Na⁺/K⁺-ATPase activity during early development and growth of an antarctic sea urchin

Although many species of marine invertebrates in polar regions show lecithotrophic (nonfeeding) development (Thorson, 1950), some of the most common benthic invertebrates in Antarctica have planktotrophic (feeding) larval forms (Pearse et al., 1991). For instance, larvae of the echinoid Sterechinus neumayeri in McMurdo Sound, Antarctica, have planktotrophic larval stages that can be present for several months in a water column that has very low concentrations of phytoplankton (Rivkin, 1991). A common theme in the reproductive biology of species from northern latitudes is that the timing of spawning is related to the bloom of phytoplankton, with the bloom often being the ‘trigger’ that initiates spawning (e.g. sea urchins, Starr et al., 1993). The situation for Antarctic echinoderm species is different, however, because spawning occurs well in advance of the phytoplankton bloom, resulting in feeding larval stages that may have to depend for a long period upon the initial maternal investment of energy reserves in the egg. Shilling and Manahan (1994) calculated the potential life span of larvae on the basis of metabolic rates and the amount of energy reserves in eggs, and estimated that Antarctic sea urchin larvae (Sterechinus neumayeri) could survive for months in the absence of exogenous foods.

What physiological mechanisms might account for the low metabolic rates reported for developing stages of Antarctic echinoderms? Understanding these mechanisms whereby Antarctic larvae achieve such long life spans in the absence of food is important not only for the study of Antarctic larvae but also for other larval forms that have to survive in environments where food supply is low (e.g. larvae in the deep sea). Ion pumps are a major determinant of metabolic rate in cold environments (Hochachka, 1988). The Na⁺ pump (Na⁺/K⁺-ATPase) is a primary active transporter that requires ATP. For developing stages of the temperate sea urchin Strongylocentrotus purpuratus, the physiologically active fraction of total Na⁺/K⁺-ATPase accounts for 40 % of larval metabolic rate (Leong and Manahan, 1997), with a maximum potential of 77 %. Clearly, regulation of the activity of the Na⁺ pump can dramatically alter whole-organism energy metabolism. The dynamics of this process during the development of Antarctic sea urchins is the focus of the current study.

Adult Sterechinus neumayeri (Meisner) were collected from McMurdo Sound, Antarctica, during their natural spawning season of austral spring (October). Spawning was induced by intracoelomic injection of 0.5 mol l⁻¹ KCl. Fertilization was performed by pooling the eggs from several females with the sperm from one male. Fertilized eggs were placed at a concentration of 7 individuals ml⁻¹ in 200 l culturing vessels (Nalgene) containing 0.2 μm (pore size) filtered, ambient sea water taken from McMurdo Sound. The rearing temperature was maintained at −1.5 °C by immersing the 200 l vessels in large aquarium tanks containing flow-through sea water continuously pumped from McMurdo Sound. Cultures were kept suspended using vertically mixing paddles (Plexiglas) driven by electric motors (5–10 revs min⁻¹). The culture sea water was replaced every 4 days, at which time all embryos (or larvae) were siphoned onto 80 μm mesh screens (23 cm in diameter to reduce flow). Under these rearing conditions, development was synchronous, and larvae reached the first feeding stage (pluteus) 22 days after fertilization.

Once the pluteus stage had been reached, the effects of food availability on Na⁺/K⁺-ATPase activity in larvae were studied by adding to the culture water an equal mixture of Rhodomonas sp. and Dunaliella tertiolecta (total algal concentration of 15×10³ cells ml⁻¹). Larvae were fed until 60 days after fertilization, during which time the change in enzyme activity was measured. In a parallel 200 l culture, unfed larvae were maintained in filtered sea water with no algal enhancement. Larvae in both treatments had very low mortality (<0.5 % day⁻¹) throughout the 60 days of the experiments.

