"Science, Faculty of"@en . "Zoology, Department of"@en . "DSpace"@en . "UBCV"@en . "Doll, Christopher Joseph"@en . "2009-04-06T19:45:02Z"@en . "1993"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "The high sensitivity of the mammalian brain and the insensitivity of the turtle brain to O\u00E2\u0082\u0082\r\ndeprivation led to the use of cortical slice preparations in both species being utilized for a\r\ncomparative study of anoxia tolerance. To assess anoxic survival, intracellular recording\r\ntechniques were employed. Turtle neurons survived both anoxia (aCSF equilibrated with\r\n95% N2 /5 % CO\u00E2\u0082\u0082) and pharmacological anoxia (anoxia + 1mM NaCN) for 180 min. with\r\nno measurable degradation. Rat pyramidal neurons responded with a decrease in whole cell\r\nresistance followed by transient hyperpolarization and a subsequent depolarization to a zero\r\nmembrane potential (41.3 \u00C2\u00B1 6.5 min., anoxia; 25.8 \u00C2\u00B1 12.6 min., pharmacological anoxia).\r\nPharmacological ischemia (pharmacological anoxia + iodoacetate 10 mM) caused a rapid\r\ndecrease in whole cell resistance, transient hyperpolarization, and a rapid depolarization in\r\nboth turtle (4.6 \u00C2\u00B1 1.1 min.) and rat (3.1 + 0.5 min.) neurons. Ouabain perfusion caused a\r\nrapid depolarization in the rat cortical neuron (8.6 \u00C2\u00B1 1.1 min.), but no initial decrease in\r\nwhole cell resistance or a hyperpolanzation.\r\nCalorimetric measures converted to ATP utilization rate indicated that the turtle cortical\r\nslice has an initial ATP utilization of 1.72 \u00CE\u00BCmoles ATP/g/min. which agrees closely to in\r\nvivo whole brain metabolic measures. This value supports a 9 fold lower metabolic rate\r\ncompared to analogous guinea pig cortical slice preparations. Based on heat depression\r\nmeasures, resulting ATP utilization estimates indicated a metabolic depression of 30 %\r\n(nitrogen) and 42% (pharmacological anoxia). Heat flux changes over pharmacological\r\nanoxia, support a large initial Pasteur effect which gradually declines over the 120 min. insult\r\ninterval. Activities of hexokinase and lactate dehydrogenase were similar between the rat\r\nand turtle cortical slice (25 \u00C2\u00B0C), but the turtle cortex only expressed 80 % of the activity of\r\nthe rat cortex for citrate synthase. Surprisingly, the turtle cortical slice did not exhibit a\r\nchange in any measured adenylate parameter up to 120 min. of anoxia or pharmacological\r\nanoxia. Significant changes did occur in [ADP], ATP/ADP ratio, and energy charge at 240\r\nmin.\r\nIn order to assess difference in ion leakage in both the turtle and rat pyramidal neurons,\r\nintracellular recording techniques for short term anoxia (120 min.) and whole cell patch\r\nclamp techniques (on cell populations) for long term anoxia (6 -9 hrs.) were utilized. Both\r\ntechniques indicated that turtle cortical pyramidal cells did not change in conductance (whole\r\ncell conductance or specific membrane conductance) with anoxia. Whole cell patch clamp\r\ntechniques supported a 4.2 fold higher specific membrane conductance in rat pyramidal\r\nneurons compared to turtle neurons at the same temperature (25 \u00C2\u00B0C) which was accentuated\r\nby temperature so that rat pyramidal neurons at 37\u00C2\u00B0C were 22 times more conductive than\r\nturtle neurons at 15\u00C2\u00B0C. A conductance Q\u00E2\u0082\u0081\u00E2\u0082\u0080 of 1.9 was measured for both turtle (15-25\u00C2\u00B0C)\r\nand rat (25-35\u00C2\u00B0C) pyramidal neurons. To asses pumping activity capacity, Na\u00E2\u0081\u00BA-K\u00E2\u0081\u00BA\r\nATPase activity was measured in cortical slices of both species. At the same temperature (25\r\n\u00C2\u00B0C) a 2.3 fold higher activity was measured in the rat cortex compared to the turtle\r\nsupporting the patch clamp results of a lower normoxic specific membrane conductance in\r\nthe turtle cortex.\r\nTaken together these results support that the turtle brain is able to survive anoxia through\r\nan enhanced glycolytic capability, a low normoxic brain metabolism with the ability to\r\nfurther depress metabolism during anoxia. Electrophysiological techniques support reduced\r\nion pumping through reduced ion leakage as one mechanism for a depressed normoxic\r\nmetabolic rate in the turtle cortical slice but do not support further down regulation of\r\nchannel activity with anoxia."@en . "https://circle.library.ubc.ca/rest/handle/2429/6828?expand=metadata"@en . "3452894 bytes"@en . "application/pdf"@en . "MECHANISMS OF ANOXLA TOLERANCEINTHE TURTLE CORTEXbyCHRISTOPHER JOSEPH DOLLB.A., The University of Hawaii, 1986A THESIS SUBMI1ThD IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe faculty of Graduate Studies(Department of Zoology)We accept this thesis as conformingto the required standardThe University of British ColumbiaOctober 1993\u00C2\u00A9 Christopher Joseph Doll, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)___________________________Department of ZoologyThe University of British ColumbiaVancouver, CanadaDate October, 1993DE-6 (2/88)ABSTRACTThe high sensitivity of the mammalian brain and the insensitivity of the turtle brain to 02deprivation led to the use of cortical slice preparations in both species being utilized for acomparative study of anoxia tolerance. To assess anoxic survival, intracellular recordingtechniques were employed. Turtle neurons survived both anoxia (aCSF equilibrated with95% N2 /5 % COD) and pharmacological anoxia (anoxia + 1mM NaCN) for 180 mm. withno measurable degradation. Rat pyramidal neurons responded with a decrease in whole cellresistance followed by transient hyperpolarization and a subsequent depolarization to a zeromembrane potential (41.3 \u00C2\u00B1 6.5 mm., anoxia; 25.8 \u00C2\u00B1 12.6 mm., pharmacological anoxia).Pharmacological ischemia (pharmacological anoxia + iodoacetate 10 mM) caused a rapiddecrease in whole cell resistance, transient hyperpolarization, and a rapid depolarization inboth turtle (4.6 \u00C2\u00B1 1.1 mm.) and rat (3.1 + 0.5 mm.) neurons. Ouabain perfusion caused arapid depolarization in the rat cortical neuron (8.6 \u00C2\u00B1 1.1 mm.), but no initial decrease inwhole cell resistance or a hyperpolanzation.Calorimetric measures converted to ATP utilization rate indicated that the turtle corticalslice has an initial ATP utilization of 1.72 pmoles ATP/g/min. which agrees closely to invivo whole brain metabolic measures. This value supports a 9 fold lower metabolic ratecompared to analogous guinea pig cortical slice preparations. Based on heat depressionmeasures, resulting ATP utilization estimates indicated a metabolic depression of 30 %(nitrogen) and 42% (pharmacological anoxia). Heat flux changes over pharmacologicalanoxia, support a large initial Pasteur effect which gradually declines over the 120 mm. insultinterval. Activities of hexokinase and lactate dehydrogenase were similar between the ratand turtle cortical slice (25 \u00C2\u00B0C), but the turtle cortex only expressed 80 % of the activity ofthe rat cortex for citrate synthase. Surprisingly, the turtle cortical slice did not exhibit achange in any measured adenylate parameter up to 120 mm. of anoxia or pharmacologicalanoxia. Significant changes did occur in [ADP], ATP/ADP ratio, and energy charge at 240mm.IIIn order to assess difference in ion leakage in both the turtle and rat pyramidal neurons,intracellular recording techniques for short term anoxia (120 mm.) and whole cell patchclamp techniques (on cell populations) for long term anoxia (6 -9 hrs.) were utilized. Bothtechniques indicated that turtle cortical pyramidal cells did not change in conductance (wholecell conductance or specific membrane conductance) with anoxia. Whole cell patch clamptechniques supported a 4.2 fold higher specific membrane conductance in rat pyramidalneurons compared to turtle neurons at the same temperature (25 \u00C2\u00B0C) which was accentuatedby temperature so that rat pyramidal neurons at 37\u00C2\u00B0C were 22 times more conductive thanturtle neurons at 15\u00C2\u00B0C. A conductance Qio of 1.9 was measured for both turtle (15-25\u00C2\u00B0C)and rat (25-35\u00C2\u00B0C) pyramidal neurons. To asses pumping activity capacity, Na-KATPase activity was measured in cortical slices of both species. At the same temperature (25\u00C2\u00B0C) a 2.3 fold higher activity was measured in the rat cortex compared to the turtlesupporting the patch clamp results of a lower normoxic specific membrane conductance inthe turtle cortex.Taken together these results support that the turtle brain is able to survive anoxia throughan enhanced glycolytic capability, a low normoxic brain metabolism with the ability tofurther depress metabolism during anoxia. Electrophysiological techniques support reducedion pumping through reduced ion leakage as one mechanism for a depressed normoxicmetabolic rate in the turtle cortical slice but do not support further down regulation ofchannel activity with anoxia.111TABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS ivLIST OFTABLES viiLIST OF FIGURES viiiABBREVIATIONS xACKNOWLEDGMENTS xiiCHAPTER 1: ANIMALANOXIA TOLERANCE-AN INTRODUCTION 1Preface 1A Species Overview 1Turtle Whole Body Anoxic Adaptations 3Tissue Hypoxic Response 5Energy And Anaerobiosis 5Insights into Brain Anoxia Tolerance 9Anoxic Membrane Coupled Function 10Thesis Overview 11CHAPTER 2: EFFECTS OF ANOXIA AND PHARMACOLOGICAL ISCHEMIAON TURTLE AND RAT CORTICAL NEURONS 13Preface 13Introduction 13Mammalian CNS Response 13Turtle CNS Anoxic Response 15Methods 15Tissue Preparation 15Data Acquisition 16Fluid Composition 17Results 17ivAdditional Statistics .26Discussion 27CHAFfER 3: A BIOCHEMICAL AND MICROCALORMET\u00E2\u0080\u0099RIC STUDY OF THETURTLE CORTEX 33Preface 33Introduction 33Methods 35Slice Preparation 35Artificial Cerebrospinal Fluid 36Adenylates 36Chromatography 37Microcalorimetry 37Enzymatic Methods 39Results 40Calorimetry 40Enzymatic Analysis 47Discussion 48Adenylates 48Normoxic Metabolic Rates 48Anoxic Metabolic Rates 49Anoxic Gap .50Glycolysis 50Enzymatic Analysis 53Conclusion 54CHAFfER 4: A CRITICAL TEST OF CHANNEL ARREST 55Preface 55Introduction 55VMethods .57General .57Patch Clamp Tissue Preparation 57Patch Clamping Techniques 57Data analysis 58Na-K-ATPase Activity 59Results 59Intracellular Recording 68Patch Clamp Results 68Na-K-ATPase Activity 69Discussion 70CHAPTER 5: A THEORETICAL APPROACH TO ANOXIA TOLERANCE- ACONCLUSION 74Preface 74Introduction 74An Interpretive Model 74Mammalian Brain 76Turtle CNS Response 79Metabolic Rate 80Ectothermy vs. Endothermy 84LITERATURE CITED 86APPENDIX A: TECHNIQUES 98APPENDIX B: TERMONOLOGY 103viLIST OF TABLES1. Effect of anoxia and pharmacological anoxia on membrane potential and actionpotentials of turtle neurons 202. Enzyme activities in the cortex of the rat and turtle 463. Turtle cortical pyramidal cell patch clamp values 634. Rat cortical pyramidal cell values 645. Effect of temperature on membrane ion leakage 656. Na-K-ATPase activity in Turtle and Rat Cortex 677. Glucose consumption measures 81viiLIST OF FIGURES1. Theoretical response of an oxygen conformer and an oxygen regulator to varyingoxygen partial pressures 62. Fermentation pathways identified in vertebrates and invertebrates 73. Time to depolarization for turtle and rat pyramidal cortical neurons in