Blood concentrations of lactate, glucose and corticosterone in dispersing hatchling sea turtles

Once they emerge from their nest, sea turtle hatchlings frantically crawl down the beach towards the sea and then engage in a frenzy swim, which rapidly takes them off-shore to reach oceanic currents (Bustard, 1972; Salmon and Wyneken, 1987). During this period hatchlings face high near-shore predator pressure that decreases as they move further offshore (Gyuris, 1994; Gyuris, 2000; Salmon et al., 2009; Whelan and Wyneken, 2007). Reptiles such as sea turtles have a limited aerobic capacity (Bennett, 1978) and during intensive exercise they supplement their energy requirements with anaerobic metabolism (Bennett, 1982) with a subsequent production and accumulation of lactate (Gleeson, 1996). Studies on the natural history and physiology of nest emergence, beach crawl and the off-shore frenzy swimming dispersal of hatchling sea turtles indicate that these processes are extremely vigorous and energy demanding (Wyneken and Salmon, 1992; Wyneken, 1997; Booth, 2009; Pereira et al., 2012). Although primarily fuelled by aerobic metabolism (Wyneken, 1997; Booth, 2009; Pereira et al., 2012), tissue lactate concentrations increase during these activities indicating that anaerobic metabolism also plays a major role at this time (Dial, 1987; Baldwin et al., 1989; Hamann et al., 2007).

Sea turtle hatchlings typically exhibit frantic swimming behaviour when they first enter the water, but this decreases in the first few hours before they settle into a sustained swimming effort for up to 24 hours (Wyneken and Salmon, 1992; Wyneken, 1997; Booth, 2009; Pereira et al., 2011; Pereira et al., 2012). However, the intensity of swimming varies between species (Wyneken and Salmon, 1992; Pereira et al., 2011). Green turtle (Chelonia mydas) hatchlings are the most vigorous swimmers followed by loggerhead turtle (Caretta caretta) hatchlings, with flatback turtle (Natator depressus) and leatherback turtle (Dermochelys coriacea) hatchlings being less vigorous (Wyneken and Salmon, 1992; Pereira et al., 2011; Pereira et al., 2012). These differences in swimming intensity are expected to be reflected in differences in the use of anaerobic metabolism and therefore differences in blood lactate concentrations.

In reptiles, aerobic and anaerobic metabolism is influenced by steroid hormones, which control metabolic processes (Nelson, 2005; Neave, 2008). Glucocorticoid hormones influence physiological processes such as lipolysis, gluconeogenesis, lactate release from muscles (Neave, 2008) and increases in blood glucose concentration (Munck and Koritz, 1962). Glucocorticoids are involved in adjusting physiological process in response to environmental cues particularly in animals that undergo seasonal changes as part of their life history, such as migration, breeding and hibernation (McEwen and Wingfield, 2003). Emerging from a nest, crawling down the beach then entering a frantic swim may be a series of events that are influenced to some degree by glucocorticoids (Hamann et al., 2007). In reptiles, corticosterone (B) is one of the major glucocorticoids produced (Nelson, 2005) and is known to vary in concentration at different stages of sea turtle life history including the natal dispersal phase (Hamann et al., 2002; Hamann et al., 2005; Hamann et al., 2007). Hence, as with blood lactate concentrations, the patterns of exercise activity exhibited by sea turtle hatchlings are expected to result in changes in blood corticosterone concentration.

In this study we investigate blood lactate, glucose and corticosterone concentration in hatchling turtles through the beach crawl and early frenzy swim to test the hypothesis that these parameters will vary in a systematic way with the varying exercise intensity of different species. In particular, we predict that loggerhead turtles experience intermediate changes in these physiological variables as a result of their moderate swimming effort and that flatback hatchlings experiences the smallest changes because of their relative low swimming effort (Pereira et al., 2011; Pereira et al., 2012). This is the first study to report data regarding changes in concentrations of blood lactate, glucose and cortiosterone in flatback hatchling turtles and data regarding glucose and corticosterone concentrations in loggerhead hatchling turtles during the early stages of the dispersal phase.

This study was conducted at Mon Repos Conservation Park (24°48S, 152°27E), located on the east coast of the Australian mainland (Pfaller et al., 2009). Annual nesting of loggerhead (Caretta caretta) and flatback (Natator depressus) turtles occur every year at this rookery (Limpus et al., 1984; Limpus and Limpus, 2003; Limpus, 2007; Limpus, 2008).

Data collection took place between the 21st January and the 18th February 2011. The beach was patrolled throughout the night to monitor nests of loggerhead and flatback turtles. When hatchlings were found emerging from a nest, a total of 18 hatchlings were collected from the nest and transported dry to the laboratory in a bucket lined with sand, a process that took around 15 minutes. During this transport time hatchlings were actively crawling around inside the bucket, a behaviour that is similar to their terrestrial movement from the nest to the sea.

