Effect of water temperature on diel feeding, locomotion behaviour and digestive physiology in the sea cucumber Apostichopus japonicus

Trade in sea cucumbers dates back several centuries owing to their health benefits and medicinal properties. In response to a dwindling supply from wild stocks and increasing demands causing a surge in market prices, commercial aquaculture of sea cucumbers has begun recently in many countries (Purcell et al., 2014; Toral-Granda et al., 2008). Apostichopus japonicus (Selenka 1867) is one of the most important commercial aquaculture species. The A. japonicus industry has now overtaken the traditional shrimp and fish aquaculture industries, and A. japonicus generates the highest single-species output value and profit in China (Yang et al., 2015). It is distributed mainly along the coasts of Northeastern Asia, including eastern Russia, northern China, Japan, North Korea and South Korea (Hamel and Mercier, 2008; Okorie et al., 2008). Usually, A. japonicus is cultured in shallow seas (sea ranching) or ponds (cofferdam). In recent years, intensive indoor culture has expanded rapidly, becoming an important rearing method in China (Yang et al., 2015). Understanding the behavioural and physiological features of A. japonicus under different environmental conditions is beneficial not just to understand the natural behaviour of the species, but also to improve the rearing techniques for sea cucumbers.

Water temperature represents the most pervasive aspect of the environment affecting marine ectotherms, varies markedly on a variety of spatial and temporal scales and is one of the most important environmental factors affecting growth and physiological performance in aquatic animals (Buentello et al., 2000). Previous studies have demonstrated the effects of water temperature on growth (An et al., 2007; Dong and Dong, 2006; Dong et al., 2006; Yang et al., 2005), immune response (Wang et al., 2008) and body composition (Dong et al., 2006) of A. japonicus. However, our understanding of the behavioural responses of A. japonicus to temperature variation lags well behind our understanding of physiological responses.

The diel cycles of sea cucumbers vary among different species, with maximal activity being diurnal, nocturnal or crepuscular (Wheeling et al., 2007). However, the activity may be scheduled to coincide with regular changes in many environmental factors. The effect of light on rhythms and diel cycles in sea cucumbers has been well described (Chen et al., 2007; Dong et al., 2011; Dong et al., 2010; Mercier et al., 1999), but the effect of water temperature on their behavioural rhythms has received less attention. Temperature-induced changes in diel activity have been observed mainly in fish (Brown and Mackay, 1995; Fraser et al., 1993; Hurst and Duffy, 2005; Reebs, 2002). To our knowledge, relevant studies on sea cucumbers have been conducted mainly in the tropical species Holothuria scabra (Mercier et al., 2000; Purcell, 2010; Wolkenhauer, 2008).

Another potential technique to study diurnal variations in feeding is based on the measurement of digestive enzyme activities (Mata-Sotres et al., 2016). Previous studies have indicated that the diel digestive physiology of fish, such as sea bass (Tillner et al., 2014; del Pozo et al., 2012), sea bream (Zeytin et al., 2016; Montoya et al., 2010) and goldfish (Vera et al., 2007), was mainly entrained by the light–dark cycle, food types and feeding times. Water temperature was one of the important factors affecting the digestive physiology of sea cucumbers (Gao et al., 2009). The diel digestive rhythms of A. japonicus at different temperatures remain unclear, and it is crucial to determine the optimum feeding time and frequency in aquaculture practice.

This study was carried out under controlled laboratory conditions to ensure that the effects of temperature were separate from that of other environmental factors. In this study, we describe the behavioural patterns and determine the digestive physiology of the sea cucumber A. japonicus under different water temperatures based on laboratory experiments. In particular, we examined diel feeding activity patterns and tentacle insertion rate, as well as the change of digestive enzyme activities, across the range of temperatures likely encountered throughout the year by A. japonicus. In addition, diurnal and nocturnal ingestion rate (IR) and locomotion behaviour of A. japonicus were compared.

