Metabolic, respiratory and cardiovascular responses to acute and chronic hypoxic exposure in tadpole shrimp Triops longicaudatus

Hypoxic exposure can elicit a range of physiological and metabolic responses in aquatic crustaceans with `oxygen conformation' and `oxygen regulation' at the ends of a response continuum(Hochachka, 1988; McGaw et al., 1994; Reiber, 1995; Reiber and McMahon, 1998; Wilkens, 1999). Oxygen conformers are able to reduce metabolic demand so that O₂consumption (V̇_O2)is coupled to environmental O₂ partial pressure(P_O2)(Loudon, 1988; Hochachka et al., 1999; Boutilier, 2001). Oxygen regulators, by contrast, maintain V̇_O2 independent of environmental P_O2 down to a point where the O₂ required for aerobic processes becomes limiting(P_CRIT) (Herreid,1980; Hochachka,1988). Below P_CRIT, either tissue oxygen demand decreases or anaerobic processes become increasingly important in meeting energy requirements (Hochachka,1988). Above P_CRIT, O₂ uptake and delivery may be enhanced through a variety of compensatory mechanisms including adjusting ventilatory parameters (rates and volumes), increasing internal convective processes such as heart rate and stroke volume and changing perfusion patterns (McMahon,1988; Hervant et al.,1995; Willmer et al.,2000; Boutilier,2001; Hochachka and Lutz,2001; Hopkins and Powell,2001; Hoppeler and Vogt,2001). Additionally, some animals can modulate the O₂-binding affinity of their respiratory proteins, which enhances O₂ uptake at the respiratory surface and increases total O₂ capacitance (Wells and Wells, 1984; Kobayashi et al.,1988; Hochachka et al.,1999; Willmer et al.,2000).

Hypoxic exposure during development (embryonic or larval periods) may elicit a similar suite of compensatory responses or may result in long-term or permanent changes to the morphology and/or physiology of the animal and thus impact the adult hypoxic response (Bamber and Depledge, 1997; Bradley et al., 1999; Willmer et al.,2000; Gozal and Gozal,2001; Peyronnet et al.,2002). The physiological consequences of hypoxia-induced developmental plasticity can be a reduced metabolic rate(Hochachka et al., 1999; Boutilier, 2001), greater respiratory gas exchange surface area(Loudon, 1988; Loudon, 1989), increased respiratory capacities and/or enhanced convection and O₂ delivery mechanisms (Banchero, 1987; McMahon, 1988; Graham, 1990; Szewczak and Jackson, 1992; Childress and Seibel, 1998; Pelster, 1999; McMahon, 2001). Transcriptional regulation accounts for many of these long-term physiological and morphological changes associated with developmental exposure to hypoxia. A well-documented example of this type of developmental hypoxic response in crustaceans is the increase of existing O₂-binding proteins as well as the production of new higher-affinity proteins that result in long-term increases in O₂-uptake and capacitance(Wells and Wells, 1984; Snyder, 1987; Kobayashi et al., 1988; Hervant et al., 1995; Astall et al., 1997; Hou and Huang, 1999; Wiggins and Frappell, 2000; Barros et al., 2001).

The degree to which developmental P_O2influences adult metabolic, respiratory and cardiovascular hypoxic responses was investigated using tadpole shrimp Triops longicaudatus. Several features of tadpole shrimp make them well suited to study the effects of acute and developmental hypoxic exposures. They are often faced with environmental P_O2s below their P_CRIT, yet the mechanisms by which they tolerate such conditions are poorly understood(Horne and Beyenbach, 1971; Hillyard and Vinegar, 1972; Eriksen and Brown, 1980; Scholnick, 1995; Scholnick and Snyder, 1996). Tadpole shrimp have high metabolic rates and respiratory structures(epipodites) that are thought to be inadequate to maintain O₂uptake in their euryoxic habitats (Fryer,1988; Horne and Beyenbach,1971; Hillyard and Vinegar,1972; Scott and Grigarick,1978; Eriksen and Brown,1980; Scholnick,1995; Scholnick and Snyder,1996). A tubal, myogenic heart devoid of vasculature produces the only internal convective currents in tadpole shrimp(Yamagishi et al., 1997; Yamagishi et al., 2000). Large, extracellular hemoglobin (29 subunits) is produced in response to hypoxic exposure, but the mechanisms of this hypoxic induction are not known(Horne and Beyenbach, 1971; Scholnick and Snyder, 1996). Finally, tadpole shrimp have a comparatively rapid generation time and amenability to laboratory culture, which make them tractable organisms for developmental investigations (Fryer,1988; Horne and Beyenbach,1971; Scott and Grigarick,1978).