Each trial consisted of one terrestrial treatment (crawling) and five swimming treatments (15 minutes, 30 minutes, 1 hour, 2 hours and 4 hours). We did not collect blood at emergence because it was difficult to sample on the beach. Once in the laboratory three hatchlings were sampled for blood and classified as a ‘crawling’ treatment. The remaining 15 hatchlings were allocated to the swimming treatments. Each hatchling was placed in a separate tank (34 cm long×28 cm wide×19 cm high) filled with 16 cm of sea water at ambient temperature (25–28°C) and a low intensity light was placed at one end of the tank to encourage unidirectional swimming. Each hatchling was fitted with a lycra harness attached to a monofilament tether that was fixed to a wooden stake attached to the top of the tank. The tether length was adjusted so that hatchlings could swim freely but could not touch the sides or bottom of the tank. The stretched tether provided resistance to forward motion that encouraged continuous swimming. At the end of each sampling period three hatchlings had a blood sample taken and were then released into the ocean.

A 200–300 µl blood sample was collected from a branch of the jugular vein using a 27 gauge needle and 1 ml heparinized syringe following the method of Bennett (Bennett, 1986). Blood lactate and glucose concentrations were immediately analysed with an I-STAT one Blood Analyzer (Heska, Colorado USA) using GC4+ and 6+ cartridges (Abbott Laboratories, USA), respectively. The remaining blood was centrifuged to separate the blood cells (discarded) from the plasma that was preserved at −10°C until later analysis of corticosterone. Plasma corticosterone was analysed by radioimunoassay (RIA) as described in Bradley using a 10 µl sample (Bradley, 1987; Bradley, 1990). Corticosterone standards ranged from 0 to 1000 pg and standards were made up in methanol using corticosterone purchased from Sigma Chemical Co, USA. Radiolabelled corticosterone ([1,2,6,7-3H(N)] corticosterone) was purchased from PerkinElmer NEN Radiochemicals, USA and 20,000 dpm was added to both standards, samples and external controls. A standard curve was plotted using AssayZap software (Biosoft, Cambridge, UK). The intra- and inter- assay coefficients of variation were 8.4 and 7.1% respectively. The assay sensitivity was 10 pg.

Blood concentrations of corticosterone, lactate and glucose were analysed using a MANOVA, followed by a Repeated Measures ANOVA (each clutch was considered as just one individual that was sampled repeatedly over time). Results were reported as means ± SE and statistical significance assumed if P<0.05.

A total of 132 loggerhead hatchlings and 79 flatback hatchlings had blood samples taken although there was not enough blood in some samples to perform measurements on all three variables (Table 1). Lactate and corticosterone blood concentration data were log transformed to comply with the statistical assumptions of normality and homoscedasticity before statistical analysis. The MANOVA analysis revealed significant differences between all variables (P<0.01). The Repeated Measures ANOVA showed significant differences over time but not between species and with the exception of lactate there were no species*time interactions (Table 2).

Loggerhead turtle hatchling blood lactate, glucose and corticosterone concentration decreased significantly with time from crawling until trials ended after 4 hours of swimming (Table 2; Fig. 1). Most of these decreases in concentration occurred between crawling and two hours of swimming (Fig. 1). Flatback turtle hatchlings had a similar pattern, with blood concentrations of all variables decreasing throughout time (Table 2), although there was a slight increase in blood lactate after 1 hour of swimming (Fig. 1) and this was the cause of the interaction between species and time for this variable.

Our results show that flatback and loggerhead turtle hatchlings have a similar use of anaerobic metabolism as indicated by the similar patterns of change in blood lactate concentration. In both species, anaerobic metabolism decreased with time but the use of anaerobic metabolism persisted for longer in flatback turtle hatchlings than in loggerhead hatchlings. Dial (loggerhead turtles), Baldwin et al. (loggerhead and green turtles) and Hamann et al. (green turtles) have reported lactate concentrations at different times during the beach crawl and early stages of the frenzy swim in hatchling sea turtles (Dial, 1987; Baldwin et al., 1989; Hamann et al., 2007). Although all studies indicate anaerobic metabolism plays an important role during this period, results differ between studies due to differences in methodology and species. Dial and the current study found lactate to be highest after the beach crawl (Dial, 1987), while Baldwin et al. found lactate continued to accumulate in blood when hatchlings swam across the reef flat (Baldwin et al., 1989). Baldwin et al. reported a fall in blood lactate in hatchlings swum in an aquarium because they power-stroked only briefly and remained mostly on the surface with the head above the water, and for loggerhead turtles blood lactate concentration was just 1.7 mmol/L after 15 minutes of ‘swimming’ in the aquarium (Baldwin et al., 1989). In the current study, however, hatchlings swam almost continuously as they were tethered and could not touch the sides or bottom of the swimming container and loggerhead hatchling blood lactate concentration was four times greater averaging 6.9 mmol/L after 15 minutes of swimming. Baldwin et al. found that blood lactate concentrations peaked at ∼13 mmol/L after 15 minutes of swimming across the reef flat and that they were greater in green turtle hatchlings than in loggerhead turtle hatchlings (Baldwin et al., 1989). In contrast, Hamann et al. found blood lactate concentrations in green turtle hatchlings to be only 4.5 mmol/L after 30 minutes of swimming but to be extremely high (24 mmol/L) after two hours of swimming (Hamann et al., 2007). The pattern of blood lactate concentration change reported for green turtle hatchlings reported in Hamann et al. (Hamann et al., 2007) is unexpected because green turtle hatchlings generally decrease their swimming effort as the time spent swimming increases (Burgess et al., 2006; Booth, 2009; Ischer et al., 2009; Booth and Evans, 2011).

