Die Studie konzentriert sich auf die Plasmachemie und Hämatologie grüner Meeresschildkröten im östlichen Mittelmeer, die sich einer Rehabilitation unterziehen, und betont, wie wichtig es ist, ihre physiologischen Parameter für effektiven Schutz und Management zu verstehen. Das Mittelmeer, insbesondere das östliche Mittelmeer, ist ein wichtiger Lebensraum für diese gefährdeten Schildkröten, die zahlreichen anthropogenen Bedrohungen ausgesetzt sind. Die Forschung liefert wertvolle Referenzintervalle für Blutanalysen, die für die Beurteilung des Gesundheitszustands rehabilitierter Schildkröten von entscheidender Bedeutung sind. Die Ergebnisse zeigen auch signifikante Unterschiede in der Plasmachemie zwischen den Lebensstadien und Korrelationen mit der Körpergröße, was zu einem tieferen Verständnis der Physiologie dieser Schildkröten beiträgt. Diese Forschung schließt eine entscheidende Wissenslücke und hilft bei der Entwicklung zielgerichteter Schutzmaßnahmen zum Schutz dieser empfindlichen Meeresarten.
KI-Generiert
Diese Zusammenfassung des Fachinhalts wurde mit Hilfe von KI generiert.
Abstract
The Eastern Mediterranean Sea (EMS) is oligotrophic, presenting low biological production and high temperature and salinity, while comprising a biodiversity hotspot, with a unique emblematic threatened species array, including sea turtles. The establishment of healthy captive green turtle (Chelonia mydas) baseline blood analyte reference intervals (RIs) will contribute to improving diagnosis, treatment, rehabilitation and conservation success rates for sea turtles, by offering information with therapeutic and prognostic implications. However, baseline plasma chemistry and hematological analytes of EMS C. mydas are still unavailable. Hence, this study aimed to establish RIs for several blood analytes in general and under different life and rehabilitation stages and compare results with congeneric and conspecifics of other regions: Mediterranean loggerhead turtles (Caretta caretta) and two Atlantic green turtle populations. The study engulfed 118 blood samples obtained from 72 turtles admitted to the Israel Sea Turtle Rescue Centre between 2008 and 2020. Blood analyses included plasma chemistry and packed cell volume (PCV), and their association with body size and health status, and enabled suggesting RIs for PCV, plasma glucose and potassium concentrations, and aspartate transaminase and alkaline phosphatase activities. The PCV were significantly associated with rehabilitation. Noted differences arising via population comparison are likely associated with interspecific dietary and metabolic traits. These novel physiological results form the basis for future studies, and provide caregivers a tool potentially increasing the success of recovery, rehabilitation and conservation efforts. Further research is warranted to fill knowledge gaps concerning physiology and potential ecological impacts on this species.
Tal Starostinetsky Malonek and Aviad Scheinin have contributed equally to this work.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
The Mediterranean Sea is the largest, deepest enclosed sea worldwide, comprising a marine biodiversity hotspot with high endemism and an array of emblematic threatened species, including sea turtles and several marine mammals (Coll et al. 2010). Various organisations embrace the use of distinct species or taxa as symbols or “flagships” as a measure for informing and engaging the public, benefactors, and political agencies in conservation endeavours. Local flagship species can be instrumental in eliciting wide commitment and support from local communities, that ultimately determine the success of many conservation enterprises (Bowen-Jones and Entwistle 2002).
Sea turtles commonly serve as flagship species, since they are highly migratory and present complex life histories, and are therefore subjected to extensive anthropogenic threats worldwide, both at sea and on land, when nesting and hatching (Ullmann and Stachowitsch 2015). However, despite their importance, data on many Mediterranean Sea flagship species, including sea turtles, are essentially lacking. Gaining physiological knowledge on such animals in the Eastern Mediterranean Sea (EMS) is, therefore, warranted, to improve management and conservation efforts.
Anzeige
Three of the seven existing sea turtle species occur in the Mediterranean Sea, with two reproducing in its basin, namely the loggerhead (Caretta caretta) and green (Chelonia mydas) turtles. While the loggerhead nesting areas span all along the Mediterranean coastline, from west to east, green turtle nesting is restricted to the easternmost Mediterranean basin, mainly in Turkey and Cyprus, with several additional sites in Israel, Syria, Lebanon, Egypt, and some islands around Crete. Hatchling dispersion data and concentrations of small juvenile green turtles detected in this area, along with the high stranding rate of small juveniles on the Turkish Mediterranean coast, imply that the Levantine Basin is the main Mediterranean Sea nursery area of this species (Camiñas et al. 2020).