response tovarious pharmacological treatments 184. Effect of 180 mins. of anoxia on a continuously impaled turtle cortical pyramidalcell 195. Response of rat and turtle pyramidal cortical neurons to pharmacologicalischemia 216. Effect of perfusing solutions of anoxia and pharmacological anoxia on rat pyramidalneurons 227. Effect of perfusing iodoacetate on turtle and rat pyramidal cells 238. Effect of perfusing ouabain on rat pyramidal neurons 249. Representative chart recordings of heat dissipation 4110. The individual percent heat depression from the predicted corresponding controlvalue 4211. The average heat depression (percentage) relative to the predicted correspondingcontrol value 4312. Effects of nitrogen perfusion and pharmacological anoxia on heat dissipation andmetabolism 4413. The effect of anoxia and pharmacological anoxia on adenylates, energy charge, andadenylate ratios for turtle cortical slices 4514. Four possible scenarios for the response of cortical brain slices to pharmacologicalanoxia 5215. Examples of membrane charging curves 6016. Conductance and membrane potential changes with anoxic exposure 61viii17. Current (1)-voltage (V) plot.6218. Conductance ratios 6619. Possible scenario for the degeneration and survival of the turtle and rat neuron 7520. Adjusted metabolic rate for the mammalian cortical slice 8221. Electrophysiological recording techniques 9922. The four configurations of patch clamping 10023. The slice chamber recording set-up and perfusion system for both patch clampingand intracellular recording techniques 101ixABBREVIATIONSaCSF artificial cerebrospinal fluidATP adenosine triphosphate[ATP] adenosine triphosphate concentration[ATPJ1 intercellular adenosine tnphosphate concentrationADP adenosine diphosphateAMP adenosine monophosphateA0 offsetAl maximum voltageCa+2 calcium[Ca+2]o calcium concentration extracellular[Ca+2]j calcium concentration intracellularCaC12 calcium chloridedegrees Celsiuscm centimeterCm specific membrane capacitanceC02 carbon dioxideCN cyanideCNS central nervous systeme natural logarithmDTT 5dithiothreitolEGTA ethylene glycol bis(B-aminoethyl) etherEPSP excitatory post synaptic potentialEEG electroencephalographGABA y-aminobutyric acidGAP Glyceraldehyde 3-phosphateGQ gigaohm0m specific membrane conductanceGTP guanosine 5\u00E2\u0080\u0099-tnphosphateGw whole cell conductanceH20 waterH+ hydrogen protonHC1 hydrochloric acidHPLC high pressure liquid chromatographyI currentpotassiumKATp ATP sensitive potassium channelsKa Ca sensitive potassium channels[K1i potassium concentration intracellularpotassium concentration extracellularKC1 potassium chloridekHz kilohertzKOH potassium hydroxideIAA iodoacetic acid\u00E2\u0080\u0098AHp after hyperpolarization currentmA milliampere(s)MgCl magnesium chloridemm. minute(s)mM millimolarxmS millisecond(s)mV millivolt(s)MYA million years agoMQ megaohm(s)N2 nitrogenNa+ sodium[Na+] sodium concentration[Na+]j sodium concentration intracellular[Na+}o sodium concentration extracellularnA nanoamperesNaCN sodium cyanideNaC1 sodium chlorideNAD oxidized nicotinamide ademne dinucleotideNADH reduced nicotinamide adenine dinucleotideNa-K-ATPase sodium, potassium - ATPaseNADP oxidized nicotinamide adenine dinucleotide phosphateNADPH reduced nicotinamide adenine dinucleotide phosphateNaHCO3 Sodium bicarbonateNaH2PO4 monosodium phosphateNaOH sodium hydroxidenS nanosecond(s)nm nanometer(s)02 oxygenpA picoampere(s)pH inverse log of percent hydrogen ion concentrationPi inorganic phosphatePo2 partial pressure of oxygenQ 10 effect of a 10\u00C2\u00B0C temperature changeR least squares residualRm specific membrane resistanceRw whole cell resistanceTc time constantTris Cl tris (hydroxymethyl) aminomethaneV voltageV02 oxygen consumptionNonaiphabetical abbreviationsmicrofarad(s)cmole micromole(s)micrometer(s)micromolarmicrosemin(s)zS entropyQ ohm(s)1 ,3DPG 1 ,3-DiphosphoglyceratexiACKNOWLEDGMENTSI would like to thank both of my supervisors, Peter Hochachka and Peter Reiner, ingeneral, for their support, encouragement, constant enthusiasm, and patience. In particular, Iwould like to thank Peter Hochachka for teaching me how to identify fundamental problems,and Peter Reiner for teaching me conciseness both in experimental design and writingtechniques. Special thanks go to Andy Laycock who\u00E2\u0080\u0099s duct tape and wire nuts kept theelectrophysiology laboratory running, and Raul Suarez who was a constant source ofproductive agitation. Special thanks also go to the Medical Research Council of Canadawho\u00E2\u0080\u0099s support made this project possible. Finally, I would like to thank my firstundergraduate biology professor, James Kanz, (Texas A & M University) who taught me thatbiology was not a science but a philosophy.xiiCHAPTER 1: AMMAL ANOXIA TOLERANCE - AN INTRODUCTIONPrefaceThis chapter is intended to give the reader an overview and orientation to the subject ofanoxia in vertebrates, and how, based on this information, the focus of this thesis evolved.Parts of this chapter were excerpted form C. J. Doll (In Surviving Hypoxia, CRC Press, pgs.389-400, 1993).A Species OverviewThe graceful glide of a marine mammal beneath the ocean surface is a reflection of theextreme evolutionary morphological change which has occurred to this group of mammals.Though not sharing a common ancestor with whales, seals have also forgone a terrestrialmode of life in exchange for the sea. Whales and dolphins abandoned terrestrial life about45 m.y.a. (Gingench and Russell, 1991) compared to seals who abandoned the land about 23m.y.a. (Berta et al., 1989). Both groups have undergone considerable morphologicalchange; yet, with respect to cellular adaptations to anoxia these animals are remarkablysimilar to nondiving mammals of today (Castellini et al., 1980).Both terrestrial and marine mammals have tissues such as skeletal muscle which aretolerant to both anoxia and ischemia, however, they both also possess oxygen sensitivetissues such as the brain. Mammalian skeletal muscle is known to survive ischemia forseveral hours (Beyersdorf et al., 1991) while the brain function is noted to fail within aminute for a similar insult (Hansen, 1985). In terms of asphyxia duration, the championvertebrate homeotherm is the elephant seals reported to dive to depths of 1.5 Km. (DeLong,1991) and for times up to 120 mm. (Hindell et a!., 1991). Yet, no diving mammal or birdhas been recorded with a zero blood P02 (Kooyman, 1989). Rather, these animals rely onmechanisms which conserve and carry more oxygen to their tissues. These adaptationsinclude: (i) a large body size to minimize oxygen consumption ; (ii) the diving reflex toconserve oxygen for critical tissues; (iii) increased hematocrit/volume of blood, as well asmore hemoglobinlcell to increase blood oxygen stores; (iv) large stores of myoglobin in the1skeletal muscle to increase cellular oxygen stores; and (v) the release of oxygenated redblood cells from the spleen during a dive to provide additional oxygen to critical areas, [forreviews see (Eisner and Gooden, 1983; Hochachka, 1980; Kooyman, 1989)].That brain is considered the most sensitive tissue to hypoxia and anoxia in the vertebratehas been the topic of investigation for a considerable time (Boyle, 1670). Even thoughmarine mammals live the majority of their lives in an anoxic environment experiencinghypoxia daily, their central nervous system (CNS) appears to be only slightly better adaptedto living without oxygen as compared to the canine brain. Studies have shown CNS failureoccurring at a P02 of 10 torr (Kerem and Eisner, 1973b) for the seal brain compared to 14ton for the canine brain (Kerem and Eisner, 1973a). This difference is not very significantwhen considering: (i) the scaling effect of brain size vs. metabolic rate which gives the largerseal brain a lower metabolism (Mink eta!., 1981) and (ii) the increased capillary density ofthe seal brain allowing a greater extraction of oxygen from the blood (Eisner and Gooden,1983). Through millions of years of selective pressure for hypoxia tolerance, marinemammals do not appear to be significantly better adapted to cerebral anoxia tolerance than atypical terrestrial counter part. This observation suggests that the CNS of homeotherms mayoperate under certain limitations which disallow anoxic tolerance.The CNS in most vertebrates is highly sensitive to oxygen deprivation. The only notableexceptions are the crucian carp (Carassius carassius), the goldfish (Carassius auratus), anda few reptiles of which the most notable is the painted turtle (Chrysemys picta) (Ultsch,1985). This species can survive 6 months in anoxic water at 3 \u00C2\u00B0C (Ultsch, 1985) and 48hours at 25 \u00C2\u00B0C (Musacchia, 1959). In the fall, C. picta submerges into the anoxic sedimentsof ponds and river bottoms and overwinters until spring (Ultsch, 1989). The turtle\u00E2\u0080\u0099s anoxicresistance has presumably evolved as a survival mechanism to overwintering in anoxic waterand mud. This hypothesis is supported by the observation that long hibemators (northernspecies) exhibit greater tolerances to anoxia compared to southern species (Ultsch, 1985).2Turtle Whole Body Anoxic AdaptationsPerhaps the most striking adaptation of the turtle is its ability to control whole bodymetabolism. As indicated above, the turtle is an excellent oxygen conformer suggesting thatit has the ability to regulate metabolism during hypoxia. Jackson (1968) demonstrated thatthe turtle was able to depress its metabolism (heat production) during anoxia to 1/5 itsnormoxic rate within a few hours of submergence (Jackson, 1968). Regulation of wholebody metabolism aids the survival of the turtle in several ways: i) it allows the stores of onboard fuel (glucose and glycogen) to last longer; ii) it lessens the cellular energy (ATP)demand allowing glycolytic machinery to match cellular demand; and iii) it decreases thebuild-up of harmful toxic products.With decreasing metabolism, the second problem facing the anoxic organism is supplyingcritical tissues like the brain with enough substrate (glucose). Adaptations in this categorycan be broadly defined into two areas. First, the shunting of blood to critical areas (divingreflex), and second, the regulation of substrate utilization.Blood flow studies have indicated that the turtle (C. scripta) increases blood flow up to250% (compared to normoxia) to the brain during an anoxic insult while other organs such asliver, intestines, pancreas, and kidneys all receive large (50 - 100%) reductions (Bickler,1992b; Daves, 1989). This shunting to the brain clearly demonstrates the importance ofmaintaining a large glucose supply to this critical organ.Another adaptation in conjunction with blood shunting is the ability to regulate blood[glucose] through the mobilization of liver glycogen. Studies have shown that turtles are ableto increase blood [glucose] by up to 11 fold (compared to control) during anoxia (Clark andMiller, 1973; Daw et al., 1967; Keiver eta!., 1992; Penney, 1974). This increase is reflectedby a dramatic drop in liver glycogen (Clark and Rothman, 1987; Penney, 1974). The glucoserise can be prevented by the administration of propranolol (a 13-adrenergic receptorantagonist) supporting the role of hepatic mobilization of glycogen in this increase (Keiver3and Hochachka, 1991). Turtles also appear