the nest was determined by Pearson Product-Moment Correlation Coefficient. We found a strong and positive correlation (= 0.853, p < 0.001) between fungal CFUs and the number of failed eggs in the nest. Fusarium solani was isolated with higher frequency during this sampling period. Most of the nests from which F. solani was isolated had a low hatching success (Table 1).
Twenty-one failed eggs of D. coriacea were collected from hatched and failed nests after the incubation period. Fungal structures were observed developing on the surface of most of these eggs. Dark spots, suggesting the presence of fungi, were observed in the outer inorganic layer and on the inner organic layer of some of the eggshells analyzed (Fig. 1). Disintegration of the calcareous layer of infected eggs was also observed. Mycelial fungi were isolated from 95% of the failed eggs sampled. Mycelial fungi were isolated from the interior of 52% of the eggs analyzed; all these eggs had mycelium on their surface. Fusarium solani was the only fungi isolated from these samples. Thirty-eight percent of the failed eggs had fungi associated with their surface, but no fungi were isolated from the interior. Eggs with no fungal structures on their surfaces were free of fungi in the interior as well.
Citation: Chelonian Conservation and Biology 15, 2; 10.2744/CCB-1217.1
Phylogenetic Analysis of Fusarium solani Isolates from Dermochelys coriacea Eggs
Sequences of our F. solani isolates from eggs interior were deposited in GenBank under accession numbers JN127379, JN127381, and JN127381. From our phylogenetic analysis, 3 distinct F. solani clades were obtained (Fig. 2). Clades I and II contain F. solani isolates from plants, whereas clade III contains isolates causing lesions in animals and plants. Our isolates from D. coriacea eggs are nested within clade III, which also contains sequences from F. falciforme isolated from hawksbill sea turtle eggs. Based on this result we can assume that given the right conditions F. solani can be an important pathogen to the leatherback sea turtle eggs as well.
Citation: Chelonian Conservation and Biology 15, 2; 10.2744/CCB-1217.1
DISCUSSION
A total of 85 mycelial fungi representing at least 15 identified genera and unidentified isolates were recovered from MABC sand prior to the nesting season. Most of the fungi we identified are common in soil, saprophytic or common plant pathogens. The most abundant genus was Aspergillus with 13 isolates distributed in 5 species (i.e., Aspergillus alliaceus, Aspergillus ochraceus, Aspergillus penicillioides, Aspergillus restrictus, Aspergillus terreus). The presence of these fungi represents a threat to nesting leatherback females because they have been previously reported causing loss of turtle eggs during incubation (Solomon and Baird 1980; Eckert and Eckert 1990; Phillot et al. 2001; Phillot and Parmenter 2001b; Elshafie et al. 2007). Penicillium and Fusarium are of special interest as well for their potential as egg pathogens. Penicillium is well known for its mycotoxigenic properties, which could be detrimental for eggs under incubating conditions. Fusarium has been previously reported causing egg and hatchling mass mortality in nests (Phillot and Parmenter 2001b; Phillot 2004; Phillot et al. 2004; Sarmiento-Ramírez et al. 2010). All the other fungi were isolated in lower frequencies (0.01) during sampling may be attributable to the limited nutrient availability in the sand.
Fungi identified during leatherbacks nesting season were previously recorded from the preliminary assessment of the MABC, except for Aspergillus niger. Most of the isolates could not be identified because no reproductive structures were produced. No fungi were isolated from cloacal samples, in contrast with Phillot et al. (2002) who reported that nesting and internesting turtles had a great occurrence and diversity of cloacal fungi (Acremonium, Aspergillus, Chrysosporium, Fusarium, Mucor, Penicillium, Phialophora, Sporothrix, and Stachybotrys). Our results suggest that fungi associated with egg loss may then be from the established fungal community at the MABC.
Fusarium solani was isolated from 50% of nests sampled and was the only species able to infect egg completely (surface and interior). Miller et al. (2009) reported this species from dead leatherback hatchling's skin and carapace. It has been also documented by Phillot et al. (2004) along with F. oxysporum and P. boydii occurring in other sea turtles clutches. These three fungi have been reported to cover the entire surface of the eggs and to produce enzymes that degrade their inorganic and organic compounds (Phillot and Parmenter 2001b; Phillot 2004).
Fusarium solani has been confirmed to be extremely detrimental to sea turtle eggs and that some strains may act as primary pathogens (Phillot et al. 2006; Sarmiento-Ramírez et al. 2010). The fungus is well known as an opportunistic pathogen in immunocompromised humans and has also been reported causing serious infections in leatherback hatchlings and hyalohyphomycosis in loggerhead sea turtles (Cabañes et al. 1997; Miller et al. 2009). Other Fusarium species have been also isolated from sea turtle nests and eggs, including Fusarium moliniforme and F. oxysporum, also associated with the occurrence of bronchopneumonia in adult sea turtles (Phillot et al. 2001; Elshafie et al. 2007; Güclü et al. 2010). More recently Fusarium falciforme was documented as an emerging pathogen associated with hawksbill sea turtles hatched and unhatched eggs from Ecuador (Sarmiento-Ramírez et al. 2014).
Other species that we isolated from the surface of eggs was Scedosporium aurantiacum, member of the P. boydii species complex (Gilgado et al. 2005, 2008). This species complex is characterized for being opportunistic pathogens, particularly in immunocompromised humans, and the mortality rate is extremely high (Rainer et al. 2000; Gilgado et al. 2006; Alastruey-Izquierdo et al. 2007). It is unclear whether our species is the same that Phillot et al. (2001) reported in their study, because it was not until 2005 that S. aurantiacum was proposed as a species within the P. boydii species complex. Also, S. aurantiacum has been regularly reported from clinical samples unlike its sister species S. prolificans, which has been mostly isolated from environmental samples (Gilgado et al. 2005).
Other species identified in our study include Aspergillus ochraceus, Aspergillus niveus, Aspergillus sclerotiorum, Penicillium islandicum, Cladosporium sp., and Trichoderma longibrachiatum. All of these fungi were isolated from nests sand and egg surface but were not able to colonize the eggs' interior. Aspergillus species have been reported as a threat to sea turtle clutches for their mycotoxigenic properties and has been the most frequent species isolated from sea turtle nests in other studies (Elshafie et al. 2007). Our Aspergillus isolates may then contribute to egg failure during incubation, not by direct colonization, but by the production of mycotoxins that could diffuse through the eggshell and negatively affect embryo development.
Fungal abundance is strongly correlated to egg loss in D. coriacea nests. This does not necessarily mean that fungi are the main cause of egg loss because high humidity and low temperatures are not only detrimental to egg development but are also favorable for fungal proliferation. According to Morelock et al. (1983), most of the MABC rainfall occurs between April and October, which collides with leatherback nesting season. The combination of these two factors and the presence of the fungi may then be the cause for leatherback egg loss during our study.
Different stages of deterioration of Dermochelys coriacea unhatched eggs. Deterioration of the calcium carbonate (A), colonization of the outer inorganic layer by fungi (B), and colonization of the inner organic layer of the egg by fungi (C). (Color version is available online.)
Bayesian inference phylogram from the ITS region of the rDNA of various Fusarium species. Posterior probabilities are shown above the branches. Members of the Fusarium solani species complex (FSSC) are shown in clades I, II, and III. Clades I and II contain sequences from isolates from plants. Clade III contains isolates from plants (not highlighted), isolates from lesions in humans (orange), isolates from hawksbill sea turtle eggs (blue), and our isolates from Dermochelys coriacea eggs (green). (Color version is available online.)
Contributor Notes
Handling Editor: Jeffrey A. Seminoff