The green turtle is globally classified as “Endangered” in the IUCN Red List, and “Possibly Extinct” in Israel (Seminoff 2004). Its Mediterranean population is genetically unique, separate from its Atlantic populations, and is highly threatened and at grave risk of extinction (Ullmann and Stachowitsch 2015). Mediterranean Sea green turtles have been severely depleted, with their targeting by fisheries from the 1920s to 1970 (Hornell 1935; Sella 1982). Green turtle nesting is mostly limited to the Eastern Mediterranean, mostly in Greece, Lybia, Cyprus and Turkey (Casale and Margaritoulis 2010). Conservation actions of sea turtles on nesting beaches commenced in Cyprus in the early 1970s, in Turkey and Greece during the 1980s, and in Israel in the 1990s. Acknowledging the necessity for designated rescue facilities for sea turtles initiated the foundation of the first Mediterranean rescue centres, in Italy (1992) and Greece (1994) (Camiñas et al. 2020).
Israel’s Sea Turtle Rescue Centre (ISTRC) was established by the Israel Nature and Parks Authority in 1999. Currently, about 40 Mediterranean Sea rescue centres and first-aid stations trained in rehabilitating and supplying veterinary care to wounded sea turtles are now in place (Camiñas et al. 2020). Rescue centres also inform locals, tourists and coastal and marine officials of the local sea turtle population condition and disturbances, along with other marine environmental concerns, engage the public and conduct vital research of the physiology, demographics, parasitology and other health-related subjects, thereby expanding existing knowledge on marine turtles (Camiñas et al. 2020). Basic and specific information of Mediterranean Sea turtles is still scarce, with substantial knowledge gaps regarding annual clutch statistics and locations, age of reproductive maturity, behavioural ecological details, and survival rate, and especially so, of green turtles and concerning the Southeastern Mediterranean (Casale et al. 2018). Recent estimates of the Mediterranean Sea green turtle abundance suggest an annual average of 2200 egg clutches, with a median seasonal clutch number laid per female of three, while the female remigration interval between nesting seasons is 3 years (Casale et al. 2018). In Israel, 121 green egg clutches were laid over the last 3 years (Dr Yaniv Levy, ISTRC, personal communication), suggestive of an estimated green turtle nesting female number of 40.
Mediterranean Sea marine turtles are threatened by several factors, including marine and terrestrial habitat degradation, climate change effects [e.g., increasing air and seawater temperatures, sea level, intensified storms, maritime traffic, intensive fishing bycatch and marine pollution (Camiñas et al. 2020)]. For example, in February 2021 an oil spill near the Israeli Mediterranean coast led to severe tar coastline pollution in Israel, directly harming at least 27 sea turtles, with 21 deaths (Dr Yaniv Levy, ISTRC, personal communication). In addition to such extensive regional threats, illegal trade in sea turtle products is also noted, as in Egypt (Camiñas et al. 2020).
Anzeige
Baseline plasma chemistry and haematology values of healthy wild animals assist in optimising population management and increase our understanding of how future anthropogenic and environmental changes affect their health. Blood analytes are important diagnostic and prognostic markers of disease (e.g., chronic or acute inflammations, stress and infectious diseases) and physiological status (e.g., reproductive state and nutrition). Establishing reference intervals (RIs) for blood analytes is, therefore, important for health assessment and monitoring (Page-Karjian et al. 2015), and is paramount sick or wounded sea turtles.
Variations in green turtle plasma chemistry have been linked with body size, sex, feeding location, captivity, and presence of disease (Arthur et al. 2008), determined previously in many green turtle populations (Bolten and Bjorndal 1992; Samour et al. 1998; Hamann et al. 2006; Whiting et al. 2007; Arthur et al. 2008; Flint et al. 2010, 2019; Fong et al. 2010; Anderson et al. 2011; Vasco et al. 2014; Page-Karjian et al. 2015; March et al. 2018; Reséndiz et al. 2018), as well as in other sea turtle species (Stamper et al. 2005; Deem et al. 2006, 2009; Kakizoe et al. 2007; Casal and Oros 2009; Casal et al. 2009; Gelli et al. 2009; Fazio et al. 2012; Patterson-Kane et al. 2013; Alberghina et al. 2015; Flower et al. 2015; Kelly et al. 2015; Pinto et al. 2015; Perrault and Stacy 2018; Yang et al. 2019; Perrault et al. 2020). Scarce studies, however, are available in the Mediterranean Sea, where these parameters have only been examined in loggerhead sea turtles (Gelli et al. 2009).
