Five marine turtle species are found along the north coast of Rio Grande do Norte state, northeastern Brazil: Caretta caretta (loggerhead turtle), Dermochelys coriacea (leatherback turtle), Chelonia mydas (green turtle), Eretmochelys imbricata (hawksbill turtle), and Lepidochelys olivacea (olive ridley turtle). This area is also an important feeding and nesting area for the last 3 species (Marcovaldi and Marcovaldi 1999; Farias et al. 2019). According to the International Union for Conservation of Nature's Red List of Threatened Species (IUCN 2018), the sea turtles occurring in Brazil are classified as vulnerable (C. caretta, L. olivacea, and D. coriacea), endangered (C. mydas), and critically endangered (E. imbricata).

Marine turtles face several threats such as habitat destruction, coastal development, a fishery, pollution, anthropogenic interaction (Herbst 1994; Bugoni et al. 2001; Mast et al. 2005), and diseases, e.g., fibropapillomatosis (FP), which can interfere with hydrodynamics, locomotion, and food ingestion (Adnyana et al. 1997; Greenblatt et al. 2005).

FP is characterized by skin tumors, and the Chelonid herpesvirus 5 (ChHV5) has been considered the primary etiological agent of FP; however, there is a general consensus that FP has a multifactorial cause with the association of environmental, genetic, and biological cofactors (Work et al. 2009; Keller et al. 2014). Green turtles with numerous FP tumors may be considered super spreaders of the disease in the ocean environment (Herbst and Klein 1995; Lackovich et al. 1999; Work et al. 2014b).

FP was first reported in 1936 in a female green turtle (C. mydas) from Key West, Florida (Smith and Coates 1938). Since then, FP-tumor lesions have been described in loggerhead (Harshbarger 1991; Mascarenhas and Iverson 2008; Page-Karjian et al. 2015; Rossi et al. 2015), Kemp's ridley (Lepidochelys kempii; Barragan and Sarti 1994), hawksbill (D'Amato and Moraes-Neto 2000), olive ridley (Aguirre et al. 1999), flatback (Natator depressus; Limpus et al. 1993), and leatherback (Huerta et al. 2002) turtles. FP is most frequently reported in sea turtles at coastal feeding grounds, areas often impacted by human activities (e.g., land use, high nitrogen concentration, and onshore sewage disposal in turtle feeding areas), leading to a eutrophication process through delivery and retention of nitrogen in coastal waters (Aguirre and Lutz 2004; Ene et al. 2005; Van Houtan et al. 2010). According to Van Houtan et al. (2010), eutrophic coastal ecosystems may promote herpesvirus infections among herbivores (but see Work et al. 2014a). Juvenile green turtles feed mainly on sea grass and/or macroalgae while in coastal areas, where they spend more time than do other sea turtle species (Mortimer 1982; Bjorndal 1997). A study carried out in Espírito Santo state, Brazil, found marine algae species with fast growth and short lifecycles, indicating a degraded ecosystem; FP-affected green turtles were also observed in the region (Santos et al. 2011). The relationship between FP and eutrophic coastal habitats was also mentioned by Van Houtan et al. (2010). Therefore, the coastal behavior of juvenile green turtles could explain why they are the most commonly affected by FP.

FP tumors can be sessile or pedunculated, varying in color and texture, and classified as papillomas (epidermal), fibromas (dermal), or fibropapillomas (epidermal and dermal) depending on the proportion of epidermal or dermal proliferation (Herbst 1994). Such tumors can be located throughout the body surface, in oral, cranial, cervical, axillary, and inguinal regions, limbs, eyes, carapace, and plastron (George 1997; Herbst et al. 1999; Rossi et al. 2009; Zwarg et al. 2014; Rossi et al. 2016). Internal fibromas, myxofibromas, and fibrosarcomas were reported in lungs, heart, kidneys, muscles, and the gastrointestinal system (e.g., liver and oral cavity; Norton et al. 1990; Herbst 1994; Balazs et al. 1997; Yu et al. 2000).

