Evidence for down-regulation of ethanolic fermentation and K+ effluxes in the coleoptile of rice seedlings during prolonged anoxia.

Ethanolic fermentation, the predominant catabolic pathway in anoxia-tolerant rice coleoptiles, was manipulated in excised and 'aged' tissues via glucose feeding. Coleoptiles with exogenous glucose survived 60 h of anoxia, as evidenced by vigorous rates of K+ and phosphate net uptake and growth of roots and shoots when re-aerated. In contrast, coleoptiles without exogenous glucose showed net losses of K+ and phosphates starting 12 h after anoxia was imposed and these did not recover fully when re-aerated after 60 h of anoxia. Ethanol production (micromol x g(-1) FW x h(-1)) declined from about 7.5 during the first 12 h of anoxia to 5 or 2.2 after 48-60 h, in coleoptiles with or without exogenous glucose, respectively. Carbohydrate concentrations changed only slightly in anoxic coleoptiles with exogenous glucose due to net glucose uptake at 2.6 micromol x g(-1) FW x h(-1). Ethanolic fermentation, and therefore ATP production, may have been down-regulated after an initial period of acclimation to anoxia in coleoptiles with exogenous glucose. Maintenance requirements for energy were assessed to be 3.4-7.6-fold lower in these anoxic coleoptiles than published estimates for non-growing aerated leaf tissues. A modest part of the required economy in energy consumption would have been derived from diminished ion transport; anoxia reduced K+ and phosphate net uptake by 70-90% in these coleoptiles. K+ efflux was 10-fold lower in anoxic than in aerated coleoptiles with exogenous glucose. Using the unidirectional efflux equation, the membrane permeability to K+ was estimated to be 17-fold lower in anoxic than in aerated coleoptiles, presumably due to predominantly closed K+ channels.     

Manipulation of ethanol production in anoxic rice coleoptiles by exogenous glucose determines rates of ion fluxes and provides estimates of energy requirements for cell maintenance during anoxia

Ethanol production by anoxic, excised, 7-10 mm tips of rice coleoptiles was manipulated using a range of exogenous glucose concentrations. Such a dose-response curve enabled good estimates at which level of ethanol production (and hence by inference ATP production), injury commenced and also allowed assessments of energy requirements for maintenance in anoxia. Rates of net uptake or loss of K+ and P by these excised coleoptile tips were related to rates of ethanol production (r2 of 0.59 and 0.68, respectively). At 72 h anoxia, ATP levels in excised tips were similar at 0, 2.5, and 50 mol m(-3) exogenous glucose, despite large differences in the inferred rates of ATP production. At 96 h anoxia, tips without exogenous glucose had low ATP concentrations; these may be the cause or the consequence of cell injury. In tips without glucose, injury was indicated by losses of K+ and Cl- between 72-96 h anoxia, and during the first hour after re-aeration, while later than 1 h after re-aeration, rates of net uptake were substantially lower than for re-aerated tips previously in anoxia with exogenous glucose. Between 96 h and 124 h anoxia, ion losses from tips without exogenous glucose increased while recovery of net uptake after re-aeration was very sluggish and incomplete. The energy requirement for maintenance of health and survival of anoxic coleoptile tips, expressed on a fresh weight basis, was lower than for three other anoxia-tolerant plant tissues/cells, studied previously. However, the energy requirement on a protein basis was assessed at 1.4 micromol ATP mg(-1) protein h(-1) and this value is 2.6-5.4-fold higher than for the other plant tissues/cells. Yet, this requirement was still only 58-88% of the published values for aerated tissues. The reason for this relatively high ATP requirement per unit protein in anoxic rice coleoptiles remains to be elucidated.     