Materials and methods
Israel Sea Turtle Rescue Centre (ISTRC)
The ISTRC provides medical care to sea turtles sustaining variable afflictions, mostly of anthropogenic origin, which are found stranded onshore or at sea along the Israeli Mediterranean coastline. Such turtles are rehabilitated at ISTRC in fiberglass, polyvinyl chloride, or polypropylene tanks, containing up to 12 m3 of sea water situated in open air, with water temperatures ranging between 18 and 27 °C, depending on diurnal and seasonal changes, or adjusted with medical treatment requirement. Sea water is pumped from an underground nearshore 70 m deep well and filtered through sand before use. Turtles are fed once daily with fish, invertebrates, lettuce and/or algae, depending on species, bodyweight, and other individual requirements. Prior to release to the sea, turtles are deemed healthy and rehabilitated and fully recovered by the ISTRC professional caregivers, based on behavioural, medical, and physiological parameters (Bluvias 2008).
Sampling process
Between 2008 and 2020 some 72 green turtles were treated at the ISTRC: Two of them were still being treated during the time the research was conducted; 18 turtles died, of which two were euthanized (all turtles that died were considered “non-survivors”); 49 were released back to the sea, while three turtles were not in a condition to be reintroduced into the wild and had to be kept in captivity.
Upon admittance, physiological data were collected for each turtle, including sex (if visual determination was possible), midline curved carapace length (CCL) and body weight. The turtles were divided into three groups representing the life stages: juveniles, subadults and adults. These distinctions were made based on midline curved carapace length (CCL). Since the shift between the early oceanic life stage and the later neritic stage of Mediterranean Sea green turtles occurs when CCL is 30 cm (Cardona et al. 2010), turtles with CCL ≤ 30 cm were considered in this study as juveniles. In green turtle females nesting in Cyprus, the minimum CCL observed was 77 cm (Broderick et al. 2003). Thus, herein, turtles with CCL > 30 cm and ≤ 75 cm were defined as subadults, and as adults when CCL was > 75 cm.
Blood samples were taken by certified and trained veterinarians for deeper physiological analyses upon admission to the ISTRC, during rehabilitation there, and shortly prior to their release to sea. Samples taken within 4 days of turtle admission at the rescue centre were defined as “admitted”, and samples taken in the 7 days preceding the turtles’ release were considered “recovered”. Turtles that spent only a short period of time at the ISTRC—if because they were in relatively good condition and were quickly released (N = 26), or because they were in a very bad shape and died shortly after admission (N = 18)—were sampled only one time. Most of the remaining turtles (N = 26) were sampled once more before being released, and two other individuals were sampled three times in total. All samples were processed within one hour from sampling at the ISTRC, excluding samples of seven turtles, admitted between 2018 and 2020, none of which was sampled shortly prior to its release.
The blood sampling process required the turtles be handheld out of the water. Antiseptic (chlorhexidine and 70% ethanol) was applied by gauze pads. At each sampling, about 2 mL were obtained from sea turtles from the dorsal cervical sinuses (Wyneken et al. 2006) using 2 mL and 5 mL syringes and 22-gauge, 1.5-inch needles. The whole blood sample was then divided into two tubes and a capillary: one tube meant for biochemistry analysis (heparin-lithium tube), a second tube for CBC blood count (EDTA) and (< 0.5 mL was placed in) a capillary tube for packed cell volume (PCV) and total solid (TS) analysis. The remaining blood, between 0.5 and 1 mL, was transferred into 15 mL sterilised tubes and freeze-dried for elemental analyses—reported in a different paper.
Whole blood glucose concentration was measured immediately post sampling using a glucometer (ACCU-CHEK, Roche, Mannheim, Germany). Whole blood hemoglobin was measured using a dry chemistry analyser (Reflovet Plus, Scil Animal Care Company, Viernheim, Germany). The remaining whole blood sample was kept in lithium-heparin tubes (Greiner Bio-One, Kremsmünster, Austria). Heparinized whole blood was centrifuged in glass, heparin-containing capillaries (14,810×g for 15 min; HCEN-202 hematocrit centrifuge, MRC, Holon, Israel). The PCV was measured using a reader card. The remaining heparinized whole blood was centrifuged (1680×g for 15 min; LCEN-324 centrifuge, MRC, Holon, Israel), and harvested plasma underwent chemistry analysis (dry chemistry analyser Reflovet Plus, Scil Animal Care Company, Viernheim, Germany) following the manufacturer instructions. Chemistry analytes included glucose, cholesterol, triglycerides, potassium, hemoglobin, aspartate transaminase (AST), creatine kinase (CK) and alkaline phosphatase (ALP). In the samples of the above-mentioned seven turtles in which plasma chemistry was not analysed at the ISTRC, the PCV was measured as described above, blood was centrifuged as above, and harvested plasma was then stored at − 20 °C (for ≤ 2 months) pending shipment to the Hebrew University Veterinary Teaching Hospital Diagnostic Laboratory, Rishon Lezion, Israel. Plasma samples were shipped cooled at 4 °C, arriving within 12 h. Plasma chemistry analysis was performed using a wet chemistry autoanalyzer (Cobas 6000, Roche, Mannheim, Germany; at 37 °C).