Sea cucumbers were collected from Tianheng Sea Cucumber Farm, Shandong Province, China, and transported to the laboratory (Qingdao National Ocean Science Research Center). All individuals were acclimated in six large tanks (1000 l⁻¹) at 6–14°C, 30–32 ppt salinity, a pH of 7.8–8.2, >5.5 mg l⁻¹ dissolved oxygen and at a 14 h–10 h light–dark photoperiod. During the acclimation period, the sea cucumbers were fed once at 08:00 h using a homemade diet containing 40% Sargassum powder and 60% sea mud processed into a cylindrical form. Healthy undamaged sea cucumbers (body length: 5.2±0.6 cm, wet body weight: 22.7±1.2 g; means±s.d.) were selected and used in the series of experiments.

Five temperatures (8, 12, 16, 20 and 24°C) were selected to approximately represent the overall temperature range (0–30°C) found in the natural habitat of A. japonicus (Dong and Dong, 2006). The experiments at different water temperatures were conducted consecutively. When the ambient water temperature reached the lower experimental temperatures (8 and 12°C), the experiments were initiated. Aerated water in a 4000 l tank was heated to the corresponding higher experimental temperatures (16, 20 and 24°C) using one or two 2000 W electric heaters controlled by a thermostat, then pumped into the experimental tanks. The air temperature of the experimental room was maintained at the corresponding higher experimental water temperature (16, 20 and 24°C) by air conditioning. The sea cucumbers were fed the same food as during the acclimation period for all studies except the tentacle observation trials. The test individuals were moved into the experimental tank, where they were raised from holding temperature to the experimental temperature at a rate of 0.2°C h^–1. Each experiment began with a 24 h (8, 12 and 16°C) or 48 h (20 and 24°C) period of temperature acclimation. All experiments were conducted in a dark environment except the tentacle observation. In this study, 08:00 h to 20:00 h was defined as ‘day’ and 20:00 h to 08:00 h was defined as ‘night’.

The experiments were conducted in four glass tanks (60×60×50 cm height×width×depth). To decrease possible interference from external factors, the sides and bottom of the tanks were covered with opaque paper. Ten sea cucumbers were placed in each of the four experimental tanks for the 24 or 48 h acclimation period. Each trial was run for 4 days, after which the animals were removed and the tanks were drained, cleaned and refilled. The animals were fed once per day, and half of the water was exchanged at 08:00 h. The water depth was maintained at 30 cm during the experiment. Charge-coupled device (CCD) cameras (Hikvision, DS-2CC11A2P-IR3, China) with the infrared systems were mounted 1.5 m above the experimental tank and connected to a video recorder to record the feeding and locomotion behaviours of sea cucumbers.

Feeding behaviour was determined by observing the individuals' position in the aquarium and the decrease of food on the aquarium bottom. The proportion of feeding sea cucumbers was determined every 30 min based on the videos. Four consecutive feeding values from 30 min intervals were used to generate a single mean value every 2 h. Locomotion behaviour observations were based on analyses of the final 30 min of the video recordings made during each 2 h segment of the experiment. A grid dividing the tank into four quadrate sections was overlaid on the monitor during playback, and the number of line crossings was used as an index of horizontal distance travelled along the bottom of the tank. A sea cucumber was considered actively moving if it made a horizontal movement of more than three body lengths during the 30 min. The mean proportion of moving sea cucumbers from 08:00 h to 20:00 h was used as the day value, and the mean from 20:00 h to 08:00 h was used as the night value. Four consecutive days' values of feeding and locomotion behaviours were used to generate a single mean value.

Tentacle locomotion during feeding was recorded using an upward-looking digital camera (Canon IXUS 125 HS) located below the glass aquarium (60×60×50 cm) according to the method of Hudson et al. (2005). A lamp with dim light beneath the tank allowed observations to be made in darkness. Four individuals of each temperature treatment were observed and the rates of tentacle insertion were determined from 100 consecutive tentacle movements.

where W_o is the dried mass of the offered food (mg); W_u is the dried mass of the uneaten feed (mg); W_sc is the wet mass of the sea cucumber in the tank (g); and t is time (h).