Reversible decreases in ATP and PCr concentrations in anoxic turtle brain

A hallmark of anoxia tolerance in western painted turtles is relative constancy of tissue adenylate concentrations during periods of oxygen limitation. During anoxia heart and brain intracellular compartments become more acidic and cellular energy demands are met by anaerobic glycolysis. Because changes in adenylates and pH during anoxic stress could represent important signals triggering metabolic and ion channel down-regulation we measured PCr, ATP and intracellular pH in turtle brain sheets throughout a 3-h anoxic-re-oxygenation transition with ³¹P NMR. Within 30 min of anoxia, PCr levels decrease 40% and remain at this level during anoxia. A different profile is observed for ATP, with a statistically significant decrease of 23% occurring gradually during 110 min of anoxic perfusion. Intracellular pH decreases significantly with the onset of anoxia, from 7.2 to 6.6 within 50 min. Upon re-oxygenation PCr, ATP and intracellular pH recover to pre-anoxic levels within 60 min. This is the first demonstration of a sustained reversible decrease in ATP levels with anoxia in turtle brain. The observed changes in pH and adenylates, and a probable concomitant increase in adenosine, may represent important metabolic signals during anoxia.     

Anoxic survival of the isolated cerebellum of the turtle Pseudemis scripta elegans

Intact turtle brain provides a useful model for the study of anoxia and potential survival strategies, since this tissue maintains transmembrane ion gradients and ATP levels during prolonged anoxia and recovers functional activity afterwards. Since isolated tissues offer experimental advantages, the present study sought to determine effects of anoxia on the isolated turtle cerebellum and to define relationships between anoxia survival and glucose supply. In normoxia, the extracellular potassium ([K+]o) activity and evoked potentials were maintained with 5 mM glucose, but 20 mM glucose was required to maintain adenosine triphosphate (ATP) levels and prevent significant increases in [K+]o during anoxia. Inhibition of glycolysis by iodoacetic acid (IAA) during anoxia provoked large increases in [K+]o at all glucose levels. These results demonstrate the usefulness of the isolated turtle cerebellum for studies of anoxic survival since this tissue can maintain ATP levels and [K+]o during prolonged anoxia with 20 mM glucose in the artificial cerebrospinal fluid medium. They also suggest the presence of a Pasteur effect at least during the transition to a hypometabolic state.     

Effects of In Vitro Anoxia and Low pH on Acetylcholine Release by Rat Brain Synaptosomes

Acetylcholine and choline release from rat brain synaptosomes have been measured using a chemiluminescent technique under a variety of conditions set up to mimic anoxic insult, including conditions of low pH (6.2) and the presence of lactate plus pyruvate as substrate. Lactate plus pyruvate as substrate consistently gave higher respiration rates than glucose alone, but with either substrate (glucose or lactate plus pyruvate) the omission of Ca2+ caused an increase in respiration whereas a low pH caused a decreased respiration. Acetylcholine release under control conditions (glucose, pH 7.4) was Ca2+-dependent, stimulated by high K+ concentrations, and decreased significantly during anoxia but recovered fully after a period of postanoxic oxygenation. Low pH (6.2) suppressed K+ stimulation of acetylcholine release, and after a period of anoxia at low pH the recovery of acetylcholine release was only partial. With lactate plus pyruvate as substrate, the effects of anoxia and/or low pH on acetylcholine release and its subsequent recovery were exacerbated. Choline release from synaptosomes, however, was not affected by anoxic/ionic conditions in the same way as acetylcholine release. At low pH (6.2) there was a marked reduction in choline release both under aerobic and anoxic conditions. These results suggest that acetylcholine release per se from the nerve is very sensitive to anoxic insult and that the low pH occurring during anoxia may be an important contributory factor.     