Statistical analyses
Results were tested for normality by the Shapiro–Wilk's test, and for equality of variance by an F test. Two group results were compared using the Wilcoxon's test. Comparisons of > 2 groups were performed using the Kruskal–Wallis test, and the post-hoc comparison P-values were adjusted by the Benjamini–Hochberg method using the “dunn.test” R package. The paired Wilcoxon signed-rank test was used for comparing results upon admission and prior to release to sea. The association of two continuous variables was determined using Spearman’s rank correlations. The receiver operator characteristic (ROC) curve, with its area under the curve (AUC), was used to examine the predictive prognostic performance of blood analytes found significantly associated with the overall outcome of the turtle treatment (R software, pROC package) (Robin et al. 2011).
All tests were 2-tailed, and in all of them P ≤ 0.05 was considered significant. Statistical analyses were performed using R Version 4.0 software, stats R package (R Core Team 2020).
Additionally, the suggested RIs were calculated using Reference Value Advisor 2.1 (Geffré et al. 2011), following the American Society for Veterinary Clinical Pathology (ASVCP) guidelines (Friedrichs et al. 2012). The Reference Value Advisor 2.1 software uses standard or robust methods, following Horn et al. (1998), and Box–Cox transformation, and provides RIs with 95% confidence intervals (CIs).
Results
The median CCL of recovered turtles was 30.5 cm (ranging between 18.5 and 84.5 cm). The median rehabilitation period was 76 days (ranging between 1 and 1093 days).
Plasma chemistry and packed cell volume
Plasma chemistry and PCV of 118 blood samples, obtained from 72 green turtles admitted to the ISTRC between 2008 and 2020 were analysed, of which 46 samples were obtained upon admission, 47 during rehabilitation, and 25 just prior to their release. The analyses found that PCV (%) had a mean of 30 and an SD of 4.6; Glucose (mg/dL) expressed a mean of 113 (SD = 25); Hemoglobin (g/dL): 10.8 (2.5); Triglycerides (mg/dL): 199 (99); and Total protein (g/L): 6.6 (1.3). The full 8-analyte plasma chemistry and PCV results per individual turtle are presented in Supplementary Table S1.
Blood analyte results and proposed reference intervals for recovered turtles
The plasma chemistry and PCV results of 25 recovered turtles, deemed healthy, with their suggested RIs and upper and lower 90% CIs are presented in Table 1. The 90% CIs calculated for five analytes exceeded the recommended 20%.
Table 1
Packed cell volume and plasma chemistry analyte results, with proposed reference intervals, of recovered Eastern Mediterranean Sea green turtles (Chelonia mydas)
Analyte (units)
N
Meana (SD)
Median (range)
Lower RLb 90%CI
RIb
Upper RL 90%CIb
Method
Packed cell volume* (%)
26
30 (4.6)
28 (19–37)
16–20
18–37
35–40
S
Glucose* (mg/dL)
24
113 (25)
108 (65–164)
52–77
64–170
152–189
ST
Hemoglobin* (g/dL)
13
10.8 (2.5)
10.8 (6.2–14.4)
CK (U/L)
18
833 (234–74,400
AST (U/L)
25
327 (96–1074)
77–140
99–1110
755–1679
RT
ALP (U/L)
24
62 (21–327)
17–24
19–238
153–352
RT
Triglycerides* (mg/dL)
16
199 (99)
174 (77–412)
Cholesterol (mg/dL)
19
155 (117–289)
Total protein* (g/L)
12
6.6 (1.3)
6.6 (4.3–8.6)
Potassium (mmol/L)
24
4 (3.1–6.4)
2.8–3.1
2.9–7.2
5.7–9.8
RT
N number of turtles; SD standard deviation; RI reference interval; RL reference limit; 90%CI 90% confidence interval; S standard method; ST standard method after Box-Cox transformation; RT robust method after Box-Cox transformation; CK creatine kinase; AST aspartate transaminase; ALP alkaline phosphatase
*Normal distribution
aMean and standard deviation are shown only for analytes distributed normally
bReference intervals were calculated only for analytes with N > 20
Analytes comparison with non-Mediterranean Sea conspecifics and other Mediterranean Sea turtle species
The PCV and plasma chemistry of the recovered turtles in this study (correlations of which are shown in Fig. 1), along with previous findings of wild turtles from two Atlantic green turtle populations (Bolten and Bjorndal 1992; Page-Karjian et al. 2015) are presented in Table 2. The mean and SD are not presented when results exhibited a non-normal distribution. AST activity in this study was higher than in the two Atlantic green turtle populations (Bolten and Bjorndal 1992; Page-Karjian et al. 2015), but lower than in Mediterranean loggerhead turtles (Gelli et al. 2009). Triglyceride and cholesterol concentrations in this study were higher than in Mediterranean loggerhead turtles (Gelli et al. 2009) (threefold higher triglycerides concentration), but similar to that of Atlantic green turtles (Bolten and Bjorndal 1992; Page-Karjian et al. 2015). Total protein concentrations in the present study were higher than in Mediterranean loggerhead turtles (Gelli et al. 2009) and Atlantic green see turtles in the Bahamas (Bolten and Bjorndal 1992), and similar to that of Atlantic green population in Puerto Rico (Page-Karjian et al. 2015).