For each temperature treatment, 72 sea cucumbers were introduced into 12 tanks (45×35×30 cm), acclimated for 1–2 days and maintained for 10 days. The sea cucumbers were fed excessively at 08:00 h every day. At the end of the experiment, six individuals were chosen randomly every 2 h during a 24 h cycle (12 sampling points) from the tanks and dissected to obtain intestines. The intestines were then cut longitudinally and washed thoroughly in ice-cold 0.84% normal saline. After rinsing, the gut was blotted dry with filter paper, frozen quickly in liquid nitrogen and then stored at −80°C until analyzed. Two replicate samples were mixed into one sample to make sure each one was enough to determine the activities of lipase and amylase enzymes.

Activities of lipase and amylase were measured using commercial assay kits (Nanjing Jiancheng, Bioengineering Institute, Nanjing, China). Lipase activity was assayed by the simplified turbidimetric assay. Amylase activity was assayed via iodine spectrophotometry. Activities of lipase and amylase were expressed as U g⁻¹ protein and U mg⁻¹ protein, respectively.

All statistical analysis was performed with the SPSS 19.0 for Windows statistical package. A two-way repeated measures (RM) analysis of variance (ANOVA) followed by Bonferroni test for post hoc multiple comparisons was used to test the effect of the interaction between time and temperature on the proportion of feeding sea cucumbers. The time of tentacle insertion, IR, proportion of moving sea cucumbers and mean digestive enzyme activities at different temperatures and digestive enzyme activities at different times were subjected to one-way ANOVA followed by post hoc multiple comparisons with Tukey's test. The differences between day and night of IR and the proportion of moving sea cucumbers were compared using an independent sample t-test. The probability level of 0.05 was used for rejection of the null hypothesis, and all data were presented as means±s.d.

A two-way repeated measures ANOVA showed that there was no significant interaction between temperature and time (interaction, F_44,132=0.66, P=0.944) on the proportion of feeding sea cucumbers. However, both temperature and time had an independent significant influence on the proportion of feeding sea cucumbers (temperature, F_4,12=122.08, P<0.001; time, F_11,33=8.30, P<0.001). Concerning the effect of temperature alone, the maximum proportion of feeding sea cucumbers was observed at 16°C (23±4%), with no significant difference when compared with 12 and 20°C groups (post hoc analysis with a Bonferroni adjustment test, P>0.05). However, the proportions of feeding sea cucumbers in these three groups were significantly higher than those of the 8 and 24°C groups (P<0.05). The proportion of feeding sea cucumbers at 24°C was the lowest, significantly lower than that observed at 8°C (Fig. 1, P>0.05).

All temperature treatments displayed a similar feeding pattern, with the proportion of feeding sea cucumbers increasing gradually from 14:00 h to 20:00 h, then reaching feeding peak before decreasing (Fig. 1). The 8 and 24°C treatments showed a feeding peak at 02:00 h to 04:00 h, whereas the feeding peak of other treatments occurred during 00:00 h to 02:00 h (Fig. 1).

Frame-by-frame analysis of the video footage showed that the feeding process of A. japonicus involved several coordinated movements. Usually, 50% of all tentacles (10) extended from the oral crown and moved towards the sediment surface. Each tentacle contained several tree-like branches, and each branch comprised several sub-branches. As the tentacle approached the sediment, it slowly opened outwards, presumably to maximize the available surface area for subsequent particle adhesion. The tentacle was then pressed onto the sediment surface and spread out. Once the tentacle contacted enough food particles, it retracted and closed the distal branched apparatus inwards and upwards towards the mouth. The tentacle released and the food was dropped into the oral mouth (Fig. 2). The sea cucumber would repeat the action continuously during its feeding.