Measurement of K⁺-dependent p-nitrophenyl phosphate activity is a highly sensitive method of measuring the activity of Na⁺/K⁺-ATPase (Skou, 1974) and was used to optimize the assay conditions for this enzyme in S. neumayeri. The activity of K⁺-dependent p-nitrophenyl phosphate is defined as the K⁺-dependent hydrolysis of phosphoric anhydrides, such as p-nitrophenyl phosphate (pNPP), by Na⁺/K⁺-ATPase in the absence of Na⁺ (Esmann, 1988). To measure enzyme activity, samples (approximately 0.1 mg of protein) of tissue homogenates were incubated at 15 °C (optimal temperature for the highest signal, see Fig. 3) in a reaction mixture (900 μl) containing a final concentration of 30 mmol l⁻¹ histidine (pH 8.0), 150 mmol l⁻¹ KCl, 20 mmol l⁻¹ MgCl₂, 10 mmol l⁻¹ pNPP and 2.5 mmol l⁻¹ EGTA. A reaction blank was also prepared with the same reagents as the reaction mixture, except that 150 mmol l⁻¹ KCl was substituted with 150 mmol l⁻¹ NaCl (control). The reaction rate was linear with time, and the reaction was allowed to proceed for 60 min before it was terminated by the addition of 100 μl of ice-cold trichloroacetic acid (TCA, final concentration of 5 %). Tris base (2 ml, 500 μmol l⁻¹) was then added to raise the pH of the medium, which aided in the detection of p-nitrophenol. This compound is a product of the hydrolysis of pNPP and is intensely yellow at high pH, allowing for its measurement at 410 nm (Beckman DU-640 spectrophotometer). The difference in activity between the reaction mixture (150 mmol l⁻¹ KCl without NaCl) and the reaction blank (150 mmol l⁻¹ NaCl without KCl) is the K⁺-pNPPase activity.

Homogenization of tissues often results in a mixture of inside-out and right-side-out membrane vesicles (Jørgensen and Skou, 1971). Two commonly used agents were used to check for the possibility of concealed (‘latent’) activity of Na⁺/K⁺-ATPase in homogenates of S. neumayeri. A range of concentrations of alamethicin (Jones et al., 1980; Xie et al., 1989; Ritov et al., 1993) and sodium deoxycholate (Jørgensen and Skou, 1971; Cortas and Edelman, 1988; Hwang and Tsai, 1993) was used on tissue homogenates of mesenchyme blastula (day 10) and prism-stage embryos (day 15). Alamethicin is not soluble in water, so a stock solution (10 mg ml⁻¹) was prepared in 60 % ethanol. The effects of alamethicin on K⁺-pNPPase activity were determined by adding 1–30 μl of the stock solution to tissue homogenates containing approximately 1 mg of protein. Controls were prepared by incubating tissue homogenates in the presence of an equivalent volume of 60 % ethanol with no alamethicin. Tissue homogenates were pre-incubated in the appropriate alamethicin concentrations at 15 °C for 20–30 min before the K⁺-pNPPase assay was conducted. The effects of sodium deoxycholate on enzymatic activity were examined by dissolving sodium deoxycholate in water and adding a range of concentrations to sea urchin homogenates (1 mg protein ml⁻¹). As with alamethicin, tissue homogenates of mesenchyme blastulae and prism-stage embryos were pre-incubated with sodium deoxycholate at 15 °C for 20–30 min before the K⁺-pNPPase assay began.

Total (maximum potential) Na⁺/K⁺-ATPase activity during the development of S. neumayeri was determined by measuring the ouabain-sensitive rate of generation of inorganic phosphate (P_i) from ATP in the presence of tissue homogenates. Total Na⁺/K⁺-ATPase activity was measured as described by Leong and Manahan (1997) with the following modifications: a concentration of 14 mmol l⁻¹ ouabain was required for enzyme inhibition in S. neumayeri (see Fig. 1A), and this was used for all in vitro Na⁺/K⁺-ATPase assays; maximal activity occurred at pH 8.0 (see Fig. 1B), a value that was used in all subsequent assays; alamethicin was used at a final concentration of 200 μg mg⁻¹ protein (see Fig. 2A,B); and all measurements were conducted at 15 °C (see Fig. 3A). The protein content of embryos and larvae was also determined (using a modified Bradford assay; Jaeckle and Manahan, 1989).