Table 2
Comparison of this study’s results with published blood analytes in two other green turtle (Chelonia mydas) populations from the Atlantic Ocean in conventional units
Analyte
Mediterranean green turtlesa (N = 72)
Atlantic green turtlesb (N = 100*)
Atlantic green turtlesc (N = 67**)
Median
Range
Mean
SD
N
Mean
SD
Range
95% RI
PCV (%)
28
19–37
30
4.6
26
35
3
26–42
24–53
Glucose (mg/dL)
108
65–164
113
25
24
114
15
87–167
48–120
Hemoglobin (g/dL)
10.8
6.2–14.4
10.8
2.5
13
CK (U/L)
833
234–7440
18
153–1992
AST (U/L)
327
96–1074
25
178
50
31–389
72–204
ALP (U/L)
62
21–327
24
43
16
13–95
28–100
Triglycerides (mg/dL)
174
77–412
199
99
16
172
85
43–413
Cholesterol (mg/dL)
155
117–289
19
217
53
73–365
49–215
Total protein (g/dL)
6.6
4.3–8.6
6.6
1.3
12
5.1
0.8
2.6–6.9
3.3–7.1
K (mmol/L)
4
3.1–6.4
24
5.3
0.6
4.1–6.9
3.7–5.8
N number of individuals; Med. the Mediterranean Sea; RI reference interval
Spearman’s rank correlations between packed cell volume and plasma chemistry analytes in recovered green turtles (C. mydas). Spearman’s ρ is represented by circle size and color with a color-coded legend
×
Comparison between life stages and correlations with body size
AST and ALP activity in juveniles (median AST, 371 U/L; median ALP, 71 U/L) were higher (P = 0.026 and P = 0.011, respectively) than in subadults (median AST, 255 U/L; median ALP, 40 U/L). In adult turtles, no ALP measurements were available, while AST activity was measured only in a single turtle, precluding comparisons with other age groups (Fig. 2). We also observed significant positive correlations between CCL and PCV (ρ = 0.475, P = 0.019) and cholesterol (ρ = 0.543, P = 0.02), and significant negative correlations between CCL and AST (ρ = − 0.471, P = 0.02), ALP (ρ = − 0.687, P < 0.001) and potassium (ρ = − 0.437, P = 0.033; Fig. 3).
Fig. 2
Aspartate aminotransferase (AST) and alkaline phosphatase (ALP) activity data of recovered green turtles from the Eastern Mediterranean Sea in partition to life stages. *Denotes statistically significant (P ≤ 0.05) difference between groups
Fig. 3
Statistically significant (P ≤ 0.05) Spearman’s rank correlations between curved carapace length (CCL) data and packed cell volume (PCV), potassium and cholesterol concentration, and aspartate aminotransferase (AST) and alkaline phosphatase (ALP) activity data of recovered green turtles from the Eastern Mediterranean Sea. The dashed line represents the trend line and the shaded area the 95% confidence interval of the trend line
×
×
Blood analytes at different treatment stages and in association with rehabilitation length
We examined the manifestation of treatment stage in blood indices in two ways: (a) a paired test for individuals sampled both upon admission at the rescue center and before their release; and (b) an independent test for newly admitted turtles, rehabilitating turtles, and recovered turtles. The results of both tests are summarized in Table 3.