Tadpole shrimp were reared under normoxic (19–21 kPa O₂),moderate (10–13 kPa O₂) or severe (1–3 kPa O₂) hypoxic conditions to determine whether differences in developmental P_O2 were sufficient to change adult metabolic, respiratory or cardiovascular system physiology and hypoxic sensitivity. Compensatory respiratory and cardiovascular processes that enhance O₂ uptake, internal convection and perfusion should increase in response to hypoxic exposure. We hypothesized that tadpole shrimp reared under hypoxic conditions would have decreased metabolic sensitivity to P_O2 change, increased physiological responses to hypoxic exposure and increased hemoglobin concentration and O₂-binding affinity relative to those reared under normoxic conditions (Wells and Wells,1984; Snyder,1987; Kobayashi et al.,1988; Hervant et al.,1995; Astall et al.,1997; Hochachka et al.,1999; Hou and Huang,1999; Barros et al.,2001; Boutilier,2001).

Sediments containing tadpole shrimp (Triops longicaudatus LeConte)cysts were collected from an ephemeral pool in Brownstone Basin (12.7 km west of Las Vegas, NV, USA; 1425 m elevation; 36.2500° N, –115.3750°W). Water P_O2 and temperature in Brownstone Basin pool ranged from 2 to 32 kPa O₂ and 17.6 to 31.7°C,respectively (Harper, 2003). Three 150-liter insulated aquaria (normoxic=19–21 kPa O₂,moderate hypoxic=10–13 kPa O₂ and severe hypoxic=1–3 kPa O₂) were set up in the laboratory using deionized water and sediments taken from the pool. A model GF-3/MP gas flowmeter (Cameron Instruments Company, Ontario, Canada) was used to regulate a mixture of N₂ and room air. Oxygen partial pressure in each aquaria was monitored for a 48-h period each week using data-logging dissolved O₂ meters (Model 810 Orion Dissolved Oxygen Meter; Orion Research,Boston, MA, USA).

Each aquarium was equipped with ultraviolet (3% UVB and 7% UVA) enhanced lights (Super UV Reptile Daylight Lamp; 20 W; Energy Savers Unlimited Inc.,Chicago, IL, USA) that established a 13 h:11 h L:D cycle. A timed circulating water bath (Model VT513; Radiometer, Copenhagen, Denmark) and heat exchange coil were used to cycle temperature with the lights. The aquaria started to warm two hours after the lights were turned on and continued for five hours each day to establish a 23–28°C temperature cycle. Dry sediments(100 g) containing tadpole shrimp cysts were added to each aquarium weekly. Animals ate algae, detritus and small invertebrates that hatched from the sediment. Tadpole shrimp were identified as Triops longicaudatus(Sassaman, 1991). Aquaria were drained and refilled monthly.

Standard metabolic rate was measured when animals were active because they were infrequently quiescent. To assess the confounding effect of P_O2 on activity, individual animals(N=10) were placed in a marked cylindrical 10-ml flow-through chamber and videotaped during progressive hypoxic exposure. Animals were acclimated(30 min) to the chamber under normoxic conditions. A gas flowmeter (Model GF-3/MP) was used to regulate the mixture of N₂ and air. Chamber P_O2 was decreased from 20 to 2 kPa O₂ at a rate of 5 kPa O₂ h^–1. The number of times the animal crossed defined marks on the chamber was averaged over one-minute intervals to produce an index of activity in response to P_O2.

Animals reared under normoxic (N=10) and severe hypoxic(N=10) conditions were used to assess anaerobic lactate metabolism in tadpole shrimp. Lactate concentration was measured for five animals pre-treatment and five animals after exposure to severe hypoxic conditions (2 kPa O₂) for 12 h at 23°C. Lactate concentrations in the experimental chamber water were also determined. Hypoxic conditions were maintained using a gas flowmeter (Model GF-3/MP) to control the mixture of N₂ and room air. Hemolymph was collected in glass capillary tubes by dorsal heart puncture. Lactate concentrations were determined for 10 μl water samples and 10 μl hemolymph samples mixed with 1.0 ml lactate reagent solution (#735-10; Trinity Biotech, St Louis, MO, USA). Hemolymph lactate concentrations were measured enzymatically (Sigma Diagnostics, St Louis, MO,USA; Sigma Lactate Kit #735) at 540 nm.