Effects of β₂-agonists on force during and following anoxia in rat extensor digitorum longus muscle

Electrical stimulation of isolated muscles may lead to membrane depolarization, gain of Na(+), loss of K(+) and fatigue. These effects can be counteracted with β(2)-agonists possibly via activation of the Na(+)-K(+) pumps. Anoxia induces loss of force; however, it is not known whether β(2)-agonists affect force and ion homeostasis in anoxic muscles. In the present study isolated rat extensor digitorum longus (EDL) muscles exposed to anoxia showed a considerable loss of force, which was markedly reduced by the β(2)-agonists salbutamol (10(-6) M) and terbutaline (10(-6) M). Intermittent stimulation (15-30 min) clearly increased loss of force during anoxia and reduced force recovery during reoxygenation. The β(2)-agonists salbutamol (10(-7)-10(-5) M) and salmeterol (10(-6) M) improved force development during anoxia (25%) and force recovery during reoxygenation (55-262%). The effects of salbutamol on force recovery were prevented by blocking the Na(+)-K(+) pumps with ouabain or by blocking glycolysis with 2-deoxyglucose. Dibutyryl cAMP (1 mM) or theophylline (1 mM) also improved force recovery remarkably. In anoxic muscles, salbutamol decreased intracellular Na(+) and increased (86)Rb uptake and K(+) content, indicating stimulation of the Na(+)-K(+) pumps. In fatigued muscles salbutamol induced recovery of excitability. Thus β(2)-agonists reduce the anoxia-induced loss of force, leading to partial force recovery. These data strongly suggest that this effect is mediated by cAMP stimulation of the Na(+)-K(+) pumps and that it is not related to recovery of energy status (PCr, ATP, lactate).     

Carbohydrate metabolism in the central nervous system of the megalobulimus oblongus snail during anoxia exposure and post-anoxia recovery

The effects of anoxic exposure and the post-anoxia aerobic recovery period on carbohydrate metabolism in the central nervous system (CNS) of the land snail Megalobulimus oblongus, an anoxia-tolerant land gastropod, were studied. The snails were exposed to anoxia for periods of 1.5, 3, 6, 12, 18, or 24 hr. In order to study the post-anoxia recovery phase, snails exposed to a 3-hr period of anoxia were returned to aerobic conditions for 1.5, 3, 6, or 15 hr. Glycogen and glucose concentrations in the CNS, hemolymph glucose concentration, and glycogen phosphorylase (active form, GPa) activity in the CNS were analyzed. Anoxia does not significantly affect the concentration of CNS glucose but induces hyperglycemia and a reduction of CNS GPa activity. The glycogen concentration was decreased at 12 hr of anoxia; however, by 18 and 24 hr in anoxia, the glycogen content was not significantly different from basal control values. During the post-anoxia period, the reduction in GPa activity and the increased hemolymph glucose concentration induced by anoxia returned to control values. These results suggest that the CNS of M. oblongus may use hemolymph glucose to fulfill the metabolic demands during anoxia. However, the hypothesis of tissue metabolic arrest cannot be excluded.     

Adaptative response of Vitis root to anoxia

The effect of anoxia on the energy economy of root cells was studied by measuring heat production, ethanol and ATP production, K(+) fluxes and electrical activity in two Vitis species, V. riparia and V. rupestris, that differ in their tolerance to anoxia. Anoxia triggered a marked decrease of metabolic activity (measured by microcalorimetry) and of ATP levels in both species. In V. riparia after the first 2 h of anoxia, the decrease in the rate of heat production was not associated with a further significant decrease in ATP content, whereas in V. rupestris the ATP level continued to decrease until very low values were reached. The concomitant increase in the rate of ethanol production did not compensate for the decreased aerobic ATP supply. In V. rupestris, anoxia typically led to energy deficit and ATP imbalance, together with the subsequent disruption of ion homeostasis and cell death. In V. riparia, the strong decrease in K(+) membrane permeability together with the fast down-regulation of the electrical signals allowed the cells to avoid severe ion imbalances during prolonged anoxic episodes.     