to have large reserves of liver glycogencompared to non-anoxia tolerant species (Hochachka and Somero, 1984).One of the most detrimental effects of anoxia on the turtle is the accumulation of harmfulend products. Numerous studies have documented a decrease in blood pH and subsequentrise in [lactate] (up to 200 mM) occurring during anaerobiosis in the turtle (Gatten, 1981;Herbert and Jackson, 1985; Robin eta!., 1981; Ultsch and Jackson, 1982). Additionally,studies have shown that if blood pH drops below 1 pH unit of resting, the animal is not likelyto fully recover (Herbert and Jackson, 1985). Long term anoxic survival necessitates theneed for the accumulation of lactate, but the organism can only tolerate limited amounts ofeither lactate or the accompanying drop in pH. One solution as discussed above is tominimize metabolism, and thus minimize the accumulation of these end products. In part,the turtle can remove some protons and lactate through the urine (Ultsch, 1989). However,this is a wasteful strategy since upon recovery, the turtle could no longer utilize the largeremaining energy reserve of lactate for re-establishment of glycogen stores or furtheroxidation. Another mechanism would be to increase buffering capacity of both blood andintracellular fluid, but the evidence does not support enhanced buffering capacity in eithercompartment (Ultsch, 1989; Ultsch and Jackson, 1982). What the turtle does appear toutilize is the mobilization of counter positive strong ions (Ca2and Mg2)to ionicallybalance the formation of lactate anions (Herbert and Jackson,1985; Jackson, 1982b). Thisbalance depresses the drop in pH of the blood significantly (Herbert and Jackson, 1985). Theconcentrations of both of these ions increase greatly with submergence, with calciumreaching 100 mequiv/1 (3X control) and magnesium about 30 mequiv/l (2X control).(Jackson,1982a) The close correlation in the rise of these two ions with lactate suggests that this is acontrolled response (Ultsch, 1989). The unphysiologicially high blood [Ca2]appears to beachievable through the formation ofCa+2lactate complexes which are believed to bind 2/3of the Q2formed (Jackson, 1982b). High blood [Ca2]has been observed in severalreptiles. The highest recorded concentrations are in the ovulating snake (Thamnophis4sauritus) which can achieve a 90 mM blood [Ca2](Dessauer and Fox, 1959). The highblood Ca2 solubility is achieved through the use of phosphoprotein (Dessauer, 1970). Thus,the turtle may use a similar mechanisms along with lactate complexes to aid in dissolvingblood Ca2 (Jackson, 1982a).Tissue Hypoxic ResponseBoth animals and individual organs can be classified broadly into two groups with respectto their response to varying oxygen tension (P2)(oxygen conformers or oxygenregulators). Oxygen conformers decrease their oxygen consumption (\u00E2\u0080\u0098O2) as the oxygenpartial pressure (P2)is decreased where as oxygen regulators retain a relatively constantV02 with decreasing o2 (Prosser, 1986; Schmidt-Nielsen, 1979; Fig. 1). Mammalianskeletal muscle which survives long term anoxia and ischemia is classified as an oxygenconformer (Whalen et at., 1973), but the mammalian brain (including the seal brain), whichhas little or no tolerance to oxygen deprivation is a stringent oxygen regulator (Jones andTiystman, 1984; Kerem and Eisner, 1973a; Kerem and Eisner, 1973b; Kitner et al., 1984).Interestingly, anoxia tolerant invertebrates (Herreid, 1980; Mangum and Van Winkle,1973)and vertebrates (Fry and Hart, 1948; Glass et at., 1983; Jackson and Schmidt-Neilsen,1966; Processer et at., 1957) tend to exhibit oxygen conformity suggesting that anoxiatolerance, in part, is related to the ability to regulate metabolic rate (Hochachka, 1980;Hochachka, 1986). Regulation of brain metabolic rate will be discussed in further detail inChapters 4 and 5.Energy And AnaerobiosisUnder normal aerobic circumstances, brain tissue consumes exogenous glucose as anenergy supply (Hawkins, 1985; Macmillan and SiesjO, 1972), and the turtle brain appears tobe no exception (P\u00C3\u00ABrez-Pinz\u00C3\u00B3n et al., 1992b; Robin et at., 1979). Aerobically catabolizedglucose (per mole) yields approximately 36 moles of ATP:Glucose+36P+36ADP+6O2 6C02+36ATP\u00C3\u00B744H050_____.>P02Figure 1. Theoretical response of an oxygen conformer and an oxygen regulator to varyingoxygen partial pressures.I oxygen regulatorsoxygen conformers6GlucoseFigure 2. Fermentation pathways identified in vertebrates and invertebrates. Modified fromHochachka and Guppy, 1987.GAP\u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094- NAD+\u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094\u00E2\u0080\u0094 _NADHcI 1,3DPGIIIIIIIIINADH\u00E2\u0080\u0094 NADNAD+ _ \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u0094 -NADHNADH(pIII\u00E2\u0080\u0098II\u00E2\u0080\u0099NADVlactate2ATPtauropine2ATPoctopine2ATPMalate///\u00E2\u0080\u0098\b4 Ethanol acetate2ATP 4ATPSuccinate4 ATPpropionate6 ATP2ATPalanopinealanine 2 ATP2ATP7However, when a tissue becomes anoxic, it no longer has oxygen as an electron acceptor andtherefore can no longer use the Krebs cycle and electron transport system and still maintainmitochondria redox balance (Hochachka, 1980). However, glycolysis can still functionbecause cytosolic redox balance can be maintained through the lactate dehydrogenase (LDH)reaction:NAD NADHPyruvate -4 LactateLactate DehydrogenaseHowever, various invertebrates (helminths and bivalves being particularly inventive) havedeveloped several different fermentative reactions allowing for increased yields of ATP (Fig.2) (Hochachka and Guppy, 1987). Surprisingly, the only two pathways which appear to beutilized by vertebrates to any extent is the ethanol pathway (found in carp) (Shoubridge andHochachka, 1981) and the lactate pathway (found in all vertebrates), both yielding a net of 2ATP (Fig. 2). Turtles do not appear to use any of the other fermentative pathways besideslactate (Buck and Hochachka, In Press; Hochachka eta!., 1975; Robin, et al., 1979). Theadvantage of these fermentative pathways is that despite oxygen lack the cytosol can remainin redox balance, a small amount of ATP can be produced (Fig. 2), and relatively nontoxicend products are formed. However, the tissue has gone from a condition of producing 36ATP/glucose to 2 ATP/glucose (in the case of glycogen, there are 3 ATP/glucosyl unit):ADP ATPGlycogen G1P G6P (into the glycolytic pathway)PhosphorylaseThe turtle brain appears to use glycogen only as a last resort (Clark and Miller, 1973). Theloss of significant ATP production suggests two possibilities for the turtle brain if ATPsupply is to meet anoxic demands: (i) glycolytic flux must increase (Pasteur effect), (ii) theATP tissue demand must be decreased (metabolic arrest).Pasteur EffectThe Pasteur effect was named after its discoverer (Louis Pasteur) who first noticed aninhibition of carbohydrate (glucose) consumption when oxygen concentrations were high and8an increase in glucose consumption when oxygen concentrations were low (Pasteur, 1861).Given that the ATP production from 1 mole of glucose under anoxic conditions is 1/18 thatof normoxic conditions, one would expect that glycolysis would be increased 18 fold if ATPsupply is to meet demand (12 fold if glycogen were used). However, this does not appear tobe the case with most tissues (specifics for brain tissue will be discussed in chapter 3). Oneof the largest observed Pasteur effects (15 fold) is found in bull sperm (Hammerstedt andLardy, 1983). There are two basic reasons that a tissue may not exhibit a full Pasteur effectand yet still maintain energy (ATP) balance. First, the tissue may depress its metabolic rateas it enters into the anoxic state, thus ATP demand has diminished and a lower glycolyticflux will meet cellular ATP consumption. Secondly, the assumption that all of the glucoseunder normoxic conditions is fully oxidized may not be correct (Lynch and Paul, 1983)which may result in a substantial overestimation of energy utilization.Insights into Brain Anoxia ToleranceBased on the above information three mechanisms can be hypothesized to play a centralrole in turtle neuronal anoxia tolerance: (i) a Pasteur effect; (ii) a low normoxic metabolicrate; (iii) further down regulation of metabolism with anoxia (metabolic arrest).The presence of a Pasteur effect would appear to be crucial if energy balance is to remainconstant. Maintenance of energy balance is most critical in the transition phase where thecell must switch from a highly aerobic state to anaerobiosis. This rapid transition causes anonequilibrium with regards to cellular function and energy balance (especially with respectto pharmacological anoxia, see Appendix B). Thus, the Pasteur effect would serve as atemporary buffer allowing the neuron to make necessary metabolic adjustments. Asdiscussed above, if metabolism can be lowered significantly (12 - 18 fold depending on theanoxic substrate), then the presence of a Pasteur effect may be transitory. The initial Pasteureffect may be subsequently down regulated to allow normal or reversed glycolytic flux (areversed Pasteur effect).9With regards to a reduced metabolic rate, several critical points can be made for anoxicsurvival: (i) a low ATP demand would aid the cell in maintaining cellular energy supply for agiven glycolytic flux; (ii) a low metabolism would decrease the build-up of toxic endproductssuch as lactate anions and protons; (iii) a low ATP demand would allow a given amount ofon board substrate (glucose or glycogen) to last longer; and (iv) since the turtle brain relies onexogenous glucose for its substrate (Clark and Miller, 1973; P\u00C3\u00A9rez-Pinz\u00C3\u00B3n, et al., 1992b), alower metabolic demand would better enable membrane glucose transporters to meet theglycolytic demand. Membrane glucose transport rate has been proposed as a rate limitingstep in the catabolism of glucose in the rat brain (Furler et a!., 1991).Anoxic Membrane Coupled FunctionThe mechanisms proposed above all share one common goal: the coupling of energyproduction to meet energy demand. If supply can meet demand, then an energy deficit willbe avoided and normal neuronal function can continue. However, not all systems maintainenergy balance during anoxia. Both Artemia embryos and locusts are known to lose energybalance during anoxia (Hand and Gniager, 1988; Wegener, 1987; Wegener et a!., 1987), andthus, completely shut down their systems. The turtle, on the other hand, maintains somedegree of alertness during anoxia (Ultsch, 1989) and thus, must maintain some degree ofbrain function.The neuron is an electrical cell. Thus, the maintenance of ion gradients is crucial ifelectrical function is to continue throughout an anoxic bout. The disadvantage tomaintenance of ion gradients is the energy cost. It has been estimated that in the normalconscious mammalian brain, maintenance of ion gradients may consume as much as 50-60% of metabolism (Hawkins, 1985). Ion homeostasis is maintained through the use of ATPutilizing ion pumps (Na+-K+-ATPase and the Ca2 -ATPase). One possible scenario is thedown regulation of ion channels during anoxia \u00E2\u0080\u0098channel arrest\u00E2\u0080\u0099 (Hochachka, 1986;Hochachka, 1987; Hochachka and Guppy, 1987; Lutz et a!., 1985). Down regulation of ionchannels (leakage or voltage-gated