The spatial–temporal assessment of FP prevalence in Brazil conducted by Projeto TAMAR showed a variation of 0.04% (data obtained from 2004 to 2014; n = 2431) to 32.59% (period between 2000 and 2014; n = 2940) in the Rio Grande do Sul and Ceará states, respectively (Baptistotte 2016). Studies in the Espírito Santo bay, Espírito Santo state, found low algae quality and diversity and high environmental degradation, demonstrated by a high FP prevalence in green turtles (Santos et al. 2010b). According to Van Houtan et al. (2010), anthropogenic-derived nitrogen is an important environmental factor (from sewage injection wells, urbanization, aquaculture and fishponds, and agriculture), and a potentially significant FP-promoter (but see Work et al. 2014a).

Considering that FP is one of the main threats to the conservation of green turtles, this study was aimed at determining its incidence in stranded individuals found in the Potiguar Basin, Rio Grande do Norte and Ceará states, northeastern Brazil. We describe the presence of FP in stranded sea turtles in this region and provide tumor scores to describe the severity of affliction in individual turtles. We hope that these data will serve as a baseline for future studies of FP in Brazil and elsewhere in the Atlantic Ocean.

METHODS

Study Area. — This research was carried out in northeastern Brazil, between Cai çara do Norte (5°4′1.15″S, 36°4′36.41″W) in Rio Grande do Norte state (RN) and Icapuí (4°38′48.28″S, 37°32′52.08″W) in Ceará state (CE), areas known as Potiguar Basin, covering an approximate extension of 300 km.

Since 2010, the Projeto Cetáceos da Costa Branca–Universidade do Estado do Rio Grande do Norte (PCCBUERN) in Brazil has conducted the Beach Monitoring Program in the Potiguar Basin (Programa de Monitoramento de Praias da Bacia Potiguar [PMP-BP]). The PMPBP is in compliance with the environmental constraint enforced by the Brazilian Institute of the Environment and Renewable Natural Resources (IBAMA) over the activities operated by PETROBRAS (Petróleo Brasileiro S.A.; agreement no. 2500.005657510.2).

Formed by crystalline basement rocks (Soares et al. 2003), the monitoring area was divided into 2 sections due to its wide extension, access condition (presence of bays, urban and rocky beaches), local infrastructure, and environmental characteristics. Sector 1 was divided into Stretches A (Grossos–Icapuí; 122 km) and B (Areia Branca–Porto do Mangue; 102 km) and has tourism, artisanal fishing, and saliniculture as its main economic activities. Sector 2 was divided into Stretches C (Guamaré–Macau; 104.2 km) and D (Galinhos–Caiçara do Norte; 61.6 km), where fishing (artisanal or professional, depending on the beach), wind energy, and gas and petroleum exploitation represent the local economic activities (Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Individuals and FP Tumors. — We analyzed data obtained during daily monitoring conducted from January 2012 to December 2015. The monitoring covered all 4 stretches and was conducted by trained field team using a 4-wheel (i.e., quad) motorcycle and a portable global positioning system (GPS) to record the location of stranded sea turtles (dead or debilitated individuals).

Curved carapace length (CCL; measured from the nuchal to notch between supra-caudal scales; Bolten 1999) was recorded in all individuals using a flexible tape, and the species of each turtle was identified when possible. The CCL was used to classify the individuals as juveniles (CCL up to the value that precedes the smallest registered size for nesting females in the largest and closest nesting area for the species in Brazil) and adults (CCL ≥ 90 cm for C. mydas [Almeida et al. 2011], CCL ≥ 83 cm for E. imbricata [Santos et al. 2010a], CCL ≥ 62.5 cm for L. olivacea [Silva et al. 2007], CCL ≥ 86.5 cm for C. caretta [Lima et al. 2012], and CCL ≥ 139 cm for D. coriacea [Thomé et al. 2007]). Unfortunately, owing to advanced autolysis, species identification was not possible in some of the cases and, on those, CCL measurement was not performed.