The effect of prolonged anoxia at 3°C on tissue high energy phosphates and phosphodiesters in turtles: A 31P-NMR study

Selected tissues (skeletal muscle, heart ventrical, and liver), sampled from turtles (Chrysemys picta bellii) at 3°C either under normoxic conditions or after 12 weeks of anoxic submergence were quantiaatively analysed for intracellular pH and phosphorus metabolites using ³¹P-NMR. Plasma was tested for osmolality and for the concentrations of lactate, calcium, and magnesium to confirm anoxic stress. We hypothesized that, in the anoxic animals, tissue ATP levels would be maintained and that the increased osmolality of the body fluids of anoxic turtles would be accounted for by a corresponding increase in the concentrations of phosphodiesters. The responses observed differed among the three tissues. In muscle, ATP was unchanged by anoxia but phosphocreatine was reduced by 80%; in heart, both ATP and phosphocreatine fell by 35–40%. The reduction in phosphocreatine in heart tissue at 3°C was similar to that observed in isolated, perfused working hearts from turtles maintained at 20°C but no decrease in ATP occurred in the latter tissues. In liver, although analyses of several specimens were confounded by line-broadening, neither ATP nor phosphocreatine was detectable in anoxic samples. Phosphosdiesters were detected in amounts sufficient to account for 30% of normoxic cell osmotic concentration in heart and 11% and 12% in liver and muscle, respectively. The phosphodiester levels did not change in anoxia. Heart ventricular phosphodiester levels in turtles at 3°C were significantly higher than those determined for whole hearts from turtles at 20°C. ¹H, ¹³C and ³¹P NMR analyses of perchloric acid extracts of heart and skeletal muscle from 20°C turtles con firmed that the major phosphodiester observed by NMR in these tissues is serine ethanolamine phosphate. We conclude that the three types of tissues studied differ substantially in their ability to maintain levels of ATP during anoxia, and that liver may continue to function despite NMR-undetectable levels of this metabolite. In addition, we conclude that phosphodiesters do not serve as regulated osmolytes during anoxia, and that the functional significance of their high concentrations in turtle tissues remains uncertain.     

Nucleotide Levels Do Not Critically Determine Survival of Maize Root Tips Acclimated to a Low-Oxygen Environment

We tested the hypothesis that ATP levels and energy charge determine the resistance of maize (Zea mays) root tips to anoxia. We focused on root tips of whole maize seedlings that had been acclimated to low O2 by exposure to an atmosphere of 3% (v/v) O2 in N2. Acclimated anoxic root tips characteristically have higher ATP levels and energy charge and survive longer under anoxia than nonacclimated tips. We poisoned intact, acclimated root tips with either fluoride or mannose, causing decreases in ATP and energy charge to values similar to or, in most cases, below those found in nonacclimated anoxic tips. With the exception of the highest fluoride concentration used, the poisoned, acclimated tips remained much more tolerant of anoxia than nonacclimated root tips. We conclude that high ATP and energy charge are not components critical for the survival of acclimated root tips during anoxia. The reduced nucleotide status in poisoned, acclimated root tips had little effect on cytoplasmic pH regulation during anoxia. This result indicates that in anoxic, acclimated root tips either cytoplasmic pH regulation is not dominated by ATP-dependent processes or these processes can continue in vivo largely independently of any changes in ATP levels in the physiological range. The role of glycolytic flux in survival under anoxia is discussed.     

Role of fructose 2,6-bisphosphate in the stimulation of glycolysis by anoxia in isolated hepatocytes.

1. Incubation of hepatocytes from fed or starved rats with increasing glucose concentrations caused a stimulation of lactate production, which was further increased under anaerobic conditions. 2. When glycolysis was stimulated by anoxia, [fructose 2,6-bis-phosphate] was decreased, indicating that this ester could not be responsible for the onset of anaerobic glycolysis. In addition, the effect of glucose in increasing [fructose 2,6-bisphosphate] under aerobic conditions was greatly impaired in anoxic hepatocytes. [Fructose 2,6-bisphosphate] was also diminished in ischaemic liver, skeletal muscle and heart. 3. The following changes in metabolite concentration were observed in anaerobic hepatocytes: AMP, ADP, lactate and L-glycerol 3-phosphate were increased; ATP, citrate and pyruvate were decreased: phosphoenolpyruvate and hexose 6-phosphates were little affected. Concentrations of adenine nucleotides were, however, little changed by anoxia when hepatocytes from fed rats were incubated with 50 mM-glucose. 4. The activity of ATP:fructose 6-phosphate 2-phosphotransferase was not affected by anoxia but decreased by cyclic AMP. 5. The role of fructose 2,6-bisphosphate in the regulation of glycolysis is discussed.     