The in vivo physiological activities of Na⁺/K⁺-ATPase in living embryos and larvae were determined as the difference in transport rates of K⁺ (using ⁸⁶Rb⁺ as a radiotracer) in the presence and absence of ouabain (Leong and Manahan, 1997). The in vivo transport assays were conducted at −1.00±0.02 °C. The very low solubility of ouabain in sea water at low temperature prevented the use of 14 mmol l⁻¹ ouabain at −1.0 °C (14 mmol l⁻¹ is required for complete inhibition of the Na⁺/K⁺-ATPase activity in S. neumayeri; Fig. 1A). Instead, a final concentration of 6 mmol l⁻¹ ouabain in sea water was used for all the in vivo⁸⁶Rb⁺ transport assays. The residual Na⁺/K⁺-ATPase activity that was not inhibited at 6 mmol l⁻¹ (<20 %) was compensated for when calculating the final in vivo activity of the enzyme (using the inhibition curve presented in Fig. 1A).

Complete inhibition of Na⁺/K⁺-ATPase in Sterechinus neumayeri was obtained at 14 mmol l⁻¹ ouabain (Fig. 1A), and this concentration was selected for subsequent in vitro enzyme assays. The high sensitivity of the K⁺-dependent p-nitrophenyl phosphate (K⁺-pNPPase) reaction was used to determine the effect of pH and demasking agents. The optimal pH for enzyme activity was broad, ranging between pH 7.12 and pH 8.12 at 15 °C, with further increases in pH resulting in a rapid decrease in activity (negligible activity above pH 9.5: Fig. 1B). Alamethicin resulted in a concentration-dependent increase in measurable enzyme activity in homogenates of both blastula and prism embryos (Fig. 2A,B). At 200 μg alamethicin mg⁻¹ protein, the K⁺-pNPPase activity doubled (blastula 1.8-fold; prism 2.0-fold). In contrast, sodium deoxycholate was not as effective in demasking the enzyme. A small increase in activity was observed, but increasing concentrations of deoxycholate severely inhibited the enzyme activity in early and late-stage embryos (blastula, Fig. 2C; prism, Fig. 2D). Alamethicin at 200 μg mg⁻¹ protein was used as the permeabilizing agent for all subsequent assays because sodium deoxycholate was ineffective.

The effect of increasing temperature on the activity of the Na⁺/K⁺-ATPase in S. neumayeri was also determined using the K⁺-pNPPase assay. A Q₁₀ of 2.9 was measured over the temperature range tested of −1 °C to 15 °C (Fig. 3). Because there was no temperature inhibition of enzyme activity at 15 °C (the highest assay temperature tested), all enzyme assays were performed at this temperature to yield higher signals.

Both total (Fig. 4A) and in vivo (Fig. 4B) Na⁺/K⁺-ATPase activities increased during development. The inset in Fig. 4B shows that the increase of in vivo Na⁺/K⁺-ATPase activity through early development seen for one culture in Fig. 4B was also observed for two other independent cultures (spawned from different adults of S. neumayeri). The rate of increase of total enzyme activity was 3.9 times faster for fed larvae than for embryos and unfed larvae (Fig. 4A: ANOVA comparison of slopes, variance ratio (VR)=9.94, F_0.01(1,35)=8.98, P ⩽ 0.01). The raw data for total Na⁺/K⁺-ATPase activity were measured at 15 °C, but were recalculated with the appropriate value for Q₁₀ (2.9; Fig. 3A,B) to express the corresponding activity at −1.0 °C, as given in Fig. 4A. This allowed a direct comparison of the total enzyme activity with the percentage of the enzyme that was utilized (in vivo rates) at the physiological temperature of −1.0 °C (see Fig. 6A).

Fed larvae had a protein deposition rate of 8.1 ng larva⁻¹ day⁻¹ (Fig. 5A). By the end of the 2 month period of starvation, unfed larvae had lost 35 % of their initial (as measured on day 6) protein content, with a slow linear rate of protein loss of −1.26 ng protein larva⁻¹ day⁻¹ (Fig. 5A; by analysis of variance, ANOVA, the slope is significantly different from zero, VR=69.57, F_0.001(1,90)=13.0, P ⩽ 0.001). Ontogenetic changes in enzyme activity were scaled to protein content in fed and unfed larvae. The protein-specific activity of total Na⁺/K⁺-ATPase increased with development time at a rate that was not significantly different for fed and unfed larvae (Fig. 5B). Prior to pooling the regression data for fed and unfed larvae as shown in Fig. 5B, the individual regressions for fed and unfed larvae were first compared (with or without data for embryos) and found not to be significantly different (ANOVA comparison of slopes for fed and unfed: VR=3.69, F_0.05(1,40)=5.42; P ⩽ ù0.05). Total Na+/K+-ATPase activity per larva increased during feeding as a function of the rate of protein growth (Fig. 5C). Enzyme activity did not change as a function of protein content for unfed larvae (Fig. 5C, slope not significant by ANOVA, see legend).