According to the PCCB-UERN protocol, FP tumors found on green turtles stranded in 2015 were quantified and analyzed according to 1) anatomical distribution (eyes, head, cervical region, forelimbs, hind limbs, carapace, plastron, and inguinal region); 2) size (A: < 1 cm; B: 1–4 cm; C: > 4–10 cm; D: > 10 cm; Work and Balazs 1999); and 3) Southwest Atlantic Fibropapillomatosis Score (FPSSWA; mild, moderate, and severe; Rossi et al. 2016).

Statistical Analysis. — We used the chi-square (x2) test to detect differences in the occurrence and proportion of strandings in different categorical variables and the additive model (Kendall and Stuart 1983) to evaluate seasonality through the distribution of the time-series according to the number of stranded sea turtles per month (with and without FP). We did not consider the non–species-identified turtles in these analyzes. Significance level was 0.05 for all tests, and analyses were performed using PAST version 2.14 and R software (R Core Team 2018).

RESULTS

Marine Turtle Strandings.—During the study period, 2688 sea turtles (128 alive) representing the 5 species that occur in Brazil were reported, with green turtles representing 87.4% of encountered turtles (n = 2350). The lowest incidence of strandings was recorded in 2013 (n = 629) and the highest in 2014 (n = 732) (Table 1). Statistical analysis revealed differences in the number of strandings among years (p = 0.01765; x²3 = 10.11). In regard to the monitored stretches, most of the individuals were found along stretch A (n = 1260) followed by B (n = 837), D (n = 298), and C (n = 293).

Table 1. Sea turtle species found during monitoring in the Potiguar Basin, Rio Grande do Norte and Ceará, Brazil, between 2012 and 2015. FP = fibropapillomatosis.

View Fullscreen

A total of 2480 individuals were classified according to their life-stage categories, and CCLs ranged 18–132 cm for green, 32–104.5 cm for loggerhead, 3.6–98.5 cm for hawksbill, and 11.6–74 cm for olive ridley turtles; we found only 1 adult leatherback turtle (CCL = 149 cm). Most individuals were classified as juveniles (n = 2207): 2073 green turtles, 97 hawksbill turtles, 22 olive ridley turtles, and 15 loggerhead turtles (Table 2).

Table 2. Number of sea turtles found during monitoring in the Potiguar Basin, Rio Grande do Norte and Ceará, Brazil, between 2012 and 2015, according to their stage of development. FP = fibropapillomatosis.

View Fullscreen

Incidence of FP in the Study Area. — During the study period, 610 stranded marine turtles (18 alive) presented FP tumors (22.7%), with the majority being green turtles (99%; n = 604) followed by 2 olive ridley turtles (0.3%), 1 hawksbill turtle (0.16%), and 1 loggerhead turtle (0.16%). Two unidentified individuals presented FP tumors (0.3%) (see Table 1).

Considering only data on green turtles—the species that represents 99% of the FP-affected individuals in this study—we identified an increase of relative FP frequency from 13.2% in 2012 (65/494) to 35.3% in 2015 (228/646) (Fig. 2A). Time-series analysis of FP-free individuals showed a high stranding probability in October, November, and December, suggesting a seasonal stranding pattern. FP-affected individuals had a less-evident stranding pattern in the same months (Fig. 2B–C). There was a significant difference in the number of strandings among years (p = 0.01765, x²₃ = 10.11). The highest relative frequency of affected green turtles was observed in Stretch C (48.1%, 115/239) followed by Stretches A (29.3%, 333/ 1135), B (18.8%, 132/704), and D (8.8%, 24/272) (p < 0.0001, x²₃ = 129.14) (Table 1).

View Figure

• Download as Powerpoint
• Download Figure

Considering the analysis of life-stage data, FPaffected green turtles represented 26.2% of stranding events (602/2304); 592 were classified as juveniles and 10 as adults (see Table 2). The proportion of individuals with FP was higher in juveniles (28.6%) than in adults (4.3%) (p < 0.0001, x²₁ = 67.962) (Table 2). The proportion of juveniles and adults were similar among years (p = 0.5225, x²3 = 2.2483) and different among stretches: 88.8% of juveniles in Stretch A, 87.8% in B, 93.7% in C, and 97.3% in D (p < 0.0001, x²3 = 25.296).