Table 3
Comparison by paired Wilcoxon signed-rank test and a Kruskal–Wallis independent test of plasma chemistry analytes of green turtles from the Eastern Mediterranean Sea upon admission at the rescue centre, during rehabilitation, and upon recovery
Analyte
Paired test
Independent test
N
Turtles upon admission
Turtles upon recovery
Turtles upon admission
Turtles during rehabilitation
Turtles upon recovery
Median
Median
N
Median
N
Median
N
Median
Total protein (g/dL)
7
5.1
6.5
38
4.5a
22
5.2a
12
6.6b
AST (U/L)
10
417
319
CK (U/L)
18
7997b
27
1106a
16
769a
Potassium (mmol/L)
38
3.5a
30
3.8a,b
24
4b
N number of samples
a,bMedians sharing a common superscript are not significantly different
The paired test’s results showed significant increase in total protein levels (W = 1, P = 0.031) from admission to recovery, and a significant decrease in AST levels (W = 48, P = 0.037).
The independent test’s results also showed significantly higher total protein levels in recovered than in admitted (χ2 = 12.659, df = 2, P-adjusted = 0.001) and rehabilitating turtles (χ2 = 12.659, df = 2, P-adjusted = 0.049). Additionally, CK levels were significantly higher in admitted than in rehabilitating (χ2 = 12.75, df = 2, P-adjusted = 0.004) and recovered turtles (χ2 = 12.75, df = 2, P-adjusted = 0.004), and potassium levels significantly higher (χ2 = 6.653, df = 2, P-adjusted = 0.03) in recovered than in admitted turtles.
In correlation tests between different blood analytes and the percent of total rehabilitation time of turtles that eventually recovered and were released back to the sea, a statistically significant negative (though weak) correlation was determined with plasma CK activity (ρ = − 0.379, P = 0.006) and significant positive correlations with total protein (ρ = 0.388, P = 0.01) and triglyceride (ρ = 0.437, P = 0.001) concentrations were observed (Fig. 4).
Fig. 4
Statistically significant (P ≤ 0.05) Spearman’s rank correlations between percent of total rehabilitation time and biochemical analytes of green turtles from the Eastern Mediterranean Sea. The Y axes are in log10 scale. Percent of total rehabilitation time is represented by circle color with a color-coded legend
×
Potential predictors of rehabilitation success
To identify analytes with a predictive capacity for rehabilitation success rate, we initially tested data taken during the first 10% of the rehabilitation period for differences between non-survivor turtles and those that eventually recovered and returned to sea, which yielded insignificant results. We then proceeded to test data from 10 to 65% of rehabilitation time for difference in analytes between treatment outcome groups. There were insufficient data for group comparisons of hemoglobin, triglycerides, cholesterol, and total protein concentration, and CK activity. The PCV was significantly (P = 0.003) higher in recovered, released turtles (N = 21; median, 25%; range, 9–28%) than in non-survivors (N = 7; median, 14%; range, 18–34%). The ROC analysis for the PCV during 10–65% of the total rehabilitation period of turtles in the centre as a predictor of recovery and release yielded and AUC of 0.84 (Fig. 5).
Fig. 5
Receiver operating characteristic curve of PCV data of green turtles from the EMS during 10–65% of their total rehabilitation time
×
Discussion
This study described the PCV and several plasma chemistry analytes of green turtles, upon admission to a rehabilitation centre, during treatment and upon recovery, and suggests RIs for PCV, plasma glucose and potassium concentrations, as well as AST and ALP activity. The results of this study for recovered turtles from the EMS were compared with published data for a Mediterranean loggerhead population and two Atlantic green turtle populations. We examined plasma biochemical and hematological profile trends associated with body size and health status. PCV values were found to be indicative of recovery success rate after the initial treatment and during the rehabilitation phase.
This study encountered several challenges from, among other things, small sample sizes, a scarcity of published data, and differences in methodologies and reporting formats used in previous studies. The results published herein should, therefore, be regarded as of preliminary nature, and they comprise the initial parts of an ongoing research, expected to further elaborate and enhance these products.
Reference intervals of blood analytes in recovered turtles and comparison with non-Mediterranean Sea conspecifics and other Mediterranean Sea turtle species
Suggested RIs with 90% CI for their upper and lower limits were computed for five plasma chemistry analytes, while medians and ranges are provided for recovered green turtles for the remaining nine analytes. With relatively low numbers of green turtles present in the EMS and admitted in the ISTRC, the cohort size was limited. Nevertheless, as rehabilitation and research efforts are ongoing, and the database is continuously expanding, and we expect to compute additional RIs and validate the presently proposed ones. However, it is important to keep in mind the effects of captivity and feeding on analyte concentrations and activity in the process of RIs determination for turtles that went through rehabilitation.