Ventilatory rate and amplitude were measured in response to hypoxic exposure in order to assess the effects of developmental P_O2 on adult ventilatory hypoxic response. Adult tadpole shrimp from each rearing group (N=13 per rearing group)were held in a 30-ml flow-through chamber (temperature, 25°C). Tadpole shrimp were secured in the chamber with an applicator stick and cyanoacrylate glue on the lateral carapace. They were inverted in the chamber to allow the ventral surface to be viewed. Movements of the respiratory appendages were videotaped (60 Hz sampling speed) under a dissecting microscope (Leica Stereozoom 6 Photo) using a video camera (Oscar Color Camera Vidcam), super VHS video recording system (Panasonic PV-54566) and Horita time code generator(VG 50; Horita Co., Inc., Mission Viejo, CA, USA). Tadpole shrimp were acclimated for 30 min and then exposed to four environmental P_O2s (20, 13.3, 10 and 1 kPa O₂) in random order. Thirty minutes was allowed for acclimation once a new P_O2 was achieved. Animals were not returned to normoxia between exposures. Time-encoded video was analyzed frame-by-frame on an editing tape player (Panasonic AG-DS550) to determine ventilation rate and amplitude. Ventilation rate (frequency) was measured as number of appendage beats per minute. The amplitude of appendage beats was determined as the mean distance that the 4th and 5th appendages separate during five subsequent ventilatory strokes (Harper,2003).

Heart rate, stroke volume and cardiac output were measured in response to hypoxic exposure in order to assess the effects of developmental P_O2 on adult cardiac hypoxic responses. Adult tadpole shrimp from each rearing group (N=13 per rearing group) were secured in a 30-ml flow-through experimental chamber as previously described. Animals were acclimated for 30 min and then exposed to five environmental P_O2s (26.7, 20, 13.3, 10 and 4 kPa O₂) in random order. Thirty minutes was allowed for acclimation at each P_O2. Cardiac contractions were videotaped as previously described. Heart rate (fh; beats min^–1) was measured when the time-encoded video was advanced frame-by-frame on an editing tape player (Panasonic AG-DS550, Cypress, CA). The tadpole shrimp heart was modeled as a cylinder with a volume ofπ r²h, where r is half the width of the heart and h is length. Images of the heart were collected during maximal [end diastolic volume (EDV)] and minimal [end systolic volume (ESV)]distention. Those images were dimensionally analyzed using Scion Imaging software (National Institutes of Health, Bethesda, MD, USA). Stroke volume(Vs; μl beat^–1) was calculated as the difference in heart volume between EDV and ESV. Cardiac output(Q̇; μl min^–1) was calculated as the product of fh and Vs.

Hemoglobin is the major protein in tadpole shrimp hemolymph(Horne and Beyenbach, 1971). Protein concentrations of animals from each rearing group (N=10 per rearing group) were determined to assess the influence of developmental P_O2 on hemoglobin production. Protein concentrations were determined for tadpole shrimp reared under normoxic conditions and exposed to severe hypoxic conditions for 5, 7 and 10 days(N=7 per day). Likewise, protein concentrations were determined for tadpole shrimp reared under severe hypoxic conditions and exposed to normoxic conditions for 5, 7 and 10 days (N=7 per day). Protein concentrations were determined using a Micro BCA Protein Assay Reagent Kit (#23235; Pierce,Rockford, IL, USA). Protein standards (2.0 mg ml^–1 BSA in a solution of 0.9% saline and 0.05% sodium azide) were diluted with tadpole shrimp saline (5.84 mg NaCl, 7.45 mg KCl, 11.09 mg CaCl₂, 9.52 mg MgCl₂, 3.65 mg HCl and 1 ml H₂O)(Yamagishi et al., 2000) to form solutions with final BSA concentrations of 200, 40, 20, 10, 5, 2.5, 1 and 0.5 μg ml^–1. Working reagent was prepared by mixing 12.5 ml Micro BCA Reagent MA (sodium carbonate, sodium bicarbonate and sodium tartrate in 0.2 mol l^–1 NaOH) and 12 ml Micro BCA Reagent MB[bicinchoninic acid (4.0%) in water] with 0.5 ml Micro BCA Reagent MC (4.0%cupric sulfate, pentahydrate in water). One milliliter of each standard was added to appropriately labeled test tubes. A water blank and tadpole shrimp saline were used as controls. In each test tube, 1.0 ml working reagent was added and mixed. The tubes were covered with Parafilm™ and placed in a 60°C water bath for 60 min and then cooled to room temperature (23°C). Absorbance was measured at 562 nm, with corrections made for reference. A standard curve was produced to obtain hemoglobin concentrations of hemolymph samples.