channels) would indirectly conserve energy through the10reduction of ion pumping. This mechanism of energy reduction would allow anoxic brainfunction and still conserve energy. Such a channel arrest hypothesis will be explored inChapter 4.Thesis OverviewThe extreme anoxia tolerance exhibited by C. picta and the high sensitivity of themammalian brain to 2 depletion led to the use of the turtle and rat as comparative models.The lack of literature in regards to anoxia and hypoxia in the cortex and the easy dissection ofthese tissues in both species suggested an ideal model for comparative studies. Earlierstudies (Connors and Kreigstein, 1986) had identified similar neurons (pyramidal andstellate) in both the turtle and rat cortex again suggesting the appropriateness of this tissue fora comparative study. As a result, this thesis will focus on turtle and rat cortical tissue inorder to understand some of the mechanisms which Cpicta utilizes to cope with both energyand oxygen depletion.Chapter 2 examines turtle and rat neuronal function during anoxia. Prior to thepublication of this study, nothing was known with respect to how individual turtle neuronswere responding to anoxia. This study was made possible through the use of intracellularrecording techniques (see Appendix A) which allows the measurement of individual neuronalparameters such as membrane potential, cell resistance, and action potential parameters.Chapter 3 focuses on the brain biochemistry (energy balance and glycolytic enzymes) andthe metabolism of the turtle cortical slice preparation. This chapter will examine whether thecortical slice possesses a low resting metabolic rate and whether this metabolic rate is furtherreduced with anoxia.Chapter 4 will investigate the role of \u00E2\u0080\u0098channel arrest\u00E2\u0080\u0099 in the turtle slice again usingelectrophysiological techniques, intracellular recording and patch clamping techniques (seeAppendix A) to measure whether the turtle pyramidal neuron exhibits a lower resting ionleakage compared to the rat counter part and whether this leakage is further down regulatedwith anoxia as a possible explanation for metabolic arrest.11The thesis concludes in Chapter 5. This chapter presents two hypotheses. First, a unifiedtheory is presented which explains how the turtle neuron maintains membrane integrityduring anoxia in contrast to the rat neuron. This theory, in part, is based on the observeddifferences in brain metabolic rates of the turtle and the rat. A second hypothesis is thenpresented which explains why there is a metabolic difference between these two species brainpreparations.12CHAPTER 2: EFFECTS OF ANOXIA AND PHARMACOLOGICAL ISCHEMIA ONTURTLE AND RAT CORTICAL NEURONSPrefaceThis chapter is excerpted (in part) from a paper published by C. J. Doll, P. W.Hochachka, and P. B. Reiner (Am. J. Physiol. 260: R747-R755, 1991). This chapterexamines how the cortical pyramidal cell responds to anoxia and similar insults usingintracellular recording techniques. Prior to the publication of this paper, no intracellularrecording had been done on the anoxic turtle brain. The results obtained in this study werethe foundation for the chapters to follow.IntroductionMammalian brains are very sensitive to lack of 02. When blood P02 reaches criticallevels, brain function ceases (Hansen, 1985). Of particular interest is the mechanism bywhich this rapid failure occurs because of the brain\u00E2\u0080\u0099s importance to survival. Due to theconsiderable amount of literature in this field, this chapter will begin with a review of someof the relevant observations that have been made in this area. This broad overview will aid inthe interpretation of the results and discussion to follow.Mammalian CNS ResponseWhen mammalian brain tissue becomes ischemic!anoxic, it typically responds with alarge Pasteur effect (Drewes and Gilboe, 1973; Kauppinen and Nichols, 1986; Lowry eta!.,1964). Unfortunately, energy supply does not appear to meet demand resulting in [ATP]declining to 25% of control within the first minute in the rat brain (Lowry, et al., 1964;Ridge, 1972). The initial response of the mammalian brain appears to be a flat EEG followedby changes in the extracellular environment beginning with a slight increase in {K+]0followed by a massive decrease of [Na+Jo and [Ca+2]oand increases in [K+]0 (Hansen,1978; Hansen, 1982; Hansen, 1985; Sick eta!., 1987; Suzuki eta!., 1985). These eventsoccur within the first minute of the ischemic insult and cause brain dysfunction ultimatelyleading to the death of the neuron. These initial events resemble those observed in spreading13depression which is brought about when the brain receives a severe trauma (Hansen andZeuthen, 1981; Kraig and Nicholson, 1978). Accompanying these changes in theintracellular and extracellular environment is the sudden loss of neuronal membraneresistance (Rm) and membrane potential (M.P.) (Hansen, 1982).With regards to anoxialhypoxia, there has been considerable work done with thehippocampus. Interestingly, anoxia/hypoxia appears to cause similar changes in the neuronthat ischemia does, but over a much longer time frame (Fujiwara eta!., 1987; Higashi et at.,1988). The slower time frame to complete membrane depolarization of the neuron duringthis insult has allowed the study of individual events taking part in the cell depolarization.Currently clamp methods have revealed 3 basic conductance and current changes when ahippocampal cell is held at -70 mV with KC1 electrodes. An initial inward current flowcorresponding to a slight or no depolarization of the neuron (phase 1) followed by a muchlarger outward flow of current and increase in membrane conductance (Gm) (phase 2)corresponding to a hyperpolarization, and (phase 3) a variable net inward flow correspondingto the final depolarization of the membrane to a 0 mV resting potential (Krnjevic andLeblond, 1989; Leblond and Krnjevic, 1989). Efforts to try and unravel this complexity ofcurrents is still being sought. Current evidence suggests that in the CA 1 region of thehippocampus, the second phase (the outward flow corresponding to a hyperpolarization), ismost likely carried by Ca2 dependent K current (IAHp) due to its blockade by carbacholand CsCl (Cummins et al., 1991; Krnjevic, 1993; Krnjevic and Xu, 1990). However,whether the1AHp is fully responsible for the hyperpolanzation in all neurons is stillspeculative. Recently, glibenclamide was shown to prevent the anoxic hyperpolarization inthe CA3 region. However, this blockage of the hypoxic hyperpolarization by glibenclamideappears to be an indirect action related to the suppression of hypoxic glutamate release (BenAri, 1990). Recent unpublished data suggests that talbutamide but not glibenclamide maysuppress the outward current in CA 1 (J. M. Godfraind and K. Krnjevic, unpublished). Thus,no definitive conclusions can be made at this time.14Phases one and three correspond to a net inward flow of current leading to neuronaldepolarization. Phase 2 can also be identified with this inward flow if the outward currentflow (IK+) is blocked with CsCl or carbachol (Krnjevic and Leblond, 1989). The reversalpotential of this current is around -40 - 0 mV suggesting the contribution of a nonselectivecation channel (Partridge and Swandulla, 1988; Yellen, 1982). However, phase 3 has beenassociated with complex events such as neurotoxicity (Benveniste et at., 1984; Choi, 1988a;Clark and Rothman, 1987) and uncoupling of the Na-K-ATPase pump (Fujiwara, et al.,1987; Krnjevic, 1993) making this phase exceptionally complex and far from understood atthis timeTurtle CNS Anoxic ResponseIn marked contrast to the rat brain, the turtle brain responds to anoxia with only a slightincrease in [K+10with concentrations rising less than 2.5 fold over a 48 hour period in vivo(Lutz, et al., 1985; Sick et at., 1982) In addition, there is evidence for an initial Pasteureffect possible decreasing with time (Kelly and Storey, 1988). With ischemia(Sick et a!.,1985) or the addition of iodoacetic acid (P\u00C3\u00A9rez-PinzOn, et al., 1992b) (glycolytic inhibitor),the turtle brain responds with an increase in extracellular [K+]0 similar to the anoxic/hypoxicrat brain. Most importantly, the turtle brain retains the ability to maintain [ATPI. Bothcerebellum (P\u00C3\u00A9rez-Pinz\u00C3\u00B3n, et al., 1992b), cortex (Bickler, 1992b) and whole brain (Lutz eta!., 1984) in vivo have been shown to maintain [ATPJ during anoxic insults.The goals of this chapter are two fold. First, to examine the response of the rat corticalpyramidal neurons to the anoxic / ischemic insult. Secondly, to examine intracellularly, thechanges which occur to the turtle pyramidal neuron during anoxia and ischemia.MethodsTissue PreparationYoung male Wistar rats 50-100 g were anesthetized with halothane, decapitated, thebrain rapidly dissected free and immersed in precooled oxygenated artificial cerebrospinalfluid (aCSF). After a few minutes precooling, the brain was bisected along the midline. A15block containing frontal - parietal cortex was dissected free, glued with cyanoacrylate to amounting block and sliced (400 pm thickness) on a vibratome. Slices were stored at roomtemperature 22 \u00C2\u00B0C in a holding chamber for at least 1 hr. until their use in the recordingchamber at 25 \u00C2\u00B0C or 35\u00C2\u00B0C. The recording chamber was a modification of a previous design(Haas et a!., 1979) (refer to appendix A for more detail) in which slices were continuouslysuperfused with aCSF at a flow rate of 1.5 to 2.0 mI/mm. Turtles (Chrysemys picta) rangedin weight from 250 to 600g. The dissection has been previously discussed in (Connors andKreigstein, 1986). In brief, turtles were cold anesthetized prior to decapitation. The brainwas dissected free and immediately placed in precooled aCSF. After cooling ( 2 mm.) theturtle brain was bisected along the midline. The cortex (whole) was then dissected free in anintact sheet. The cortical sheet was then used whole or divided in half depending on the size.Subsequent storage of slices was identical to that discussed above for the rat brain.Data AcquisitionFor a more general discussion of intracellular recording techniques, the reader is referredto Appendix A and B). Intracellular recordings were carried out using 1.2 mm o.d.micropipettes filled with 2 M-KCI with resistance ranging from 40 - 90 MQ connected to anAxoclamp 2A amplifier. Data were acquired using the Pclamp suite of programs and anAxolab 1100 interface, which also served to generate current commands. Data were alsoindependently digitized at 49 kHz and stored on videotape for off-line analysis.Pyramidal neurons in the turtle were identified by location, action potential size andduration, as well as input resistance and time constant as previously discussed (Connors andKreigstein, 1986). Pyramidal neurons in the rat were identified by location as well as actionpotential duration and size as discussed elsewhere (McCormick et at., 1985). Criteria for ahealthy neuron included positive going action potentials, a minimum of 40 MQ resistance forrat neurons and 100 MQ resistance for turtle neurons, and stable membrane potential for 15minutes.16Fluid CompositionThe rat aCSF consisted of NaCI 126 mM, KC1 2.5 mM, MgCl 1.2 mM, CaCI2 2.5 mM,NaH2PO4 1.2 mM, Glucose 2 mM, NaHCO3 25 mM and phenol red .03 mM