1H-NMR study of the metabolome of an exceptionally anoxia tolerant vertebrate, the crucian carp (Carassius carassius)

The crucian carp (Carassius carassius) can tolerate anoxia for days to months, depending on the temperature. In this study, we applied ¹H-NMR-based metabolomics to polar extracts of crucian carp brain, heart, muscle and liver samples obtained from fish exposed to either control normoxic conditions, acute anoxia (24 h), chronic anoxia (1 week) or reoxygenation (for 1 week following chronic anoxia) at 5 °C. Spectra of the examined tissues revealed changes in several energy-related compounds. In particular, anoxic stress resulted in decreased concentrations of phosphocreatine (muscle, liver) and glycogen (liver) and ATP/ADP (liver, heart and muscle) and increased concentrations of lactate (brain, heart, muscle) and beta-hydroxybutyric acid (all tissues). Likewise, increased concentrations of inhibitory compounds (glycine, gamma-amino butyric acid or GABA) and decreased concentrations of excitatory metabolites (glutamate, glutamine) were confirmed in the anoxic brain extracts. Additionally, a decrease of N-acetylaspartate (NAA), an important neuronal marker, was also observed in anoxic brains. The branched-chain amino acids (BCAA) valine/isoleucine/leucine increased in all anoxic tissues. Possibly, this general tissue increase can be due to an inhibited mitochondrial function or due to protein degradation/protein synthesis inhibition. In this study, the potential and strength of the ¹H-NMR is highlighted by the detection of previously unrecognized changes in metabolites. Specifically, myo-inositol substantially decreased in the heart of anoxic crucian carp and anoxic muscle tissue displayed a decreased concentration of taurine, providing novel insights into the anoxia responses of the crucian carp.     

High glucose protects single beating adult cardiomyocytes against hypoxia

In the heart, the opening of sarcolemmal ATP-sensitive K(+) (K(ATP)) channels seems to be crucial for the cardiac protection against hypoxia/ischaemia. In the present study, we have exposed cardiomyocytes under hypoxia to high extracellular glucose (30 mM). Under these conditions, intracellular concentration of 1,3-bisphosphoglycerate has increased confirming stimulation of glycolysis. Perforated patch-clamp electrophysiology revealed that hypoxia induces whole-cell K(+) current in cardiomyocytes more efficiently in the presence than in the absence of high glucose. Glucose significantly promoted survival of cardiomyocytes exposed to hypoxia. HMR 1098, an antagonist of sarcolemmal K(ATP) channels, inhibited glucose-induced activation of whole-cell K(+) current during hypoxia as well as glucose-mediated cytoprotection. An inhibitor of glyceraldehyde 3-phosphate dehydrogenase, iodoacetate, inhibited glycolysis in hypoxia and blocked the activation of sarcolemmal K(ATP) channels. Based on the obtained results, we conclude that the activation of sarcolemmal K(ATP) channels is involved in glucose-mediated cardioprotection.     