As expected, blood lactate concentrations of loggerhead hatchlings were lower than those reported for green turtle hatchlings, which is consistent with the less vigorous swimming behaviour of loggerhead hatchlings compared to green turtle hatchlings (Pereira et al., 2011; Pereira et al., 2012). We also expected blood lactate concentrations of flatback turtle hatchlings to be lower than loggerhead turtle hatchlings because they are less vigorously swimmers (slower power-stroke rate and less frequent power-stroking bouts) (Pereira et al., 2011). However, this was not the case, as lactate concentration was not significantly different between the species during the beach crawl and first hour of swimming. Indeed, contrary to our expectation, from two hours onward blood lactate concentration in flatback hatchlings increased slightly although it continued to decrease in loggerhead hatchlings. A possible explanation for this increase in blood lactate concentration of flatback hatchlings may be that lactate is being released from the muscles to the blood, which will then be transported to the liver and converted back into glucose (gluconeogenesis), either for immediate use or to be stored as glycogen.

The fact that anaerobic metabolism in flatback hatchlings is similar or even exceeds that of loggerhead hatchlings, but swimming effort is relatively lower in flatback hatchlings (Pereira et al., 2011), suggests that it is the aerobic component of flatback hatchling swimming effort that is relatively lower compared to loggerhead hatchlings. This hypothesis is supported by the finding that aerobic metabolism of flatback turtles decreases much more sharply than in loggerhead turtle hatchlings within the first two hours of swimming (Pereira et al., 2012).

The pattern of changes in the concentrations of glucose and corticosterone seems to reflect the changes in blood lactate concentration. Both aerobic and anaerobic muscle metabolisms are fuelled by glucose and during heavy exercise glucose is mobilised from glycogen stores in the liver. During prolonged exercise liver glycogen stores become depleted and other sources of fuel such as lipid and proteins are gradually mobilised (Brooks, 1987; Horowitz and Klein, 2000; Hamann et al., 2007). Hence in loggerhead and flatback turtle hatchlings elevated concentrations of corticosterone during crawling and early swimming are correlated with glucose being released from the liver and a consequent increase in blood glucose concentration. Subsequently blood glucose concentration falls as the liver glycogen stores are depleted.

The ‘predation risk’ hypothesis suggests that offshore swimming behaviour of hatchling sea turtles is moulded by the interaction of three factors that contribute to survivorship: predator pressure in near-shore waters, swimming distance to oceanic currents and hatchling size (Chung et al., 2009; Salmon et al., 2009). Green turtle hatchling swimming behaviour is more vigorous than that of loggerhead and flatback hatchlings (Pereira et al., 2011; Pereira et al., 2012), which indicates a strategy moulded primarily by the need for a rapid escape from near shore predator rich waters (Gyuris, 1994). Loggerhead turtle hatchlings swimming behaviour is similar to but not as vigorous as green turtle hatchlings (Pereira et al., 2011; Pereira et al., 2012; Wyneken, 1997), which suggests that is also primarily moulded by a need to rapidly escape predator rich near shore waters. Flatback hatchlings, however, experience a sharper decrease in their swim effort and aerobic metabolism within the first two hours of the swimming frenzy (Pereira et al., 2011; Pereira et al., 2012) and they maintain a similar level of swimming activity for the first four days after entering the water (Salmon et al., 2009). Our results do not allow us to make strong inferences about the role of anaerobic metabolism on dispersal strategies but it supports the idea that near-shore predation pressure is a potential driver for the frantic and energetically expensive swimming effort typical of the initial stages of the frenzy swim.

In summary the current study and previous studies (Dial, 1987; Baldwin et al., 1989; Hamann et al., 2007) indicate anaerobic metabolism is used extensively during the beach crawl and at least the beginning of the frenzy swim during natal dispersal of sea turtle hatchlings. The additional use of anaerobic metabolism allows hatchlings to maximize their physical activity and thus minimises the time they spend on the beach and in the shallow near-shore waters where predators are in their highest concentration.

This research conforms to Australian Animal Welfare laws and was approved by a University of Queensland Animal Ethics Committee (approval no. SBS/228/10). We thank Christina Burger, the Mon Repos Conservation Park staff and volunteers and the DERM Turtle Conservation Volunteers for the help provided during the field work. This study was co-financed by the European Social Fund (ESF) and Fundação para a Ciência e a Tecnologia (FCT) as part of a doctoral scholarship [SFRH/BD/44139/2008 to C.M.P.].