as a pHindicator. The aCSF for the turtle was a modification of (Connors and Kreigstein, 1986) andconsisted of NaC1 96.5 mM, KC1 2.6 mM, CaC12 2.5 mM, MgC12 2.0 mM, NaH2PO4 2.0mM, Glucose 10 mM, NaHCO3 26.5 mM, and phenol red .03 mM as a pH indicator. FinalpH of both solutions was 7.4 when saturated with 95% 02/5% C02. To mimic anoxia, thesolution was switched to a presaturated aCSF solution of 95%N2/5%C02. Forpharmacological anoxia, the aCSF solution was the same as anoxia and contained 1 mMNaCN. For pharmacological ischemia, the solution was the same as pharmacological anoxiaand also contained 10 mM iodoacetic acid (IAA) titrated to a pH of 7.4 with concentratedNaOH. For the ouabain and IAA experiments the aCSF contained 100 J4M ouabain and 10mM iodoacetic acid respectively, equilibrated with 95% 0215%C02ResultsThe results for various pharmacological treatments are illustrated in Fig. 3. This figurerepresents the mean times for cell survivability which for the purpose of this paper is the timeit takes for a cell to depolarize from its resting membrane potential toO mV once a drug wasapplied. These measurements are corrected for the lag time for the drug to reach the slice andequilibrate in the slice chamber ( 30 sec). All turtle cells were recorded at 25\u00C2\u00B0C due toproblems associated with survivability of ectothermic cells at unphysiologicially hightemperatures. All treatments were significantly different (P 0.05; Newman-Keuls multiplecomparison test) from each other except IAA vs. ouabain. All values within a group weresignificantly different (P 0.05; independent t test), except rat pharmacological anoxia 35 vs.25\u00C2\u00B0C and rat vs. turtle pharmacological ischemia 25 \u00C2\u00B0C. Rat neurons were recorded at 25as a comparison to the turtle. In addition, some insults were repeated at 35\u00C2\u00B0C in the ratin order to demonstrate the repeatability of the low temperature results at a morephysiological temperature. The average whole cell resistance were as follows: turtle 25\u00C2\u00B0C17A\u00E2\u0080\u0098120.60B.5.\u00E2\u0080\u0094. 210Figure 3. Time to depolarization for turtle and rat pyramidal cortical neurons in response tovarious pharmacological treatments: pharmacological anoxia (NaCN), anoxia (N2),iodoacetate (IAA), and ouabain (0), (A) and pharmacological ischemia, (B). All treatmentshave an n of 6 except rat 35 \u00C2\u00B0C (n = 3), turtle 25\u00C2\u00B0C NaCN (n = 5), and turtle N2 (n = 5).Data are presented \u00C2\u00B1 SE bars. Treatments without SE bars do not represent survival times.Note different time scales on ordinate.90Turtle 25\u00C2\u00B0CQ Rat25\u00C2\u00B0CRat35\u00C2\u00B0C30NaCN N2 IAA 0TPharmacological Ischemia18A2OmVC100 mSBFigure 4. Effect of 180 mins. of anoxia on a continuously impaled turtle cortical pyramidalcell. Cell initially was impaled and recorded from using oxygenated aCSF (A and B) andthen was subjected to 180 mins.. of anoxia while measurements were repeated (C and D).Membrane potential in both cases was -65 mV. Current steps were from -0.5 nA by 50 pA inA and C. A stimulating pulse of 0.5 nA was used in B and D.D19Table 1. Etlct of anoxia and pharmacological anoxia on membrane potential and actionpotentials of turtle neuronsTreatment Parameter ii Control 180 mmMembrane parameterN2 MP, mV 4 -67.8 \u00C2\u00B1 3.8 -70.0 \u00C2\u00B1 6.4NaCN MP, mV 4 -74.0 \u00C2\u00B1 4.2 -74.5 \u00C2\u00B1 3.6Action potential parametersN2 Spike Rise, mV/mS 4 92.6 \u00C2\u00B1 4.7 91.9 \u00C2\u00B1 5.5Spike fall, mV/mS 4 28.4 \u00C2\u00B1 0.9 33.9 \u00C2\u00B1 3.7Threshold, mV 4 -36.0 \u00C2\u00B1 1.5 -43.0 \u00C2\u00B1 2.5Spike amplitude, mV 4 62.7 \u00C2\u00B1 1.1 63.4 \u00C2\u00B1 1.6Spike width, mS 4 4.8 \u00C2\u00B1 0.4 4.7 \u00C2\u00B1 0.4Values are means \u00C2\u00B1 SE; n, no. of turtles at 25 \u00C2\u00B0C. Values given are an average from aset of continuously impaled cells. There were no statistical differences between any datasets (P> 0.05; paired t test). N2 , anoxia; NaCN pharmacological anoxia; MP, membranepotential.200.4nA2.0mmCAPh. Ischemia-100BAPh. Ischemia0mVFigure 5. Response of rat and turtle pyramidal cortical neurons to pharmacological ischemia(Ph. Ischemia). Treatment was applied to a rat cortical cell at 25(A) and 35\u00C2\u00B0C (B).Treatment was applied to a turtle cortical pyramidal neuron at 25 \u00C2\u00B0C and iodoacetate wasapplied after cell had been subjected to pharmacological anoxia for 180 mins. at 25 \u00C2\u00B0C (D).Time scale in (A) is representative for (A - D)A B0mV-J-100fr 0mVPh. Ischemia -100-J-100Ph. Ischemia21ABC11.1 -10mV_i -1000mV_J-ioo0.4nA____2.0 mmFigure 6. Effect of perfusing solutions of anoxia and pharmacological anoxia on ratpyramidal neurons. In (A), the perfusing solution was anoxic (25 \u00C2\u00B0C) compared topharmacological anoxia in (B) and pharmacological anoxia at 35 \u00C2\u00B0C in (C).0mY-1oo22AB4 mmIAA0mV-1000mV-100Figure 7. Effect of perfusing iodoacetate on turtle and rat pyramidal cells (rat, A; turtle, B) at25\u00C2\u00B0C.IAA20 mm4234 mm0mV-100Figure 8. Effect of perfusing ouabain on rat pyramidal neurons (100 jsM at 25 OC).ill 1jA024151\u00C2\u00B1 12MQ(n=23),rat25\u00C2\u00B0C86\u00C2\u00B15MQ(n=30),andrat35\u00C2\u00B0C66\u00C2\u00B1 13MQ(n=6).These values are similar to those reported for both turtle and rat using whole cell patchclamping (Blanton et al., 1989).The turtle neurons maintained a healthy whole cell resistance throughout the anoxic andpharmacological anoxic insults, and were able to fire action potentials for the duration of theexperiment (180 minutes) (Fig. 4). Individual turtle cells impaled under normoxic conditionswere held up to five hours under anoxia. In addition, slices were held in a holding chamberbathed with nitrogen equilibrated aCSF for up to 18 hours. These cells qualitatively showedno effect of the treatment. In order to quantitatively study the effect of anoxia on certaincellular parameters, turtle neurons were impaled under control conditions (oxygenated aCSF)and held for 180 minutes under anoxic conditions (Table 1). Cells were tested for changes inresting membrane potential (M.P.), action potential rate of rise and fall, threshold, spikeamplitude and width following various treatments. Spike amplitudes were measured fromthreshold, and the rates of spike rise and fall were measured at the maximum slopes. Theresults indicate that there were no significant changes in any of the measured parameters.When the turtle cells were exposed to pharmacological ischemia, the neurons quicklydepolarized (4.6 \u00C2\u00B1 1.1 mm.) (Fig. 3). This experiment was carried out under oxygenatedconditions (Fig. Sc) as well as anoxic conditions (Fig. 5d). In both instances, the turtlecortical neurons exhibited a slight hyperpolarization (4.6 \u00C2\u00B1 1.8 mV) preceded by orconcurrent with a loss in whole cell input resistance followed by a rapid depolarization to azero membrane potential.Rat cortical neurons responded to pharmacological ischemia similarly to the turtle, alsoexhibiting a slight hyperpolarization (8.0 \u00C2\u00B1 2.7 mV, 25\u00C2\u00B0C; 1.0 \u00C2\u00B1 0.6 mV, 35\u00C2\u00B0C) preceded byor concurrent with a loss in apparent whole cell input resistance leading again to a rapiddepolarization (3.1 \u00C2\u00B1 0.5 mm., 25\u00C2\u00B0C; 1.8 \u00C2\u00B1 0.2 mm., 35 \u00C2\u00B0C) (Figs. 3 & Sa,b). There was asignificant temperature effect for both the loss of membrane potential (Qio. 0.56) andhyperpolanzation (Q1.0.13) (P 0.05; independent t-test).25The results of anoxia and pharmacological anoxia show that in both instances the ratresponded with slight hyperpolarization (anoxia 3.2 \u00C2\u00B1 0.1 mV, 25 OC; pharmacologicalanoxia 5.5 \u00C2\u00B1 1.7 mV, 25 OC; 1.3 \u00C2\u00B1 0.3 mV, 35 0C) preceded by or concurrent with a loss inwhole cell input resistance and a slow and gradual depolarization (anoxia 41.8 \u00C2\u00B1 6.6 mm., 25\u00C2\u00B0C; pharmacological anoxia 25.8 \u00C2\u00B1 12.6 mm., 25 \u00C2\u00B0C, 9.3 \u00C2\u00B1 1.8 mm., 35\u00C2\u00B0C) (Figs. 3 & 6).Again, there was a significant effect of temperature on both the loss of the membranepotential (Q 10,0.36) and the hyperpolarization (Qio, 0.24) for pharmacological anoxia (P0.05; independent t-test).The addition of IAA by itself was applied in order to test the cell\u00E2\u0080\u0099s ability to rely on othersources of substrate besides glucose from intracellular reserves. Since IAA blocks theglycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase, the cell can no longer utilizeglucose and glycogen as a glycolytic substrate. Under these conditions, rat cells depolarizedtoO mV in an average of 6.5 \u00C2\u00B1 0.8 mm. compared to 53.5 \u00C2\u00B1 4.6 mm. for turtle cells (Figs. 3& 7). This treatment also caused a hyperpolarization (3.0 \u00C2\u00B1 1.2 mV, turtle; 4.0 \u00C2\u00B1 1.7 mV,rat).To test whether pharmacological ischemia or anoxia may simply reflect an inability of theNa-K-ATPase to function, ouabain (100 jsM) was perfused onto the rat cortical slice at 25\u00C2\u00B0C. The average time to depolarization was 8.7 \u00C2\u00B1 1.1 mm. (Fig. 8). The cells never showeda hyperpolarization, and there was no loss in membrane resistance until there was a loss inmembrane potential. The time to depolarization was faster than anoxia or pharmacologicalanoxia, but less than total metabolic inhibitionAdditional StatisticsIn addition to testing the significance of these pharmacological treatments to each other,each treatment was individually tested (Pearson Correlation Matrix) to see if there were anycorrelations within treatments corresponding to membrane potential, whole cell resistance,degree of hyperpolarization, and the time to depolarization. Although there were a fewcorrelations, there were no consistent trends throughout all of the treatments. The most26consistent correlation was with the membrane potential versus the time to depolarization withthe treatments of nitrogen and phaimacological anoxia in the rat, and with IAA in the turtle(P 0.05).DiscussionThe evidence presented in this chapter strongly supports the supposition that themechanism responsible for the rapid failure of the brain during anoxia (rat brain) andischemia (rat and turtle brain) involves the breakdown of ion gradients across the neuron.This observation is not surprising in view of the evidence discussed earlier. However, thequestion further explored here is why cortical neurons lose apparent input resistance underthese treatments.One hypothesis for the loss of membrane resistance observed here and elsewhereconcerns the loss of metabolism, and hence, the energy state of the cell (Hansen, 1982;Hochachka, 1986; Lipton and Whittingham, 1982). This hypothesis centers around the ideathat high energy phosphates are directly or indirectly responsible for maintaining iongradients through pumping and regulation of ion channels. Brain tissue is metabolically veryactive compared to other tissues. A reliance on the low ATP producing glycolytic pathwayduring anoxialischemia results in the inability to maintain intracellular [ATPJ as discussed inthe Introduction. The loss of [ATPJ would cause energy consuming processes to cease,including the ion pumping action of Na,K-ATPase andCa2-ATP se. The increasing[Ca+2]j and decreasing [ATP] could potentially activate a host of ion channels sensitive tolow [ATPJ and Ca2 including non selective ion channels (Ashford et at., 1988; Partridgeand Swandulla, 1988) as well as voltage gated ion channels (Hille, 1992). The ultimate apexof these events would be the observed anoxic/ischemic massive release of neurotransmitters(Beal, 1992; Benveniste, et a!., 1984; Choi, 1988a; Choi, 1988b; Globus et at., 1988;S\u00C3\u00A1nchez-Prieto and Gonzalez, 1988) which would only serve to perpetuate the membrane andcell degeneration. This hypothesis will be discussed in further detail in chapter 5.27Current support for this \u00E2\u0080\u0098ATP\u00E2\u0080\u0099 hypothesis is indirect, but includes such observations asbathing the slice with creatine leading to prolonged anoxic and ischemic survival (Leblondand Kmjevic, 1989), the extracellular addition of high energy compounds to neurons aftercyanide exposure leading to a partial re-establishment of the membrane ion gradient(Caidwell eta!., 1960), hypothermia protecting the neuron (Okada et at., 1988b), andcorrelating low [ATP] concentration with the loss of electrical activity (Lipton andWhittingham, 1982).Our data support the concept that metabolic inhibition, and hence, limiting ATP, plays amajor role in the loss of neuronal membrane resistance. First, the cell\u00E2\u0080\u0099s ability to maintainresistance and membrane potential is inversely related to the degree of metabolic inhibition.Thus, the use of nitrogen gave the longest degree of survival followed by pharmacologicalanoxia, IAA, and finally, the most inhibitory regime used, pharmacological ischemia (Fig. 3).The use of nitrogen presumably allows the cell to exist longer due to the fact that thecytochrome system is not immediately inhibited; in contrast, sodium cyanide leads to analmost immediate and total inhibition of the system aided by the fact that the aCSF wasequilibrated with nitrogen. The use of IAA alone is the most severe single drug treatment.The inhibition of glycolysis by IAA leaves only the low levels of endogenous fatty acids,lactate, Krebs cycle intermediates, and the reserves of high energy phosphates to fuelmetabolism. The most severe was the use of pharmacological ischemia which inhibited allenergy production and allowed the cell to use only its reserve of high energy phosphates.Secondly, the observation that the turtle lasted 10 times longer during treatment of IAAcompared to the rat also supports this statement. In an earlier paper (Suarez et a!., 1989), weshowed that in the turtle brain resting metabolic rates were approximately 1/10 to 1/20resting metabolic rate in a rat brain in vivo. Assuming a of 2 for turtle metabolism(Funk and Milsom, 1987), the expected difference between the turtle and the rat neuronaldepolarization would be 5- 10 fold since in the turtle ATP would be consumed 5- 10 times28more slowly assuming usable energy stores were similar between the two tissues (McDougaleta!., 1968).Although there have been some suggestions that cyanide does not provide a good mimicof anoxia (Aw and Jones, 1989), our data suggest the opposite. Pharmacological anoxia andanoxia provided qualitatively similar results (Figs. 3 & 6). The turtle responded well with thetreatment and illustrated no difference to either treatment suggesting that pharmacologicalanoxia is not alternately inhibiting other processes which are necessary for survival underthese circumstances.The pharmacological mimic of ischemia appears to reflect the observed extracellularobservations well (Hansen, 1982; Reiner eta!., 1990). The initial slow gradual rise in [K],as observed in the extracellular space, could be caused by an evoked K+ current resulting inthe initial loss of membrane resistance and concurrent neuronal hyperpolanzation (Krnjevicand Xu, 1989). The sequential sudden and massive loss in membrane resistance resulting inthe depolarization of the neuron would correspond to the sudden drop of [Na+] and [Ca2lin the extracellular compartment since these ions would be entering the neuron flowing downtheir electrochemical gradient. Further confirmation that pharmacological ischemia providesa plausible in vitro mimic of the in vivo response awaits simultaneous recording of changes inmembrane potential and extracellular ion concentrations.There is currently considerable evidence that one mechanism which aids the turtle brainin surviving anoxia is the ability to depress brain metabolic rate (Kelly and Storey, 1988;Lutz, et al., 1984; Robin, Ct al., 1979). Two biochemical mechanisms are actively beingstudied for the mechanism of this depression. The first involves the down regulation of ionchannels responsible for the normal ion fluxes in a resting state and is termed \u00E2\u0080\u0098channel arrest\u00E2\u0080\u0099.The metabolic depression is the result of reduced ion pump activity due to the reduced ionleakage. The prediction which is forwarded is that anoxia tolerant species will have cellmembranes which are less conductive than non anoxia tolerant species. This hypothesis willbe discussed in detail in chapter 4. The second hypothesis for the reduction of turtle brain29metabolic rate is a reduction in brain electrical activity (Lutz, Ct a!., 1985; Suarez, 1987) aptlytermed \u00E2\u0080\u0098spike arrest\u00E2\u0080\u0099 (Sick et al., 1993). Three general mechanisms could alter brainelectrical activity in the turtle brain during anoxia: i) increasing the release of inhibitoryneurotransmitters such as y-aminobutyric acid (GABA) and decreasing the release ofexcitatory neurotransmitters such as glutamate; ii) altering the physical properties of neuronssuch as the membrane potential (hyperpolanze) or a resistance (decrease) to make theneurons less excitable; and iii) directly regulating ion channels such as voltage gated Ca+2and Na+ channels.With regards to the first proposed mechanism, studies support that the anoxic turtle brainincreases substantially in [glycine], [taurine], and [GABA] inhibitory neurotransmitters(Hitzig eta!., 1985; Lutz and McMahon, 1985; Nilsson eta!., 1990). Additionally, there isevidence for the increased release of these neurotransmitters during anoxia while the releaseof the excitatory neurotransmitter glutamate remains relatively constant (Nilsson and Lutz,1991). However, increase in brain concentrations as well as release does not occur untilabout 120 minutes into the anoxic insult (Lutz and McMahon, 1985; Nilsson, et al., 1990;Nilsson and Lutz, 1991) where as reduction in electrical activity has been implicated to occurvery rapidly ( 30 mins) (Feng et al., 1988; Feng et al., 1990; Feng eta!., 1988; Sick, et al.,1982). Additionally, increases in these neurotransmitters occur to both reptiles and mammalsduring anoxia (Nilsson et al., 1991) suggesting that this may not be an adaptation but abyproduct of the insult. These data suggest that at least on the short term basis, we mustexamine different mechanisms. The data reported here can neither support nor refute thismechanism as playing a role during anoxia in the cortex.With regards to changes in physical properties of the neuron, we have tested threshold,MP, and resistance (discussed in chapter 4). There were no significant changes in any ofthese parameters (Table 1). Interestingly, studies from the cerebellum of the anoxic turtlesuggest that anoxic cells in this area of the brain lose resistance, and MP during anoxia. In30addition, spike thresholds appear more depolarized(P\u00C3\u00A9rez-Pinz\u00C3\u00B3n et a!., 1992a) suggestingthat the cerebellum may have different anoxic control mechanisms compared to the cortex.The third set of alterations involves the direct regulation of ion channels which arevoltage gated. By removing or blocking these channels, action potential generation wouldbe inhibited. Evidence in the anoxic turtle cerebellum suggests that voltage gated sodiumchannels are inhibited or possibly even physically removed from the membrane as reportedfrom binding studies(P\u00C3\u00A9rez-Pinz\u00C3\u00B3n et at., 1992c). However, our evidences in the cortexshows no change in the threshold or the rate of spike rise which would be expected if therewere fewer active voltage gated Na+ channels in this area of the brain. This observationagain suggests that regulation in the cerebellum may be different than in the cortex. One sidemechanism to this hypothesis would be to make receptors less sensitive to excitatoryneurotransmitters, (such as glutamate activation of Ca+2 channels on the postsynapticmembrane). Cortical studies using fura-2 suggest that the Ca+2 influx is 75% reduced for agiven application of glutamate during anoxia compared to normoxia (Bickler and Gallego,1993). However, the turtle brain increases its [Ca+2]oby 6 fold during an anoxic bout (Cserret at., 1988). Under these conditions, if Ca+2 channels were not down regulated in responseto neurotransmitter application, there would be a state of hyperexcitivity. Thus, whether thisdown regulation is simple a response to the 6 fold increase in the cerebral spinal fluid [Ca+2]or an attempt to decrease electrical activity is unknown.The emphasis throughout this chapter has been on the importance of supplying cells withan adequate amount of energy (ATP) in order to maintain ion homeostasis during anoxia. Ifthis theory is correct, then the turtle cell in the absence of any energy production shouldrapidly lose its membrane potential. The results of pharmacological ischemia on the turtleneuron are illustrated in Fig. 5c,d. Although this is the first study to observe turtle neuronalresponse directly to total metabolic inhibition, an extracellular rise in [K+j has been reportedwith both the addition of IAA plus nitrogen (Sick, et al., 1982) as well as with clamping thearterial supply to the turtle brain (Sick, et al., 1985). The results in all cases clearly31demonstrate the importance of glycolysis for maintaining ion honieostasis across theneuronal membrane. Statistically, the turtle neuron maintains a membrane potential nolonger than the rat neuron which suggests that the turtle cell is not designed to survive on lowamounts of ATP but is designed to maintain ATP concentrations during anoxia. Thishypothesis is further strengthened by the observation of the turtle brain has been shown toexhibit a Pasteur effect (Kelly and Storey, 1988) as well as the comparatively large amountsof key glycolytic control enzymes compared to the rat at physiological temperatures (seechapter 3)Our data support metabolism playing a direct role in the maintenance of ion homeostasis,but it is unclear if the rapid failure of the neuron which accompanies anoxia and ischemia andtheir pharmacological mimics is due to a lack of energy dependent pumping, channelopening, or the release of excitatory neurotransmitters. It is conceivable that the mechanismsof failure between the two insults may not be the same. To look at one facet of this model,ouabain was used to block Na+K+ATPase and observe the resulting time to depolarization.The results indicate that the ouabain treatment (Figs. 3 & 8) did not adequately mimic any ofthe previous times of survival. In addition, there was no initial hyperpolarization in any ofthe cortical cells. This observation is in contrast to hippocampus in which a slighthyperpolanzation has been observed using lower concentrations of ouabain (Fujiwara, et aL,1987). In addition, there was no initial loss in whole cell resistance until there was a changein the membrane potential of the cell. This data suggests that although the anoxic failurecould be explained by the inhibition of the pump system, it seems unlikely that the rapidmembrane degeneration observed in pharmacological ischemia could be explained fully byion pump inhibition.32CHAPTER 3: A BIOCHEMICAL AND MICROCALORMETRIC STUDY OF THETURTLE CORTEXPrefaceThis chapter explores metabolic and biochemical mechanisms which allow the turtlebrain to survive extended bouts of anoxia. This chapter is excerpted (in part) from C. J..Doll, P. W. Hochachka, and S. H. Hand (submitted, J.E.B.). The enzymatic measures wereadapted from R. K. Suarez, C. J. Doll, A. E. Buie, T. G. West, G. D. Funk, and P. W.Hochachka (Am. J. Physiol. 