Since vertebrate blood indices are affected by species, geographic area, ecological habitat, genetically distinct populations, sex, life stage, and more (Jensen et al. 1994; Maceda-Veiga et al. 2015; Page-Karjian et al. 2015), it is not a common practice to compare measurements from a specific population with values of different species or populations from other geographical areas. However, due to the scarcity of available data on green turtles in the Mediterranean Sea, such comparisons were the only ones applicable. It is important to note that environmental and nutritional conditions directly affect sea turtle plasma chemistry. Additionally, differences in laboratory equipment, reagents and methodology might also account for plasma chemistry variations between studies (Casal et al. 2009).
AST activity can be found in many soft tissues and is non-organ-specific (Bain 2011), although is routinely employed as a biomarker of hepatic stress in turtles. In loggerhead turtles rehabilitated in Croatia, AST levels were higher just prior to turtle release than upon arrival to the rescue centre (Jakšić et al. 2022), in contrast to that observed herein, which may be due individual turtle physiological status and condition (Jakšić et al. 2022).
The differences in plasma triglycerides and cholesterol concentrations between the current results in recovered green turtles and previous ones of wild Mediterranean Sea loggerhead turtles (Gelli et al. 2009) are likely species-related, also influenced by interspecific dietary differences between omnivorous loggerheads and herbivorous green turtles, and likely not attributed to feeding differences between wild and captive turtles, as triglyceride concentrations of rehabilitated Canary Islands loggerhead turtles were similar to those of the Mediterranean Sea (Bolten and Bjorndal 1992; Casal et al. 2009; Page-Karjian et al. 2015), while the current triglyceride concentrations were similar to those reported in wild green turtles from the Bahamas (Bolten and Bjorndal 1992).
Plasma total protein concentrations in this study are somewhat higher compared to those reported in Mediterranean Sea loggerhead turtles and Atlantic green turtles, both wild and rehabilitated (Bolten and Bjorndal 1992; Casal et al. 2009; Gelli et al. 2009; Osborne et al. 2010). The reasons for this are unclear, but species related differences (regarding loggerhead turtles), and dietary and environmental variations, as well as differences in laboratory methods likely played roles in these findings.
Comparison between life stages and correlations with body size
We observed higher plasma AST and ALP activity in juvenile vs. subadult recovered turtles, and a significant negative correlation between CCL and these analytes. CCL also correlated negatively with potassium concentrations, and positively with PCV and cholesterol concentrations.
It is hard to interpret variations in AST activity without CK measurements, as AST activity is prevalent in many tissues, including muscle and liver, while CK is muscle-specific and most of its plasma activity originates from muscle cells. Thus, if AST activity is increased, with no increase concurrently of CK activity, the former is very likely not of muscle origin, and may suggest liver damage (Bain 2011). Our results indicated higher AST activity in smaller turtles, also reported in green turtles from Taiwan (Fong et al. 2010) and Kemp’s ridley turtles (Perrault et al. 2020). It has been suggested by Perrault et al. (2020) that AST activity correlates negatively with CCL due to faster tissue growth at small turtle sizes, with a gradual decrease in growth rate as turtles become larger.
This trend of growth rate also applies to the process of bone formation, and accordingly, ALP activity is higher in juvenile turtles and decreases with an increase in CCL. These findings were observed by several other researchers in green and other sea turtle species (Bolten and Bjorndal 1992; Kakizoe et al. 2007; Stacy et al. 2018).
The negative size-correlation with potassium concentrations observed in this study was also reported by Stacy et al. (2018), who proposed this may suggest limited osmoregulatory capacity (i.e., decreased potassium excretion) of smaller turtles due to underdeveloped or small salt glands, or alternatively indicate changes in bone metabolism.
Positive size-correlation with PCV, as observed herein, is widely documented in captive and wild sea turtles of various species, including green turtles (Frair 1977; Wood and Ebanks 1984; Stamper et al. 2005; Kakizoe et al. 2007; Casal et al. 2009; Basile et al. 2012; Rousselet et al. 2013; Kelly et al. 2015; Stacy et al. 2018; Perrault et al. 2020). An increase in age and body size is associated with a concomitant increase in red blood cell (RBC) volumes and decrease in RBC counts (Frair 1977). Sea turtle diving activity increases with size, making higher PCV advantageous for meeting the increasing oxygen demands in larger turtles (Stacy et al. 2018).