Although the amount of total enzyme per individual differed greatly between fed and unfed larvae (Fig. 4A), the proportion of the total potential Na⁺/K⁺-ATPase that was physiologically active in vivo did not differ during development as a function of feeding (Fig. 6A). Both fed and unfed larvae showed an increase during development in the percentage of total potential Na⁺/K⁺-ATPase that was physiologically active, from 17 % of the enzyme being utilized by early (10-day-old) gastrulae, increasing to 77 % by early-stage six-arm pluteus (48-day-old) larvae.

We calculated the percentage of metabolic rate during development of this Antarctic sea urchin that is accounted for by the energy required to sustain the Na⁺ pump. Rates of oxygen consumption during development were taken from Marsh et al. (1999) for measurements made on the same cultures used here to quantify enzyme activity. The rate of utilization of ATP by the fraction of the Na⁺ pump that was physiologically active (⁸⁶Rb⁺-determined in vivo rate) can be compared directly with the rate of oxygen consumption, when the latter is expressed as ATP equivalents (see conversions in Leong and Manahan, 1997). This comparison of respiration rate and Na⁺ pump activity is shown in Fig. 6B; the metabolic importance of in vivo Na⁺/K⁺-ATPase activity increased linearly with development, accounting for 12 % of metabolic rate at day 10 (gastrula) and 84 % by day 48 (larva).

Alamethicin and sodium deoxycholate are commonly used to demask latent Na⁺/K⁺-ATPase activity in tissue homogenates. The detergent sodium deoxycholate disrupts the lipid vesicles that are formed from membrane fragments generated during homogenization and can expose the ‘concealed’ Na⁺/K⁺-ATPase activity (Jørgensen, 1988). However, like most detergents, the critical micelle concentration for sodium deoxycholate has a narrow range, and concentrations below or above this range result in lower Na⁺/K⁺-ATPase activity. For Sterechinus neumayeri, the activity of Na⁺/K⁺-ATPase was highest at approximately 1 mg ml⁻¹ sodium deoxycholate (Fig. 2C,D), but even at this concentration only a small increase in total Na⁺/K⁺-ATPase activity was measurable. Unlike deoxycholate, alamethicin permeabilizes membrane vesicles through the formation of transmembrane channels (Cafiso, 1994, and references therein). Alamethicin can also be used to permeabilize whole, intact cells for measurement of in situ activity of various transmembrane enzymes (Xie et al., 1989; Ritov et al., 1993; Kultz and Somero, 1995). In our study, the addition of alamethicin to homogenates resulted in a twofold increase in measurable enzyme activity for mesenchyme blastulae (Fig. 2A) and prism-stage (Fig. 2B) embryos. This doubling of activity with a permeabilizing agent is not unexpected, given that in a crude homogenate there exists a random mixture of inside-out and right-side-out membrane vesicles. Our results show that alamethicin should be used in preference to sodium deoxycholate for permeabilizing membrane vesicles in tissue homogenates of S. neumayeri. Kultz and Somero (1995) also found that sodium deoxycholate was not as effective as alamethicin for studies of Na⁺/K⁺-ATPase in cells from fish gill.