Evidence of anoxia-induced channel arrest in the brain of the goldfish (Carassius auratus)

The common goldfish (Carassius auratus) is extremely anoxia tolerant and here we provide evidence that "channel arrest" in the brain of these fish contributes to ATP conservation during periods of anoxia. Whole-cell patch-clamp recordings of slices taken from the telencephalon indicated that the N-methyl-d-aspartate (NMDA) receptor, an ionotropic glutamate receptor and Ca(2+)-channel, underwent a 40-50% reduction in activity during 40 min of acute anoxia. This is the first direct evidence of channel arrest in an anoxia-tolerant fish. Because goldfish produce ethanol as a byproduct of anaerobic metabolism we then conducted experiments to determine if the observed reduction in NMDA receptor current amplitude was due to inhibition by ethanol. NMDA receptor currents were not inhibited by ethanol (10 mmol L(-1)), suggesting that channel arrest of the receptor involved other mechanisms. Longer-term (48 h) in vivo exposure of goldfish to anoxic conditions (less than 1% dissolved O(2)) provided indirect evidence that a reduction in Na(+)/K(+)-ATPase activity also contributed to ATP conservation in the brain but not the gills. Anoxia under these conditions was characterized by a decrease in brain Na(+)/K(+)-ATPase activity of 30-40% by 24 h. Despite 90% reductions in the rates of ventilation, no change was observed in gill Na(+)/K(+)-ATPase activity during the 48-h anoxia exposure, suggesting that branchial ion permeability was unaffected. We conclude that rapid "channel arrest" of NMDA receptors likely prevents excitotoxicity in the brain of the goldfish, and that a more slowly developing decrease in Na(+)/K(+)-ATPase activity also contributes to the profound metabolic depression seen in these animals during oxygen starvation.     

Comparison of in vivo energy metabolism in the brain of rainbow trout, Salmo gairdneri and bullhead catfish, Ictalurus nebulosus during anoxia

1. Rainbow trout and bullhead catfish (Ictalurus nebulosus) were exposed to anoxic water inside a plastic tube until death (12 min for trout; 62 min for catfish). Immediately upon death, the brain was removed and analyzed for metabolites, high energy phosphate compounds, and metabolic fuel while the blood was analyzed for metabolites. 2. Control bullhead brains had higher concentrations of glycogen, ATP, creatine phosphate (CrP), and glucose than control trout. 3. After anoxia bullheads showed a significant decrease in ATP, CrP, and glycogen while lactate more than doubled in concentration. 4. After anoxia trout showed a doubling in brain lactate and a decline in glycogen, but no decline in ATP or CrP. There were no changes in brain glucose, ketone bodies, or alternative anaerobic end-products in either species although an elevation in blood isobutyrate was noted. 5. Brain death in the catfish may be due to depletion of fuel for anaerobiosis and a subsequent loss of ATP. In the trout there may be other causes such as a greater permeability of its neuronal membranes and alterations in intracellular free calcium stores.     

Investigation of Na(+),K(+)-ATPase on a solid supported membrane: the role of acylphosphatase on the ion transport mechanism

Charge translocation by Na(+),K(+)-ATPase was investigated by adsorbing membrane fragments containing Na(+),K(+)-ATPase from pig kidney on a solid supported membrane (SSM). Upon adsorption, the ion pumps were activated by performing ATP concentration jumps at the surface of the SSM, and the capacitive current transients generated by Na(+),K(+)-ATPase were measured under potentiostatic conditions. To study the behavior of the ion pump under multiple turnover conditions, ATP concentration jump experiments were carried out in the presence of Na(+) and K(+) ions. Current transients induced by ATP concentration jumps were also recorded in the presence of the enzyme alpha-chymotrypsin. The effect of acylphosphatase (AcP), a cytosolic enzyme that may affect the functioning of Na(+),K(+)-ATPase by hydrolyzing its acylphosphorylated intermediate, was investigated by performing ATP concentration jumps both in the presence and in the absence of AcP. In the presence of Na(+) but not of K(+), the addition of AcP causes the charge translocated as a consequence of ATP concentration jumps to decrease by about 50% over the pH range from 6 to 7, and to increase by about 20% at pH 8. Conversely, no appreciable effect of pH upon the translocated charge is observed in the absence of AcP. The above behavior suggests that protons are involved in the AcP-catalyzed dephosphorylation of the acylphosphorylated intermediate of Na(+),K(+)-ATPase.     