No significant difference in the tentacle insertion rate was observed between the treatments of 8, 12, 16 and 20°C (one-way ANOVA, F_3,12=0.45, P=0.722). However, no tentacle locomotion was observed at 24°C. The total time for a tentacle to be placed onto the sediment, collect food, pass it to the mouth and feed again was 46.21±2.79, 44.80±2.22, 45.74±3.16 and 47.02±2.80 s at 8, 12, 16 and 20°C, respectively (Table 1).

The IR of sea cucumbers was significantly affected by water temperature (one-way ANOVA, F_4,15=31.85, P<0.001, Fig. 3A). Specifically, the IR of sea cucumbers at 16°C (4.9±0.4 mg g⁻¹ h⁻¹) was significantly higher than at other temperatures (post hoc Tukey's test, P<0.05) and IR at 24°C was significantly lower than those observed at other temperatures (P<0.05). No significant difference in IR was observed between the 8, 12 and 20°C treatments (P>0.05). Significant differences in IR were observed between day and night within all temperature treatments (t-test, P<0.05, Fig. 3B).

The proportion of moving sea cucumbers was also significantly affected by water temperature (one-way ANOVA, F_4,15=30.87, P<0.001, Fig. 4A). The proportion of moving sea cucumbers increased as the temperature increased from 8 to 16°C, then decreased above 16°C. The proportion of moving sea cucumbers at 16°C (42.9±4.4%) was significantly higher than that observed at 8, 20 and 24°C (post hoc Tukey's test, P<0.05). No significant differences were observed between 12 and 16°C (P>0.05). At 24°C, the proportion of moving sea cucumbers was significantly lower than that at 8, 12 and 16°C (P<0.05). No significant difference was observed between 20 and 24°C (P>0.05). Significant differences in the proportion of moving sea cucumbers were also observed between day and night within all temperature treatments except 24°C (Fig. 4B).

The lipase activity of sea cucumbers at all temperatures showed significant differences over time during the 24 h cycle (one-way ANOVA, P<0.05, Fig. 5A). Peak lipase activity of the sea cucumbers at 8, 12 and 24°C occurred at 24:00 h, whereas the highest activity for individuals at 16 and 20°C occurred at 22:00 h. Amylase activity showed significant differences over time at 8 and 12°C (one-way ANOVA, P<0.05), but no significant differences over time were found for the 16, 20 and 24°C treatments (P>0.05, Fig. 5B). The highest amylase activity at 8°C appeared at 22:00 h, whereas peak amylase activity for individuals at 12°C occurred at 02:00 h.

Both the mean lipase and amylase activities were significantly affected by water temperature (lipase: F_4,10=83.93, P<0.001; amylase: F_4,10=9.14, P=0.002, Fig. 5). Specifically, mean lipase activities at 12, 16 and 20°C were significantly higher than at 8 and 24°C (post hoc Tukey's test, P<0.05). The highest lipase activity was 12.88±0.20 U g⁻¹ protein at 20°C, and the lowest lipase activity was 6.92±0.34 U g⁻¹ protein at 24°C (Fig. 5A). No significant differences in mean amylase activity were found between the 8, 12, 16 and 20°C (P>0.05), but they were all significantly higher than at 24°C (P<0.05, Fig. 5B).

The responses of feeding activity to different water temperature vary among species. As a tropical species, H. scabra decreased its feeding activity from 9.8 to 0.8 h day^–1 as the temperature changed from 24 to 17°C (Wolkenhauer, 2008). The sea cucumber species in the present study, A. japonicus, is a temperate species that survives in a temperature range of 0–30°C (Dong and Dong, 2006). However, growth occurs only between 12 and 21°C, and the optimal temperature for food consumption and growth is 15–18°C (An et al., 2007; Dong et al., 2006; Yang et al., 2005). The present study showed that feeding proportion peaked at 16°C, and the proportion of feeding sea cucumbers at the highest temperature treatment (24°C) was significantly lower than in the other groups. This finding was consistent with the results obtained for IR, which demonstrated that the highest and the lowest IRs were found at 16 and 24°C, respectively. These results were supported by previous studies by Dong et al. (2006), which showed that juvenile A. japonicus displayed the highest specific growth rate (SGR) at 16–18°C and the lowest SGR at 24°C. It can thus be concluded from the present study that 16°C is the optimal water temperature for the feeding activity of A. japonicus, and high temperature (≥24°C) had negative effects on feeding activity of this species.