Hemoglobin O₂-binding affinities were measured for animals from each rearing group to determine differences dependent on developmental P_O2. Hemolymph (60 μl) was collected in glass capillary tubes from large tadpole shrimp (N=7 per rearing group) by dorsal puncture of the heart. Hemolymph was added into a small flow-through tonometer that opened into a narrow chamber (1 mm inner diameter)inside a cuvette (1 cm×1 cm×5 cm). The tonometer was placed on its side when hemolymph was added and during each equilibration step. This allowed the hemolymph to flow into the bulbous region of the tonometer. A stirring flea powered by a magnetic stirrer was placed into the tonometer to ensure thorough mixing of the hemolymph with inflowing gas mixtures. During spectrophotometric measurements, the tonometer was held upright so that the hemolymph flowed into the narrow chamber. The P_O2 of humidified inflowing gas (1000 sccm) was adjusted using a gas flowmeter (Model GF-3/MP; Cameron Instruments, Guelph,ON, Canada). Hemolymph was equilibrated for 20 min with normoxic air (20 kPa)and analyzed spectrophotometrically using a Turner spectrophotometer (Model 340; Mountain View, CA, USA) at a wavelength of 570 nm. This is the wavelength of maximal absorbance for oxy- and deoxy-hemoglobin for Triops(Horne and Beyenbach, 1974). Water was used as a reference. The absorbance of hemoglobin was determined after equilibration with air of 30, 4.0, 2.7, 1.3, 1.1, 0.8, 0.5 and 0 kPa O₂. Standard curves were constructed from the absorbance of hemoglobin at 0% and 100% saturation. Percent saturation for hemoglobin at each P_O2 was calculated using standard curves. Curve fitting of O₂ binding was calculated using SigmaStat 2.03(SPSS Inc., Chicago, IL, USA). The P₅₀ for each rearing group was determined as the P_O2 at which 50%saturation occurred (Bruno et al.,2001). Cooperativity (n_H) was calculated as the maximal slope of log[saturation/(1–saturation)] against log[P_O2](Bruno et al., 2001).

Oxygen-dependent changes in hemolymph pH were used to determine the significance of a Bohr shift in altering hemoglobin O₂-binding affinity. Hemolymph pH was measured using a PHR-146 Micro Combination pH Electrode (Lazar Research Laboratories, Inc., Los Angeles, CA, USA) inserted into the base of a flow-through chamber (20 μl). The P_O2 of inflowing humidified gas (1000 sccm) was adjusted using a gas flowmeter (Model GF-3/MP) to control a mixture of N₂ and room air. Hemolymph pH was determined after 10 min equilibration at 30, 4.0, 2.7, 1.3, 1.1, 0.8, 0.5 and 0 kPa O₂.

Oxygen-delivery potential was calculated as the product of O₂content and cardiac output and reported in μl O₂min^–1 (Ronco et al.,1991). Calculations for the determination of O₂ content and cardiac output were described above.

All statistical analyses were run with SigmaStat 2.03 unless otherwise specified. Results are presented as means ± s.e.m. with statistical significance accepted at the level of P<0.05. Multiple pairwise comparisons were made using Bonferroni t-tests when rearing effects were significant, unless otherwise specified.

The strength of the relationship between activity level and P_O2 was measured using Pearson Product Moment Correlation. Comparisons of metabolic response to graded hypoxia were made among rearing groups using one-way repeated-measures analysis of variance(ANOVA). Comparison of lactate concentrations was made between normoxic and severe hypoxic rearing groups using a Student's t-test.

Comparisons of ventilatory amplitude and frequency among rearing groups were made using Kruskal–Wallis one-way ANOVA on ranks because data had unequal variance. Multiple pairwise comparisons were made using a Tukey test.

Comparisons of heart rate, stroke volume and cardiac output to graded hypoxia were made among rearing groups using one-way repeated-measures ANOVA.

Comparisons of hemoglobin concentration were made among rearing groups using one-way ANOVA. The hemoglobin concentration of animals that had been switched from normoxic to severe hypoxic, and from severe hypoxic to normoxic conditions (after 5, 7 and 10 days), was analyzed using one-way repeated-measures ANOVA. Comparison of the O₂-binding affinity of hemoglobin from each rearing group was made using Friedman repeated-measures ANOVA on ranks because of unequal variance. Hemoglobin P₅₀was compared among rearing groups using Kruskal–Wallis one-way ANOVA on ranks because data were not normally distributed. Multiple pairwise comparisons were made using a Tukey test. Hemoglobin n_Hwas compared among rearing groups using analysis of covariance (StatView,5.0.1; StatView Software, Cary, NC, USA). Multiple pairwise comparisons were performed using a Scheffe test and StatView. Comparisons of hemolymph pH with P_O2 were made among rearing groups using one-way repeated-measures ANOVA.

Comparisons of the O₂-carrying capacity of hemolymph were made among rearing groups using Kruskal–Wallis one-way ANOVA on ranks because data did not have equal variance. Multiple pairwise comparisons were made using a Tukey test. Comparisons of the O₂-delivery potential of hemolymph were made among rearing groups using Kruskal–Wallis one-way ANOVA on ranks because data had unequal variance. Multiple pairwise comparisons were made using a Tukey test.

Comparisons of the O₂ consumption/oxygen transport coupling were made among rearing groups using one-way repeated-measures ANOVA.