Description of FP Tumors.—A total of 2158 external tumors were counted in 171 green turtles (stranded in 2015), ranging between 1 and 67 tumors/individual (12.6 ± 11.9 tumors/turtle; mean ± SD). The forelimbs were the most affected anatomical region (1071 tumors) followed by the neck (560 tumors) (Table 3). Tumors were verruciform or smooth and of varied colorations (white, pink, gray, or black) (Fig. 3). There was no association between anatomical regions and tumor size (p = 0.11, x²₂₇ = 36.1).

Table 3. Number of tumors on green turtles (Chelonia mydas) stranded in the Potiguar Basin, Rio Grande do Norte and Ceará, Brazil, between January and December 2015, according to anatomical region and size category.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Most tumors were classified as size B (1607 tumors) followed by C and A (285 and 240 tumors, respectively); only 26 tumors were classified as D (found on 18 individuals). According to FPS_SWA, 56.7% of the 171 examined green turtles were classified as mild, 32.7% as moderate, and 10.5% as severe.

DISCUSSION

During the study period (from 2012 to 2015), we recorded a total of 2688 sea turtles stranded at the Potiguar Basin, all belonging to the 5 species that occur in Brazilian waters and previously reported in northeastern Brazil (Marcovaldi et al. 2011; Farias 2014; Gavilan et al. 2016). According to Farias et al. (2019), strandings in this region are most commonly juvenile green and hawksbill turtles and adult ridley and loggerhead turtles. Our data show that green turtles composed 87.4% of the strandings (n = 2350) while in Hawaii green turtles were the identified species in 97% of the stranding events between 1982 and 2003 (n = 3732/3861; Chaloupka et al. 2008). Reports from the PMP-BP and PMP–Santos Basin showed that green turtles were the most commonly reported species in stranding events (PETROBRAS 2017, 2018).

Our results revealed an increase in FP frequency, with 13.2% of examined green turtles presenting tumors in 2012 (65/494) and 35.3% in 2015 (228/646). Studies along the east coast of the United States indicated an increase of FP prevalence from 10% in the early 1980s to over 30% in the late 1990s (Foley et al. 2005). In Hawaii, the apparent prevalence varied and persisted for decades, including an outbreak in the late 1980s, followed by a peak in the mid-1990s and a decline up to 2007 when the incidence rate decreased to 9.4% (Chaloupka et al. 2009).

The main economic activities in the Potiguar Basin are salt and oil exploitation, shrimp farming, and artisanal fishery, resulting in environmental degradation and mangrove destruction. The saliniculture industry has increased since the 18th century due to its ever-evolving technology (Costa et al. 2013). Therefore, the increased occurrence of FP-affected green turtles could be related to the increase in resource exploitation and the consequent boost in anthropogenic impacts in the area. For example, effluent discharge contributes to habitat contamination because there is no satisfactory water treatment along the monitored stretches (Cavararo et al. 2011). Areas under heavier anthropogenic impact present a higher number of FP reports (García-Fernández et al. 2009; Ikonomopoulou et al. 2009). This dynamic was previously demonstrated in a degraded ecosystem in Brazil which held a commercial port, water discharge from domestic residences, and both light and heavy coastal industry (Santos et al. 2010b, 2011). Environmental pollutants may play a role as a cofactor on the development of this disease, compromising physiological aspects and the immune response, leading to chronic stress and increased disease susceptibility, induc-ing latent virus reactivation (Aguirre et al. 1994; Herbst 1994; Aguirre et al. 1995; Herbst and Klein 1995; Adnyana et al. 1997; Miao et al. 2001; Aguirre and Lutz 2004; Keller et al. 2014). In regard to data on green turtles, the relative FP frequency was high in all monitored stretches, suggesting that the impact of human activities could be associated with the development of this disease. In Hawaii, soil-related activities, when associated with elevated nitrogen concentrations, eutrophication processes, and the presence of invasive algae, were highly correlated with animals presenting FP cutaneous lesions (Van Houtan et al. 2010).