Sea cucumbers have previously been shown to exhibit physiological adaptations that parallel the behavioural switches that occur with seasonal changes in temperature (Hamel and Mercier, 1996, 1998; Singh et al., 1999; Swan, 1961). The present study examined the direct effects of water temperature on the diurnal variation of feeding activity of sea cucumbers using infrared-light-assisted laboratory observations. Apostichopus japonicus exhibited a higher feeding proportion from 22:00 h to 06:00 h, and the feeding peak occurred at 00:00 h to 04:00 h. Also, they had higher IRs at night than during the day. Our study showed that A. japonicus demonstrated a distinct nocturnal feeding activity pattern at all temperature treatments. While water temperature significantly affected the activity levels of A. japonicus, it did not alter the overall feeding rhythm or timing of feeding peaks.

Sea cucumbers feed by extending their buccal tentacles either into or over the sediment surface (deposit feeders) or into the water column (suspension feeders) using a variety of tentacle forms (Roberts and Moore, 1997). Factors that are known to cause variation in the tentacle insertion rate of sea cucumbers include body size (Sun et al., 2015), current speed (Holtz and MacDonald, 2009; Singh et al., 1999), and food quality or concentration (Singh et al., 1998, 1999). These factors might counteract or exacerbate the effect of temperature on tentacle locomotion in the wild. This study aimed specifically to exclude those variable factors to find a possible underlying pattern in response to water temperature alone. Our results indicated that water temperature had no influence on the tentacle insertion rate, and this corresponded well with a previous report, which showed that tentacle insertion rates of Cucumaria frondosa were not significantly related to temperature (Singh et al., 1999). Combined with the results that water temperature affected the IR, we may assume that the influence of water temperature on feeding behaviour was mainly reflected in the duration of feeding. There are two possible reasons for the lack of tentacle locomotion at 24°C. First, high temperature stopped the feeding behaviour of A. japonicus. Second, the light or the presence of humans for observation exacerbated the cessation of feeding, given that A. japonicus prefers to feed in a dark environment.

The locomotion of many aquatic animals is significantly related to water temperature (Reynolds, 1977). For example, the sea urchin Abatus ingens increased their displacement activity with only a 1°C increase in temperature (Thompson and Riddle, 2005). The amount of time during the day that individuals of the sea cucumber species H. scabra spend buried (no movement) increased with decreasing temperature from 6.7 h per day at 24°C to 14.5 h per day at 17°C (Wolkenhauer, 2008). A long period of high water temperature would induce A. japonicus to enter aestivation, during which they exhibited little movement and depressed metabolic activity (Yang et al., 2015). In the present study, the proportion of moving sea cucumbers increased and then decreased with increasing temperature. Apostichopus japonicus became inactive in the present study at 24°C, which is consistent with the reported thermal threshold (∼24°C) for aestivation in the field (Yang et al., 2005). A laboratory investigation conducted by Kato and Hirata (1990) also found that A. japonicus moved more (49.6 m day⁻¹ on average) at temperatures <17°C and moved less (21.6 m day⁻¹ on average) at temperatures >18°C (Kato and Hirata, 1990). This may be because the high temperature resulted in an increased energetic cost and oxygen consumption (Dong et al., 2006, 2008), which made the sea cucumber reduce movement to save energy for other metabolic activities. In the wild, one of the primary reasons for movement by sea cucumbers is to search for food (Uthicke, 1999; Mercier et al., 1999). Therefore, the decrease in movement may in part have reduced the feeding activity.