Animal activity was not correlated with P_O2and therefore did not confound V̇_O2 measurements. Developmental P_O2 had no effect on metabolic rate or metabolic response to hypoxic exposure(Fig. 1). All groups gradually decreased V̇_O2down to a P_CRIT of 2 kPa O₂ (691±20μl O₂ g^–1 h^–1 at normoxia to 274±9 μl O₂ g^–1 h^–1 at 2 kPa O₂ in normoxic animals; 714±14 μl O₂g^–1 h^–1 at normoxia to 277±14 μl O₂ g^–1 h^–1 at 2 kPa O₂in moderate hypoxic animals; and 728±3 6 μl O₂g^–1 h^–1 at normoxia to 253±12 μl O₂ g^–1 h^–1 at 2 kPa O₂in severe hypoxic animals). Baseline lactate levels for animals reared under normoxic (2.971±0.29 mmol l^–1) and severe hypoxic(3.142±0.33 mmol l^–1) conditions were not significantly different after 12 h of severe hypoxic exposure(3.810±0.45 mmol l^–1 and 3.304±0.59 mmol l^–1, respectively). No lactate was observed in the any of the water samples.

Ventilation rates were highly variable in all rearing groups. Tadpole shrimp ventilation rates were not different among rearing groups and did not differ within groups in response to hypoxic exposure (13.3, 10 and 1 kPa O₂) (Fig. 2A). Ventilatory amplitude did not differ within groups in response to hypoxic exposure (Fig. 2B). However,ventilatory amplitudes at 1 kPa O₂ were significantly different among the groups. Ventilatory amplitude was less in animals reared under severe hypoxic conditions when compared with animals reared under normoxic conditions but only at 1 kPa O₂.

Heart rate response to hypoxic exposure was significantly different among the rearing groups. At low P_O2, fh decreased in tadpole shrimp reared under normoxic conditions from 281±6 beats min^–1 at 20 kPa O₂ to 226±5 beats min^–1 at 2 kPa O₂ (Fig. 3A). Animals reared under moderate hypoxia also showed a significant decrease in fh, from 262±11 beats min^–1 at 20 kPa O₂ to 223±8 beats min^–1 at 2 kPa O₂. However, animals reared under severe hypoxic conditions did not exhibit a change in fh with decreased P_O2 (275±9 beats min^–1at 20 kPa O₂ to 238±10 beats min^–1 at 2 kPa O₂). Tadpole shrimp reared under normoxic (0.23±0.01 μl beat^–1) and moderate hypoxic conditions (0.28±0.01μl beat^–1) had significant differences in Vs at 10 kPa O₂ (Fig. 3B). Those reared under normoxic conditions decreased Vs, whereas those reared under moderate hypoxic conditions increased Vs in response to severe hypoxic exposure. Yet neither hypoxic response was statistically significant due to the high variability of Vs; mean standard deviation was greater than the hypoxic response. Cardiac output was maintained in animals reared under severe hypoxic conditions down to 2 kPa O₂(Fig. 3C). However, animals reared under moderate hypoxic conditions decreased Q̇ from 64±6 μl min^–1 at 20 kPa O₂ to 53±2 μl min^–1 at 2 kPa O₂, and animals reared under normoxic conditions decreased Q̇ from 67±4 μl min^–1 at 20 kPa O₂ to 59±3 μl min^–1 at 2 kPa O₂.

Hemoglobin concentrations measured as protein concentration were dependent on rearing P_O2 and were significantly altered by chronic hypoxic exposure. Animals reared under normoxic (9.6±0.7μg μl^–1) and moderate hypoxic (2.8±1.4 μgμl^–1) conditions had significantly less protein(hemoglobin) than those reared under severe hypoxic (19.8±0.9 μgμl^–1) conditions (Fig. 4). Concentrations in tadpole shrimp reared under normoxic conditions and transferred to severe hypoxic conditions increased significantly after 7 days (Fig. 5). After 10 days of severe hypoxic exposure, concentrations were not significantly different from animals that had been reared under severe hypoxic conditions. Adult tadpole shrimp reared under severe hypoxic conditions and switched to normoxic conditions did not decrease the amount of hemoglobin in their hemolymph for the 10 days investigated.

The O₂-binding affinity of hemoglobin was dependent on developmental P_O2. Hemoglobin O₂-binding affinity in animals reared under normoxic conditions was less than those reared under severe hypoxic conditions but not different from animals reared under moderate hypoxic conditions(Fig. 6). Hemoglobin cooperativity was significantly different among rearing groups(Fig. 7). Animals reared under normoxic conditions exhibited more cooperativity among hemoglobin subunits than those reared under moderate and severe hypoxic conditions. Oxygen-binding affinities and cooperativity values for hemoglobin from each rearing group are given in Table 1.

Tadpole shrimp O₂-carrying capacity was dependent on developmental P_O2(Table 1). Animals reared under severe hypoxic conditions had 5.2× the O₂-carrying capacity of those reared under normoxic conditions. Maximal O₂-delivery potential for tadpole shrimp was dependent on developmental P_O2. Oxygen-delivery potential was significantly lower in animals reared under normoxic conditions (274 μl O₂ min^–1) relative to animals reared under severe hypoxia (1336 μl O₂ min^–1)(Table 1).