Effects of chemical anoxia on adrenergic responses of goldfish hepatocytes and the contribution of alpha- and beta-adrenoceptors

Adrenergic responses during normoxia and chemical anoxia were investigated in anoxia-tolerant hepatocytes from the goldfish, Carassius auratus. Epinephrine-stimulated glucose release was unaltered after 1 hr of chemical anoxia, the concentration of epinephrine required for half maximal stimulation of glucose release (K0.5(GLU)) ranging from 0.62 x 10(-8) to 2.05 x 10(-8) M. Similarly, the maximum rate of glucose release caused by hormonal stimulation was not affected by chemical anoxia. In anoxic goldfish hepatocytes [Ca2+](i) remained constant in nonstimulated cells but could be elevated by addition of epinephrine. The magnitude of this [Ca(2+)](i)-increase was dependent on the concentration of the catecholamine and this dependency was similar under normoxia (K0.5(Ca2+) = 1.17 x 10(-8) M) and chemical anoxia (K0.5(Ca2+) = 1.15 x 10(-8) M), as was the percentage of cells responding (77%) and displaying oscillatory [Ca2+]i response patterns (60%) after epinephrine addition, although the frequency of [Ca2+]i oscillations was significantly lower in anoxic cells. To analyze a possible shift in the importance of alpha- and beta-adrenoceptors during chemical anoxia, the effect of phentolamine and propranolol, alpha- and beta-adrenergic antagonists respectively, on epinephrine-stimulated glucose release was studied. Application of the alpha-antagonist caused a dose-dependent reduction of glucose-release which was similar under both conditions, whereas the sensitivity to the beta-antagonist was lowered after chemical anoxia. Taken together these results provide evidence that during chemical anoxia goldfish hepatocytes remain responsive to adrenergic stimulation and that there is a partial shift regarding the contribution of alpha- and beta-adrenergic pathways to the induction of cellular glucose release stimulated by epinephrine.     

Bone and shell contribution to lactic acid buffering of submerged turtles Chrysemys picta bellii at 3 degrees C

To evaluate shell and bone buffering of lactic acid during acidosis at 3 degrees C, turtles were submerged in anoxic or aerated water and tested at intervals for blood acid-base status and plasma ions and for bone and shell percent water, percent ash, and concentrations of lactate, Ca(2+), Mg(2+), P(i), Na(+), and K(+). After 125 days, plasma lactate concentration rose from 1.6 +/- 0.2 mM (mean +/- SE) to 155.2 +/- 10.8 mM in the anoxic group but only to 25.2 +/- 6.4 mM in the aerated group. The acid-base state of the normoxic animals was stable after 25 days of submergence. Plasma calcium concentration (?Ca(2+)) rose during anoxia from 3.2 +/- 0.2 to 46.0 +/- 0.6 mM and ?Mg(2+) from 2.7 +/- 0.2 to 12.2 +/- 0.6 mM. Both shell and bone accumulated lactate to concentrations of 135.6 +/- 35.2 and 163.6 +/- 5.1 mmol/kg wet wt, respectively, after 125 days anoxia. Shell and bone ?Na(+) both fell during anoxia but the fate of this Na(+) is uncertain because plasma ?Na(+) also fell. No other shell ions changed significantly in concentration, although the concentrations of both bone calcium and bone potassium changed significantly. Control shell water (27.8 +/- 0.6%) was less than bone water (33.6 +/- 1.1%), but neither changed during submergence. Shell ash (44.7 +/- 0.8%) remained unchanged, but bone ash (41.0 +/- 1.0%) fell significantly. We conclude that bone, as well as shell, accumulate lactate when plasma lactate is elevated, and that both export sodium carbonate, as well as calcium and magnesium carbonates, to supplement ECF buffering.