A study conducted on one Indo-Pacific species showed that Holothuria edulis had a higher movement rate at night than during the day (Wheeling et al., 2007). In our study, a nocturnal locomotion behaviour was also evident at all temperatures but 24°C. At 24°C, A. japonicus was relatively inactive and displayed a nearly equal moving proportion during the day and night. This suggests that the high temperature (>24°C) may alter diel locomotion rhythm. Similar results have also been reported in some other species of sea cucumbers. Most adult H. scabra on the surface did not follow their usual burying cycle when the water temperature was increased to more than 30°C (Mercier et al., 2000).

Even in complete darkness, A. japonicus still showed more active feeding and moving at night than during the day in the present study. Dong et al. (2011) reported that the nocturnal activity patterns (i.e. emergence and feeding at night, sheltering during the day) of A. japonicus were observed under continuous darkness as well as continuous light. Sea cucumbers have evolved multiple behavioural strategies to increase their chances of survival in nature. It is believed that nocturnal activity results in the avoidance of diurnal predation (Ebling et al., 1966; Lawrence and Hughes-Games, 1972; Nelson and Vance, 1979). Many predators have been reported to capture A. japonicus at different developmental stages, such as copepods for larval A. japonicus, and carnivorous fish, sea stars, sea urchins and crabs for juveniles and adults less than 10 cm (Yang et al., 2015). In the present experiment, sea cucumbers were kept with a natural photoperiod before the experiment began. Perhaps the length of the experiment was too short or the attempt to change their behaviour was too rapid in relation to their internal clock. Additional studies should, therefore, be conducted to determine whether longer exposure to the different external environment can change the activity patterns of A. japonicus. Thus, it can be concluded from our results and previous studies on this species that diel feeding and locomotion within a short period were not controlled by light but by an endogenous rhythm.

The effects of water temperature on the activities of digestive enzymes of the poikilotherm A. japonicus were mainly reflected in two aspects. First, changes in water temperature caused body temperature changes, which then directly altered the activities of enzymes within the digestive tract. Second, the water temperature affected the feeding behaviour, which then had an indirect impact on digestive activity (Gao et al., 2009). In this study, both lipase and amylase activities increased with the water temperature and reached peaks at 20°C, which is in line with the previous study (Gao et al., 2009). The lowest activities of both digestive enzymes were found at 24°C. The decrease in digestive enzyme activities at 24°C may therefore be related to a reduction in food intake. Maintaining a high level of digestive enzyme activity within the digestive tract would consume energy and, thus, a reduction in digestive enzyme activity during aestivation is conducive to energy conservation in sea cucumber (Gao et al., 2009). The declining trends in digestive enzyme activity observed in A. japonicus during the aestivation phase are consistent with data from lobster and shrimp, either in aestivation or under food deprivation (Comoglio et al., 2004; Johnston et al., 2004).

Digestive activity was positively correlated with both ingestion and assimilation of food (Baars and Oosterhuis, 1984). In our study, the highest activities of lipase and amylase were observed between 22:00 h and 02:00 h (depending on temperature), which was slightly earlier than their corresponding feeding peak (00:00 h to 04:00 h). The results demonstrated that A. japonicus may have the ability to optimize digestive function by secreting digestive enzymes before the feeding peak. This mechanism from ancestral sea cucumbers may allow them to concentrate on feeding in a short time period and thus lower predation risks (Sánchez-Vázquez et al., 1997). In addition, the uptake of food as a physical and biochemical trigger will further increase digestive enzyme secretion into the gut (Zeytin et al., 2016).

In conclusion, the present study highlighted the shift in the diel activity of A. japonicus in response to different water temperatures, which will refine our ecological understanding of A. japonicus. The knowledge of the preferred conditions of water temperature in A. japonicus will assist in the design of suitable holding conditions in the context of captive breeding. In addition, understanding diel activity pattern will be of strategic value to the design of the optimal feeding schedule to maximize feeding efficiency and minimize food waste.