Our results revealed that juvenile green turtles (n = 592) were more affected than were adults (n = 10), suggesting that the Potiguar Basin is an important feeding area for juveniles. Farias (2014) also found that this area is an import feeding ground for sea turtles, especially green turtles. This species occupies developmental habitats and becomes a resident to coastal areas, returning to specific sites for foraging, and only leaving as adults for reproductive migrations (Musick and Limpus 1997; Chaloupka et al. 2004; Limpus et al. 2005). Studies have indicated that FP is most frequently reported in sea turtles at coastal feeding grounds (Ene et al. 2005; Chaloupka et al. 2009), which could partially explain why this species apparently is more susceptible to FP development.

Seasonality also affected the stranding rates of sea turtles with and without FP, the latter of which occurred more frequently from October to December. Farias (2014) reported a higher number of strandings between June and January. The lobster fishing season in the study area is from June to November, and the high number of fishing vessels, as well as the presence of nets and lobster traps, could explain the increased stranding rate recorded in this period.

In accordance with previous reports in Brazil, we observed various tumor colorations and textures (verruciform and smooth) (Rossi et al. 2009; Zwarg et al. 2014) and a higher occurrence frequency on limbs than on other anatomical regions (Baptistotte 2007; Mascarenhas and Iverson 2008; Rossi et al. 2016). We also reported more tumors on forelimbs than on hind limbs, as described by Page-Karjian et al. (2014) and Rossi et al. (2016). Studies carried out in Hawaii (Work et al. 2004), Brazil (Santos et al. 2010b, 2015), and in the United States (Page-Karjian et al. 2014) also reported that the most affected region was the anterior. In contrast, green turtles studied in Indonesia presented the highest number of FP lesions in the caudal body portion (Adnyana et al. 1997). We recorded few FP tumors on the carapace/plastron, similar to the findings described by Santos et al. (2010b). In Florida, researchers found the most FP tumors around the neck and limbs (Hirama and Ehrhart 2007). Aguirre et al. (2002) highlighted that FP tumors in the oral cavity of green turtles are common in Hawaii but rare in Florida. Our study demonstrated that the cervical region presented 26.4% of all tumors, unlike an earlier report from Brazil that described that only 10% of the tumors were located in this area (Rossi et al. 2016).

The total number of tumors ranged from 1 to 67 among the examined individuals: almost 50% of afflicted turtles presented more than 10 tumors (n = 80), with 12.62 ± 11.92 per individual. Rossi et al. (2016) reported a range from 1 to 129 tumors per turtle (n = 214 green turtles), with 79% presenting more than 10 tumors and with 8.7 ± 6.8 tumors/turtle. Other research in Brazil found 40 ± 20 tumors/individual with a range of 2 to 91 tumors (Santos et al. 2010b). Studies conducted in Indonesia reported a range of 1 to 29 tumors per individual (n = 949 individuals), with an average of 5 ± 4.1 tumors and 12.6% of turtles presenting more than 10 tumors (Adnyana et al. 1997).

We classified most tumors characterized in this study as size B followed by categories C and A, respectively. A study conducted in southeastern Brazil that found 7466 FP tumors on 214 juvenile green turtles reported categories A and B as the most frequent (Rossi et al. 2016). According to the FPSSWA, most of the studied individuals were classified as mild followed by moderate. Vilca et al. (2018) used this score to correlate to polycyclic aromatic hydrocarbon concentrations, and they classified more green turtles in the moderate score followed by mild. Other studies carried out in Brazil, which used another FP score system, verified that 34.4% of captured green turtles had FP tumors with different levels of severity score (220/ 640) (Torezani et al. 2010), and 94.25% (82/87) presented tumor scores 2 or 3 (Santos et al. 2010b)

Our results provide relevant FP data on stranded sea turtles in the Potiguar Basin and help describe the frequency of this disease in the area during the several years of this study as well as the anatomic distribution of FP tumors and severity score for turtles in the region. Considering that the Potiguar Basin is an area of rich biodiversity and provides important feeding and nesting habitat for sea turtles, our work can contribute to the development of appropriate mitigation measures such as greater control over the use of fishing gear and reinforcement of public policies regarding marine ecosystem conservation.