Cholesterol concentrations were shown to increase with increasing body size in green, loggerhead, and Kemp’s ridley sea turtles, possibly in relation to dietary shifts and changes in hormone levels and reproductive state (Kakizoe et al. 2007; Arthur et al. 2008; Perrault et al. 2020), similar to the current findings.
Comparison between treatment stages and correlations with percent of rehabilitation time
In this study, plasma total protein, potassium, and triglycerides concentrations increased from rehabilitation up to recovery, while plasma AST and CK activities decreased significantly, consistent with previous results in Atlantic green turtles rehabilitated in GA, USA (Bloodgood et al. 2019). This change in plasma total protein and potassium concentrations likely reflects malnutrition or digestive disfunctions, cachexia and a negative energy balance upon admission, with gradual improvement in body condition and energy balance over the rehabilitation with a protein-rich and high-fat diet and administration of medical care. These findings agree with previous reports of stranded loggerhead turtles sustaining malnutrition and restricted food intake exhibiting significantly lower total protein, potassium, and triglyceride concentrations than healthy foraging turtles (Deem et al. 2009; Casale and Margaritoulis 2010; Osborne et al. 2010; Alberghina et al. 2015).
AST and CK are intracellular leakage enzymes found in myocytes. Their concomitant high activities in turtles upon admission, decreasing later during rehabilitation, towards recovery, are consistent with muscle damage occurring due to trauma or myopathy prior to admission, which gradually improved during treatment. Increased CK activity was also reported in stranded loggerhead turtles compared to healthy turtles (Deem et al. 2009; Flower et al. 2015), similar to AST activity (Casal et al. 2009).
Potential predictors of rehabilitation success
In samples obtained during the rehabilitation phase, PCV was significantly higher in turtles which ultimately recovered than in non-survivors. The ROC curve AUC of PCV during 10–65% percent of the total rehabilitation time as a predictor of the overall survival (0.84), indicates that PCV is a good prognostic marker (Ekelund 2012).
Pagano et al. (2019) examined and compared hematological and chemistry analytes of survivor and non-survivor loggerhead turtles stranded and rehabilitated in Sicily. In that study, the PCV was not measured, but the RBC count and hemoglobin concentration were recorded. Severe anemia was noted in all turtles upon admission, and both analytes were significantly lower in non-survivor turtles than in survivors.
No hematological or biochemical differences were detected to indicate a prognosis in the data of admitted turtles, likely due to all turtles suffering from poor health and possible anaemia. However, later during rehabilitation, turtles with lower PCV displayed lower chances of surviving, which is useful in alerting turtle caregivers of individuals requiring more intensive care and monitoring, as PCV measurements are simple, cost-effective, and widely available.
It should be noted, however, that the non-survivor sample size (N = 7) was three times smaller than the survivor sample size (N = 21), indicating the need for further assessments in this regard for the Mediterranean population.
Conclusions
This study describes several plasma chemistry analytes and the PCV of green turtles, from their admission to the rescue centre, and through their rehabilitation up to their recovery, and preliminary RIs for several analytes in recovered turtles were calculated. Results were compared with those of conspecifics in other geographical areas and with other species in the Mediterranean Sea and other areas, and similarities and differences linked to varying physiological factors are discussed. Associations of plasma chemistry analytes and PCV with turtle size and health status were determined. Additionally, PCV measured during rehabilitation was a good predictor of survival and chance of recovery, which might provide turtle caregivers clinically useful information.
This study’s novel physiological data of a sentinel endangered species in the EMS ecosystem will contribute to its better conservation in the area, serving as future data basis, useful for research and care of green turtles. Large knowledge gaps regarding the physiology of these animals and the environmental effects on it remain unaddressed, and further research is essential to shedding light on matters such as their population condition and trend, migration patterns, life history, anthropogenic effects, and more.
Acknowledgements
The authors would like to thank Maya Lalzar, Eli Shemesh, and Olga Rybak for their assistance in this study. We would also like to express our appreciation to the European Cooperation in Science & Technology (e-COST), CA22160, for their assistance in this research.
Declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
This work has received approval for research ethics from The Ministry of Environmental Protection, Israel, and a proof/certificate of approval is available upon request.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Plasma chemistry and hematology of Eastern Mediterranean Sea green turtles undergoing rehabilitation
verfasst von
Tal Starostinetsky Malonek Aviad Scheinin Natascha Wosnick Nadav Davidovich Peleg Itay Shlomi Piha Itamar Aroch Dan Tchernov Yaniv Levy Rachel Ann Hauser-Davis Danny Morick