The ratio of V̇_O2/Q̇_O2was dependent on developmental P_O2. Oxygen consumption/transport ratio was significantly higher (consistently above 1) in animals reared under normoxic conditions relative to those reared under moderate or severe hypoxic conditions (Fig. 8). Ambient P_O2 significantly affected V̇_O2/Q̇_O2in animals reared under normoxic and moderate hypoxic conditions. Animals reared under moderate hypoxic conditions had a near coupling (1.18) of O₂ demand to cardiovascular supply at 20 kPa O₂, which decreased significantly with severe hypoxic exposure (2 kPa O₂). Animals reared under severe hypoxic conditions maintained V̇_O2/Q̇_O2below 1 at each P_O2 investigated.

The ability to regulate V̇_O2 may be related to the frequency and duration of hypoxic exposure that an animal typically encounters in its habitat (Chen et al., 2001). Tadpole shrimp inhabit euryoxic environments that can fluctuate daily by more than 26.7 kPa O₂ or remain almost anoxic (<1.3 kPa O₂) for months(Horne and Beyenbach, 1971; Scholnick, 1995). In addition,they vertically migrate from severely hypoxic (2 kPa O₂) sediments to hyperoxic (32 kPa O₂) surface water in a relatively short period of time (Scholnick and Snyder,1996). Oxygen conformation to environmental P_O2 should be energetically favorable for organisms frequently exposed to hypoxic conditions on many temporal and spatial scales (Chen et al.,2001). Tadpole shrimp would be considered oxygen conformers as they did not tightly regulate V̇_O2 when P_O2 was decreased(Fig. 1). Instead, tadpole shrimp reduced O₂ demand to match O₂ supply.

Sensitivity to hypoxic exposure may be altered in animals exposed to hypoxic conditions throughout development due to ontogenetic changes in physiological capabilities and metabolic demand(Schulte, 2001). Adult animals that develop under hypoxic conditions often have decreased metabolic rates and decreased hypoxic sensitivity relative to animals that develop under normoxic conditions (Pichavant et al.,2001; Sokolova and Portner,2001; Cech and Crocker,2002; Chapman et al.,2000; Hammond et al.,2002). Yet developmental P_O2 did not affect adult tadpole shrimp metabolic rate or metabolic response to acute hypoxic exposure. Tadpole shrimp that develop under hypoxic conditions must balance immediate O₂ demand with long-term requirements for completing their life cycle within 30–40 days(Horne and Beyenbach, 1971). Therefore, transient decreases in V̇_O2 in response to ambient changes in P_O2 may be more beneficial than permanent decreases in V̇_O2 that could ultimately limit growth and development.

Anaerobic metabolism can be used to supplement aerobic metabolism and decrease sensitivity to hypoxic exposure in many aquatic organisms(Truchot, 1980; Childress and Seibel, 1998). Tadpole shrimp do not appear to utilize anaerobic metabolic pathways that end in lactic acid production; hemolymph lactate levels did not increase after 12 h of severe hypoxic exposure (1.3 kPa O₂). However, tadpole shrimp may employ other anaerobic pathways that were not investigated by the current study (i.e. pathways ending in pyruvate, valeric acid or alanine).

Organisms from a variety of phyla, including crustaceans(Hervant et al., 1995), birds(Faraci, 1991; Maina, 2000), bats(Maina, 2000), humans(Gozal and Gozal, 2001; Hoppeler and Vogt, 2001),mussels (Chen et al., 2001),fish (Galis and Barel, 1980; Cech and Crocker, 2002) and toads (Hou and Huang, 1999)increase ventilation rate and/or volume in response to acute hypoxic exposure. Tadpole shrimp apparently lack this typical ventilatory response or perhaps the response was not observed due to the length of acclimation prior to data collection (Fig. 2). Ventilation rates did not increase in response to hypoxic exposure down to 1 kPa O₂. Further, developmental P_O2did not affect ventilation rates under normoxic conditions or in response to hypoxic exposure. Tadpole shrimp appendages are used for locomotion and the generation of feeding currents, in addition to respiratory gas exchange(Fryer, 1988). If ventilatory rates were increased in response to hypoxic exposure, it could cause the tadpole shrimp to swim faster and/or may alter the currents used for food collection. Typical ventilatory hypoxic responses may be lacking because those alterations may have adverse effects on locomotion and feeding.