257: R1083-R1088, 1991).IntroductionLong term anoxic survival of the freshwater turtle (Chrysemys picta) is well documentedin the literature with individuals having survived forced submergence for up to 6 months at 3\u00C2\u00B0C (Ultsch, 1985). The high sensitivity of the mammalian brain to anoxia and ischemia(Hansen, 1985), has lead to the turtle brain being used as a comparative model for anoxicstudies. However, only recently have healthy in vitro preparations of turtle brain becomeavailable. Three such preparations are the turtle brain stem slice preparation (Jiang et al.,1992), whole cerebellum preparation (P\u00C3\u00A9rez-Pinz\u00C3\u00B3n, et al., 1992b), and the cortical slicepreparation (Connors and Kreigstein, 1986; Chapter 2).In the previous chapter, we demonstrated that the turtle cortical neurons can surviveanoxia and still maintain a healthy membrane resistance, potential, and ability to fire actionpotentials. Several biochemical mechanisms are currently believed to aid the turtle brain insurviving anoxia, [for review see (Lutz, 1992)]. However, we believe that two adaptationsare central to helping maintain energy stability in the turtle brain: (i) an enhanced ability toproduce ATP (glycolytic capacity) coupled with the ability to increase glycolytic flux duringanoxia (Pasteur effect); (ii) a low metabolic rate with the ability to further depressmetabolism with anoxia.As discussed in Chapter 1, the Pasteur effect serves to replace some of the lost ATPproduction ( 2 vs. 36 ATP for glucose) resulting form the exclusive use of glycolysis during33anoxia. However, the Pasteur effect appears to be only temporary and never achieves thepredicted 18 fold increase which is necessary to replace all of the lost ATP production.Mammalian brain studies (in vivo and in vitro) have indicated glycolytic activation between 2and 10 fold during hypoxialanoxialischemia or simulated conditions (Borgstrom et aL, 1976;Drewes and Gilboe, 1973; Kauppinen and Nicholls, 1986; Kauppinen and Nichols, 1986;Ksiezak and Gibson, 1981; Lowry, et al., 1964; Rolleston and Newsholme, 1967a) which (inpart) could explain the inability of the rat brain to maintain [ATPJ (Lowry, et al., 1964;Ridge, 1972)during these insults (maintenance of energy balance will be discussed in furtherdetail in Chapter 5). Although in vivo turtle brain studies indicate glycolytic activation withanoxia (Kelly and Storey, 1988; Lutz, Ct al., 1984), none have indicated a full 18 foldactivation, and the one study which examined turtle brain slices in detail for glycolyticactivation failed to measure any Pasteur effect (Robin, et al., 1979). The ability of the turtlebrain to maintain [ATPJ during anoxia (Kelly and Storey, 1988; Lutz, et al., 1984) combinedwith the low or non existent Pasteur effects suggests that the turtle brain is experiencing somedegree of metabolic depression with anoxia.Metabolic depression was first measured in the turtle using calorimetry. The initialexperiments indicated that turtle whole body heat dissipation decreased 85% (Jackson, 1968).Whole body calorimetry has also been used to measure metabolic depression in other anoxiatolerant vertebrates with similar results. For ectotherms, for example, the goldfish (Carassiusauratus) depresses heat production 70% (van Waversveld et al., 1988), and the marine toad(Bufo bufo) has also been observed to reduce heat production by 80% with submersion(Leivestad, 1960). Current in vivo studies on the turtle brain based on lactate accumulationover the duration of anoxic insult suggests a severe (>80%) metabolic depression withanoxia (Kelly and Storey, 1988; Lutz, et al., 1984). However, these numbers may bemisleading due to lactate not being trapped in the brain tissue compartment (Drewes andGilboe, 1973; Hawkins, 1985; Schurr et al., 1988). As a result, the lactate pool measured34may not include the lactate washed out of the tissue compartment and therefore, may notreflect the anoxic metabolic rate.In vitro turtle brain preparations are able to survive extended periods of anoxia comparedto rat counterparts as observed in Chapter 2 and other studies (Jiang, et aL, 1992; PerezPinzon, et al., 1992b). However, whether metabolic depression occurs in the in vitropreparation as suggested by the above the in vivo studies has not been established. In thischapter, three areas of metabolic adaptation in the turtle cortical slice will be explored withrespect to anoxia: (i) metabolic rate, (ii) changes in adenylates and energy balance, (iii)glycolytic capability and activation.MethodsSlice PreparationTurtles (C. picta), 150 - 300 g were cold anesthetized before decapitation. The brain wasrapidly dissected free and immersed in artificial cerebrospinal fluid (aCSF) which had beenprecooled and equilibrated with 95% 02/5% C02. The cortical tissue was then dissectedfree as previously described (Connors and Kreigstein, 1986). Blocks of whole cortex (z 500M thickness) were stored at room temperature 22\u00C2\u00B0C in a recirculating holding chamberuntil their use in the calorimeter or incubation chamber.For the enzymatic assays, the brains (turtle and rat) were rapidly dissected free aftercervical dislocation or decapitation. The brains were dissected over ice and the cortex wasremoved. The cortex was then placed in 9 Volumes of 50 mM tris (hydroxmethyl)aminomethane (Tris)-Cl (pH 7.4 at 4 oC), 2 mM EDTA and 0.5% Triton X-100. Rat cortexwas homogenized with and UltraTurrax homogenizer 3 times 10 sec. each time. Turtlebrains were homogenized with a hand glass homogenizer and sonicated for 10 sec twice.Homogenates were the spun at 12,000 g for 5 mins. The resulting supernatant was then usedfor enzyme assays.35Artificial Cerebrospinal FluidThe aCSF for the turtle was a modification(Connors and Kreigstein, 1986) whichconsisted of (in mM) 96.5 NaCI, 2.6 KCI, 2.5 CaC12, 2.0 MgCl, 2.0 NaH2PO4, 10 glucose,26.5 NaHCO3, and 0.03 phenol red as a pH indicator. Final pH of the solution was 7.4 whensaturated with 95% 02/5% C02. To mimic anoxia, the aCSF was equilibrated with 95% N2/5% COD. Pharmacological anoxia was the same solution as anoxia, but in additioncontained 1 mM NaCN.AdenylatesSlices used in the adenylate measurements were incubated (in a recirculating slicechamber) for two hours in oxygenated aCSF before the experiment was commenced. Thisincubation period allowed slices to stabilize after dissection and was done for all experimentsincluding the calorimetry studies. After incubation, the slices were transferred to 60 mlrecirculating semiclosed chambers. The chambers were sealed at the top with an 0 ring andbolted shut. All gases flowed through gas impermeable Viton tubing. Gas at positivepressure inside the chamber was allowed to escape via a variable flow escape valve. Byadjusting the opening of the valve, a slight but constant positive pressure could be maintainedon the liquid above the slice maximizing saturation of the gas used and minimizingcontamination from the atmosphere. For anoxic experiments, chambers were tested foroxygen contamination using a Radiometer gas analyzer and 02 electrode. No 02contamination was detected compared to a sodium dithionate solution. Upon completion ofthe experiment, the tissue was removed from the chamber with a pair of forceps and blotteddry. Slices were immediately immersed in precooled (-5 \u00C2\u00B0C), preweighed vials of 5%perchiorate acid (PCA) and weighed. The procedure from removal to PCA immersion tookless than 20 seconds. The tissue was then sonicated 3 times (10 sec. duration) with 30seconds in between each burst. The vials were kept on a rock salt/ice slurry (t -5\u00C2\u00B0C). Thehomogenate was then spun at 15000 rpm for 20 minutes (-5 \u00C2\u00B0C). The resulting supernatantwas removed and adjusted to a pH of 2.5 with 3MK2C03(potassium bicarbonate). The36partially neutralized supernatant was respun at 15000 rpm for 20 minutes (-5 \u00C2\u00B0C) to removeprecipitated perchiorate salts. The resulting supernatant was removed and frozen in liquidN2. Samples were stored at -70 \u00C2\u00B0C until analysis (within one week).ChromatographyAll adenylates in this paper were measured using high performance liquidchromatography (HPLC) on an LKB 2152 HPLC controller and 2150 titanium pump coupledto a 2220 recording integrator as previously described (Schulte et at., 1992). In brief, sampleseparation was performed on an Aquapore AX-300 7 jim weak anion exchanger (Brownleelabs). Elution off the column was performed by running an isocratic solution of 60 mmol/lKH2PO4 (pH 3.2) for the first five minutes followed by a linear increasing concentration to750 mmol/l KH2PO4 (pH 3.5) for the next 10 mm. This concentration and pH wasmaintained for the next 12 mm. The column was re-equilibrated for 5 mm with the startingbuffer for the next run. Adenylates were detected using a BlO-RAD flow through UVdetector at 254 rim. Standard curves were done for all adenylates to assure linearity overrelevant concentrations. Analytical reagent grade KH2PO4was purified before use in theHPLC by running a stock IM solution through a BlO RAD Econo column packed with anionexchange resin (AOl X8, chloride form, a cation exchanger (chelex 100, sodium form) andactivated charcoal (14-60 mesh). The solution was kept at 4\u00C2\u00B0C and constantly recirculatedthrough the column by a peristaltic pump ( 3 mI/mm) for 24 hrs. Before use, the stock 1Mbuffer was diluted, pH adjusted, and vacuum filtered (0.22 jim filter).MicrocalorimetryCalorimetry was used in this study because it allowed the continuous measurement ofheat dissipation over reversible bouts of oxygen limitation with tissue from a single turtle.The only other method giving this advantage is NMR spectroscopy. However, preliminaryexperiments with NMR indicated that sufficient cortical tissue was not attainable from asingle turtle to permit adenylate concentrations to be measured during a reasonable timeframe. Additionally, the low normoxic metabolic rate of the turtle brain gave an extremely37slow spin-lattice relaxation time (TiM) making saturation transfer measurements impossible(Shoubridge eta!., 1982).Heat dissipation measurements on turtle cortical slices were performed with an 1KB2277 thermal activity monitor equipped with a 3.5 ml perfusion chamber. The perfusionaCSF was identical to that described above for normoxia, anoxia, and pharmacologicalanoxia. Perfusion aCSF was held in semiclosed glass bottles that were kept at 25\u00C2\u00B0C with acirculating water bath. Flow lines for gases and aqueous media were made of Viton andstainless steel, respectively. Flow