Both total Na⁺/K⁺-ATPase activity and the physiologically active fraction (in vivo activity) increased in fed and unfed larvae of S. neumayeri (Fig. 4A,B). The increase in Na⁺/K⁺-ATPase activity on a per-individual basis for unfed larvae (Fig. 4B) occurred despite the decrease in total body protein content during this period (Fig. 5A). Unfed larvae had a mean total Na⁺/K⁺-ATPase activity of 80.7±4.23 pmol P_i individual⁻¹ h⁻¹ (mean ± S.E.M., N=3–5), a value that was independent of the total protein content of larvae (Fig. 5C). This probably represents the lower threshold of Na⁺/K⁺-ATPase activity that has to be maintained by larvae of S. neumayeri, even under starvation conditions. Although the fed larvae showed a significant increase in total Na⁺/K⁺-ATPase activity, the percentage of the enzyme that was physiologically active (Fig. 6A) and the metabolic importance of the enzyme’s active fraction were independent of food supply (Fig. 6B). The in vivo Na⁺/K⁺-ATPase activity for both fed and unfed larvae of S. neumayeri increased with time, with this single enzyme’s requirement for ATP accounting for 12–84 % of metabolic requirements depending on the stage of development (Fig. 6B). Similar studies with temperate species of sea urchins (Strongylocentrotus purpuratus and Lytechinus pictus) found that in vivo Na⁺/K⁺-ATPase activities accounted for up to 40 % of metabolic rates (Leong and Manahan, 1997).

Regarding the role of Na⁺ pumps and hypometabolic regulation in cold environments, the available data on Na⁺/K⁺-ATPase activity in sea urchin development are consistent with the suggestion (Hochachka, 1988) that Antarctic species have lower amounts of this enzyme. When compared at the same assay temperature (15 °C), the protein-specific activities of total Na⁺/K⁺-ATPase in the prism and pluteus stages of two species of temperate sea urchins (S. purpuratus and L. pictus) were up to 2.6 times higher (P<0.001) than activities in these stages of S. neumayeri (Table 1). Earlier stages of development (morula, blastula, gastrula) did not have a significantly different enzyme activity in these three different species. The differences in specific activity of the enzyme only appeared in late-stage embryos (prism) and continued into the larval stage of the Antarctic species. These differences in activity could be caused by a variety of mechanisms in the Antarctic species, such as lower numbers of Na⁺ pumps or differences in turnover rates or the molecular activities of each individual pump (Else et al., 1996). Given that approximately half of the total pump activity is utilized in vivo by both the temperate (Leong and Manahan, 1997) and the Antarctic (Fig. 6A) species, the lower protein-specific activities of Na⁺/K⁺-ATPase indicate a reduced energy requirement for ion regulation during development of S. neumayeri. Because the activity of the Na⁺ pump is such a major component of metabolic rate, a reduction in its activity would result in lower metabolic rates in the Antarctic species. The finding of a decrease in Na⁺/K⁺-ATPase activity in the Antarctic species (compared with temperate sea urchins), but an increase in the percentage of metabolic rate (S. neumayeri 84 %; S. purpuratus 40 %) accounted for by the Na⁺ pump, has also been observed during anoxia (Buck and Hochachka, 1993). Under normoxic conditions, the activity of Na⁺/K⁺-ATPase in turtle hepatocytes accounted for 28 % of their cellular metabolic rate. In response to anoxia, the activity of the Na⁺ pump decreased by 75 % and yet this lower activity accounted for 74 % of cellular metabolic rate.

Low metabolic rates in larval forms of Antarctic echinoderms permit extended life spans in the absence of exogenous food by using the initial maternal endowment of energy reserves at a low rate (Shilling and Manahan, 1994). Also, a reduction in metabolic rate has been suggested as an over-wintering mechanism for the survival of Antarctic krill (Quetin and Ross, 1991). For Antarctic invertebrates in general, little is known (Clarke, 1991) about the cellular mechanisms of metabolism compared with what is known about fish (Torres and Somero, 1988; Crockett and Sidell, 1990). Knowing the physiological mechanisms whereby Antarctic larvae can survive months of starvation is important to understanding their ecology. In addition, this increased understanding might be applicable to larval forms from other environments where food availability and temperature are low (e.g. deep-sea larvae).

We thank Dr Adam Marsh and Tracy Hamilton for their help with culturing, Dr Alicia McDonough for her advice on the development of our protocol for the pNPPase assay and Antarctic Support Associates Inc. for providing logistic support for our research at McMurdo Station, Antarctica. This work was supported by a grant from the US National Science Foundation (OPP-9420803).