Ventilatory amplitude was used in this study as an index of ventilation volume. Ventilatory amplitude did not increase in response to hypoxic exposure in any of the rearing groups. Below P_CRIT, amplitude was lowest in tadpole shrimp reared under severe hypoxic conditions but was not significantly different from amplitude under normoxic conditions. Lower ventilatory amplitude in animals reared under severe hypoxic conditions may result from direct effects of O₂ limitation on the respiratory appendage movement or may be a strategy to minimize metabolic demand under severe hypoxic conditions (Maina,2000). Additionally, alterations in the angle of the appendages could account for ventilatory volume changes without observed changes in amplitude.

Tadpole shrimp cardiovascular responses to hypoxic exposure follow a unique pattern unlike that documented for other crustaceans(McMahon, 2001). In most crustaceans, acute hypoxic exposure results in a decrease in fh (bradycardia) and concomitant increase in Vs to maintain or increase Q̇(Reiber, 1995; McMahon, 2001). Smaller crustaceans, such as water fleas (Daphnia magna)(Paul et al., 1998) and grass shrimp (Palaemonetes pugio)(Harper and Reiber, 1999),increase fh (tachycardia) to maintain Q̇ in response to hypoxic exposure. In tadpole shrimp, however, cardiac function appears to be highly insensitive to hypoxic exposure (Fig. 3). Again, it should be acknowledged that this may be an artifact of the experimental protocol in that an acute hypoxic response may have been present but not observed because it took place during the acclimation period. Those animals reared under normoxic and moderate hypoxic conditions did not change fh or Vs when exposed to 5 kPa O₂. Below this, a bradycardia was observed and resulted in decreased Q̇. All cardiac parameters were maintained down to 1 kPa O₂ in tadpole shrimp reared under severe hypoxic conditions.

The difference between the cardiac response of tadpole shrimp and those of other crustaceans may result from differences in mechanisms of cardiac regulation. The heartbeat of many crustaceans is regulated through periodic bursting activity of cardiac ganglion, excitatory neurons that innervate the myocardium (Yamagishi et al.,2000). Hypoxia-induced bradycardia in those crustaceans may be mediated by a direct effect of lack of O₂ to the cardiac ganglion(Wilkens, 1999). Tadpole shrimp hearts have a myogenic mechanism of regulation whereby the cardiac muscle has endogenous rhythmic properties and does not rely on neural impulses to contract (Yamagishi et al.,1997; Yamagishi et al.,2000). The heart of tadpole shrimp does not respond to hypoxic exposure in the typical compensatory manner observed in other crustaceans. Myogenic regulation apparently supports cardiac function over a wide range of P_O2s, including exposure to severe hypoxic conditions.

Hemoglobin concentrations obtained in this study were comparable to previously reported concentrations for Triops(Horne and Beyenbach, 1974; Scholnick and Snyder, 1996; Guadagnoli et al., 2005). Hemoglobin concentration was higher in animals that developed under severe hypoxic conditions, yet was not proportional to developmental P_O2 (Fig. 4). Animals reared under moderate hypoxic conditions produced less hemoglobin than animals reared under either normoxic or severe hypoxic conditions. Moderate hypoxic conditions may represent the level of P_O2 in which O₂ uptake is sufficient to meet metabolic demand. Alternatively, it may represent a trigger for genetic up- or down-regulation of hypoxia-inducible genes prior to the translation of hemoglobin protein subunits.

Hypoxia-induced hemoglobin synthesis appears to be a compensatory response that allows T. longicaudatus and several other branchiopods to regulate O₂ uptake and transport in their euryoxic habitats. Hypoxic exposure (4–5 kPa O₂) induced more than a 10-fold increase in hemoglobin concentration within 10 days in adult D. magna(Goldmann et al., 1999) and a threefold increase within three weeks in adult brine shrimp (Artemia salina) (Gilchrist,1954). Hypoxic exposure (1–3 kPa O₂) induced a significant increase in protein concentrations within seven days in adult T. longicaudatus (Fig. 5). Since hemoglobin is the only major protein in tadpole shrimp hemolymph, protein concentration was accepted to be representative of hemoglobin concentration (Horne and Beyenbach, 1971). Within 10 days, hemoglobin concentrations increased to the levels observed in tadpole shrimp reared under severe hypoxic conditions. The induction of hemoglobin synthesis observed in T. longicaudatus was less than the induction observed in A. salinaor D. magna. These results indicate that tadpole shrimp exposed to hypoxic conditions acclimate by increasing hemoglobin concentration. Once the Hb is produced, though, it remains even when animals are returned to normoxic water (Fig. 5)(Guadagnoli et al., 2005).

Hemoglobin O₂-binding affinity was enhanced (decreased P₅₀) in tadpole shrimp reared under severe hypoxic conditions relative to those reared under normoxic conditions(Fig. 6). Differences in O₂-binding affinity are thought to have resulted from changes in subunit assembly of the functional hemoglobin molecule(Guadagnoli et al., 2005). This has been observed in other branchiopods such as D. magna and A. salina (Bowen et al.,1969; Waring et al.,1970; Sugano and Hoshi,1971; Kobayashi et al.,1988; Goldmann et al.,1999). Specific assembly of hemoglobin subunits could be dependent on environmental P_O2, internal chemistry or protein concentration (Sugano and Hoshi,1971; Kobayashi et al.,1988; Fago and Weber,1995; Goldmann et al.,1999). Since each subunit differs in O₂-binding affinity and cooperativity, changes in assembly directly affect O₂-binding affinity of the functional hemoglobin molecule(Kobayashi et al., 1988). The observed differences in hemoglobin O₂-binding affinity and cooperativity support the hypothesis that different P_O2 levels induce different hemoglobin isoforms or differential subunit assembly in T. longicaudatus (Figs 6, 7).

Tadpole shrimp reared under severe hypoxic conditions appear to have an enhanced ability to transport and possibly store O₂ obtained from the environment due to their increased O₂-carrying capacity(Table 1). Tadpole shrimp reared under severe hypoxic conditions had a fivefold increase in hemolymph O₂-carrying capacity compared with those reared under normoxic conditions. Increased hemolymph O₂-carrying capacity resulted from increased hemoglobin concentration and O₂-binding affinity observed in those animals. Hemoglobin with high O₂-carrying capacity can often serve a storage function (Fago and Weber, 1995). Tadpole shrimp frequently surface and expose their respiratory appendages to the air–water interface. Tadpole shrimp may obtain and store O₂ from the higher P_O2 surface water for use during feeding and hunting in severely hypoxic regions of the pool. Surfacing behavior has been shown to increase with decreased P_O2, further supporting a storage function of tadpole shrimp hemoglobin(Scholnick and Snyder,1996).

Oxygen-delivery potential, which takes into account hemolymph O₂content and Q̇, appeared to be dependent on developmental P_O2. Maximal O₂-delivery potential for tadpole shrimp was lowest in animals reared under normoxic conditions and highest in those reared under severe hypoxic conditions (Table 1). However, this was not a direct effect of developmental environment since hemoglobin synthesis could be increased throughout the life of the animal. Adult tadpole shrimp reared under normoxic conditions significantly increased hemoglobin concentrations in response to severe hypoxic exposure.

The coupling of V̇_O2 with cardiovascular transport reveals the level of coordination in tissue O₂ demand relative to O₂ delivery(Territo and Burggren, 1998). In tadpole shrimp, differences in cardiovascular contribution observed among the rearing groups were due to increased hemoglobin concentration and O₂-binding affinity since there was no observed cardiovascular hypoxic response. Oxygen consumption of tadpole shrimp reared under normoxic conditions was not as dependent on cardiovascular transport as it was in tadpole shrimp reared under hypoxic conditions(Fig. 8). The coupling of V̇_O2 with transport was consistently above 1 in tadpole shrimp reared under normoxic conditions; however, V̇_O2/Q̇_O2decreased in response to hypoxic exposure. This reduction suggests that V̇_O2 was reduced relative to cardiac output since (1) cardiac output did not increase in response to hypoxic exposure and (2) hemoglobin levels remain constant during acute hypoxic exposure. Animals reared under normoxic conditions increased convective processes to supply O₂ demands or relied on diffusional processes when exposed to severe hypoxic conditions.

Cardiovascular contribution to O₂ delivery was greatest in animals reared under severe hypoxic conditions. Oxygen delivery was enhanced in those animals via increased hemoglobin concentration and O₂-binding affinity. The increased O₂ delivery to the aerobic, metabolically active heart muscle may have supported cardiac contraction at normoxic rates under hypoxic conditions. Tadpole shrimp reared under severe hypoxic conditions maintained fh and Q̇ when exposed to severe hypoxic conditions while the other rearing groups did not. It was concluded that as tadpole shrimp produce more hemoglobin, they increase O₂ delivery relative to O₂ supply and enhance cardiac function during hypoxic exposure.

Developmental P_O2 may not be sufficient to induce permanent changes in adult tadpole shrimp physiological capabilities. Metabolic rate, ventilatory rate and metabolic response to acute hypoxic exposure were independent of developmental P_O2. Differences in cardiac response to hypoxic exposure among the rearing groups were probably due to compensatory increases in hemoglobin concentration and O₂-binding affinity rather than to developmental differences. Based on the results of this study, hypoxia-induced hemoglobin synthesis represents an effective compensatory mechanism that allows tadpole shrimp to flexibly regulate O₂ uptake and transport under euryoxic conditions. Differences that result from increased hemoglobin concentration should not be considered developmental effects until it is shown that (1) hemoglobin type or subunit arrangement is dependent on developmental P_O2 and not directed by ambient P_O2 during hemoglobin synthesis or (2)hypoxia-inducible genes were switched on during development and are continually expressed in the adult even if the animal is exposed to normoxic conditions.