rate of the perfusion media through the slice and chamberwas set at 15.0 mI/hr. and the temperature was maintained at 25\u00C2\u00B0C for all experiments. Heatdissipation in microwatts was time corrected by the calibration unit of the calorimeter; typicaltime constants were t1 = 600s and t2 = 17100s. Heat measurements were taken every 60sand stored on an IBM XT personal computer. For all experiments, blanks (no slices) wererun in parallel and subtracted from the measurements obtained with slices.Two cortical slices from a single turtle were prepared as described above. Aftercalibration of the instrument, the slices were placed in the perfusion chamber and loweredinto the calorimeter. Heat dissipation was then monitored for a minimum of 1 hr before datacollection was begun. Total time from dissection to collection of data was approximately 2hrs. The normoxic solution was switched over to nitrogen perfusion or pharmacologicalanoxia once heat dissipation of the slices had stabilized to less than 0.2 jsW/min change.During perfusion with nitrogen saturated medium, previous measurements indicated a periodof 60 - 90 mm. was needed at similar flow rates to reduce oxygen content below 0.5% airsaturation in the excurrent flow (Hand and Gniager, 1988). For nitrogen perfusionexperiments, nitrogen equilibrated aCSF was switched to normoxic aCSF to monitorrecovery. The experiment was terminated once heat dissipation had returned to predictedlevels of heat dissipation or stabilized.38Enzymatic MethodsFor hexokinase, the following recipe was used: 5 mM glucose (omitted for control), 5mM ATP, 5 mM MgC12, 0.5 mM NADP, 5 mM dithiothreitol (DTI\u00E2\u0080\u0099), and 50 mMimidazole-CI (pH 7.5). For lactate dehydrogenase, the recipe was as follows: 1 mM pyruvate(omitted for control), 0.15 mM NADH, 5 mM DTT, and 50 mM imidazole-Ci (pH 7.0). ForCitrate Synthase the recipe contained 0.5 mM oxaloacetate (omitted for control), 0.3 acetylCoA, 0.1 mM dithiobisnitrobenzoic acid, and 50 mM Tris-CI (pH 8.0). Total volume ofassay was 1.0 ml. Measurements were done at 25\u00C2\u00B0C turtle, and 37\u00C2\u00B0C rat and started by theaddition of 10-20 p1 of crude supernatant. All Enzymatic assays were done using a PyeUnicam SP-1800 spectrophotometer with water jacketed cuvette holders and a chart recorderfor data acquisition.Calorimetric Data AnalysisData collected with IBM XT was imported into CA-Cricket Graph Ill for the Macintosh.Based on blank runs after the removal of slices, any baseline drift or signal from bacterialcontamination (due to high glucose in the aCSP) was apportioned linearly over the entireexperiment using CA - Cricket Graph III.The heat depression data was then graphed similar to Fig. 9a and b. These data were thencurve fit by hand to estimate aerobic values over the nitrogen perfusion period. Data (controland treatment) over the anoxic period were then digitized using Sigmascan. Percentage heatdepression was then calculated. ATP turnover estimates were derived by a conversion factorof .76 pmoles ATP/g/minlmW for aerobic heat dissipation and .86 pmoles ATP/g/minlmWfor anaerobic heat dissipation with the following assumptions and conversion factors:Aerobic Conditions1) The turtle slice is using only (glucose or glycogen) as a fuel (Clark and Miller, 1973;P\u00C3\u00A9rez-PinzOn, et al., 1992b).2) The aerobic catabolism of glucose to CO2 and H20 yields -469 - 476 kJ/mole 02(Gnaiger and Kemp, 1990).3) For glucose, 6 moles of ATP are produced for every mole of 02 consumed.39Anaerobic Conditions1) The turtle slice is using only glucose as a fuel (see ref. to 1 above).2) Lactate is the only anaerobic end product produced (Robin, et al., 1979)3) The catabolism of glucose to lactate yields -70 U/mole lactate (Gnaiger and Kemp,1990).4) Production of one mole of lactate from glucose provides one mole of ATP.ResultsCalorimetryTypical calorimetry recordings are illustrated in Figs. 9a,b. Results demonstrated thatcortical cells depress heat flow under both treatments (Fig. lOa,b). Both nitrogen perfusionand pharmacological anoxia produced rapid declines in heat dissipation, but pharmacologicalanoxia produced a significantly (P 0.05) more rapid effect (32.8 \u00C2\u00B1 18SE% vs. 8.6 \u00C2\u00B1l.8% by 20 minutes; Fig 14a), due to the slow washout of oxygen. By 60 minutes into theinsult, treatments had decreased the rate of change to 6.3 \u00C2\u00B1 08SE %/hr (nitrogen perfusion)vs. 3.1 \u00C2\u00B1 08SE (pharmacological anoxia). Nitrogen perfusion followed by normoxiareestablishes heat flow to predicted control level in all trials similar to Fig. 9a. Average heatdepression (percentage of control) by the termination of the insult was significantly less inthe nitrogen perfusion group compared to the pharmacological group (36.3 \u00C2\u00B1 2.6 vs. 49.3 \u00C2\u00B1l.6; Fig. lib).Observations of heat flow depression indicated highly significant differences between thecontrol heat dissipation and both treatment groups by 120 mins (Fig. 12a). Similarly, theATP utilization rates based on these heat dissipation values indicated large differencesbetween control and treatment groups by the end of the insult (nitrogen perfusion control,1.36 \u00C2\u00B1 vs. 120 mm., 0.97\u00C2\u00B1 0.09w; pharmacological anoxia control, 1.42 \u00C2\u00B1 vs.120 mm., 0.82 \u00C2\u00B10\u00E2\u0080\u00A27SD pmoles ATP/g/min; Fig 12b). The ATP metabolism was based on aconversion factor of .76 pmoles ATP/g/min./mW for normoxia and .86 pmolesATP/g/min./mW for anoxia (see Methods). Because heat output under aerobic conditions40A. Normoxia Nitrogen NormoxiaCC?)\u00E2\u0080\u00A2 i.:\u00E2\u0080\u0094 I i i i I i i I i i I i i i I I . I I0 40 80 120 160 200 240 280 320Time (mins)B2.25-Normoxia Pharmacological AnoxiaE -1.50--CC. 1.25--C?)\u00E2\u0080\u0098 i.()()\u00E2\u0080\u0094 I , I0 40 80 120 160Time (mins)Figure 9. Representative chart recordings of heat dissipationover nitrogen perfusion (A) andpharmacological anoxia (B) for turtle cortical slices (25 \u00C2\u00B0C). Vertical lines across charttracings indicate the time at which the indicated solution entered the calorimeter chamber.41A453525155-5Figure io. The individUal percent heat depression from the predicted orrespondmg controlvalue over nitrogen perfuSiOn (A) and pbar acological anoxia (B) for all tested slices. Time0 indicates the point at which the test solution first entered the calorimeter slice chamber.0Duration of Nitrogen Perfusion (mm.)B840 20 0Duration of Pharmacoiogjcaj Anoxia (mm.)42\u00E2\u0080\u00A240Cl).C.\u00E2\u0080\u0094Cl)C.\u00E2\u0080\u0094Cl)_________Cl)Figure 11. The average heat depression (percentage) relative to the predicted correspondingcontrol value. (A) The heat depression at 20 minutes into the nitrogen perfusion orpharmacological (ph. anoxia) insult. (B) The average percentage of heat depression from thecorresponding predicted control value at the end of the insult (120 mins). For both (A andB), values are derived from Fig. 10. An asterisk denotes a significant difference (P 0.05,unpaired t- Test) from the corresponding treatment. See results for exact numbers. Data(nitrogen perfusion, n=4; pharmacological anoxia, n=3) are mean values \u00C2\u00B1 SE.AB*0Nitrogen Ph. Anoxia*T6050-40.30-20-10-0-rNitrogen Ph. Anoxia43A25.*11.0.I7 7O.0\u00E2\u0080\u0094Initial Nitrogen Ph. AnoxiaControl13 Treatment2.0-*00.0-Initial Nitrogen Ph. AnoxiaFigure 12. Effects of nitrogen perfusion and pharmacological anoxia on heat dissipation andmetabolism (A and B respectively) compared to the corresponding predicted control value.The initial value in both (A and B) denotes the condition of the slice at the beginning of theexperiment. In both figures, an asterisk denotes a significant difference from thecorresponding control (P 0.001; paired t-Test). The ATP turnover is calculated from A (seemethods for further detail). Data (anoxia, n=4; pharmacological anoxia, n=3) are meanvalues \u00C2\u00B1 SD.44A D2.5 40.\u00E2\u0080\u00940-\u00E2\u0080\u0094 ControlT Anoxiao 120 240 0 120 240Time (mins)B E0.3 1.________:________0 120 240 0 120 240C F0 120 240 0 120 240Figure 13. The effect of anoxia and pharmacological anoxia on adenylates, energy charge,and adenylate ratios for turtle cortical slices (25 \u00C2\u00B0C). Time sequence in (A) and symboldefinitions in (D) are the same for all graphs. An asterisk denotes a significant difference forthat particular mean value compared to the corresponding control mean value (P 0.05;Tukey HSD test). Data (n=4) are mean values \u00C2\u00B1 SE. Concentrations are expressed asjmo1es/g wet wt. (ATP; ADP) and pmoles /g wet wt (AMP; IMP).45Table 2. Enzyme activities in the cortex of the rat and turtleEnzymes Rat (37 C) Turtle (25 OC) Turtle (37 OC)bHexokinase 25.7 \u00C2\u00B1 1.9 21.7 \u00C2\u00B1 1.2 43.4Lactate Dehydrogenase 217.1 \u00C2\u00B1 14.6 177.2 \u00C2\u00B1 7.5 354.4Citrate Synthase 60.9 \u00C2\u00B1 2.8 17.2 \u00C2\u00B1 0.4 34.4aValues are means \u00C2\u00B1 SE expressed in units of activity; n =6.bValues are mathmatically obtained from turtle (25 \u00C2\u00B0C) values, assuming a Qio of 2.46tended to drop slowly over the experiment, an initial control ATP turnover rate (1.72 \u00C2\u00B1O.O6 pmoles ATP/g/min; n =7) was calculated for comparison to other experimentalpreparations.Adenylates were measured in this experiment for two reasons. First, although somestudies have been done on the whole turtle brain with respect to these metabolites, very littleis known on individual tissue compartments in the brain and their changes with anoxia.Secondly, one assumption made in calculating ATP turnover directly from heat dissipationwas that energy intermediates remain in steady state (Shick et al., 1983). This information isimportant not only for the metabolic rate determination, but also as an indicator of the healthof the tissue. ATP, ADP, and AMP showed no significant changes over the control datapoints (P 0.05; Fig. 13a,b,c). The 240 minute [ADP] levels for both treatment groups weresignificantly higher compared to the corresponding control time point (P 0.05; Fig 13b).IMP did not change significantly under control or treatment conditions (P> 0.05; Fig. 1.3d).Differences among adenylates exhibited higher statistical significance when expressed asenergy charge (E.C.) or adenylate (ATP/ADP) ratios. Both anoxia and pharmacologicalanoxia produced significant differences in E.C. and ATP/ADP ratios at time 240 minutescompared to equivalent control values (P 0.05 Fig. 13e,f).Enzymatic AnalysisThe results of the enzymatic study of turtle and rat cortical slices are illustrated in Table2. At physiological temperatures, turtle (25 0C) and rat (37 OC), both species expressedapproximately the same amount of enzyme activity for lactate dehydrogenase andhexokinase, but the turtle cortex expressed approximately 30% of the acitivity of the ratcortex for Citrate Synthase. When the results were normalized for temperature (Qio 2) theturtle brain expressed twice the activity of hexokinase and lactate dehydrogenase and onlyabout 60 % of the activity for citrate synthase found in the rat cortex.47DiscussionThere were four goals to this study: i) to measure adenylate pools during anoxia, ii) toestablish a metabolic rate for the turtle cortical slice, iii) to establish whether the turtlecortical slice further depresses metabolism with simulated anoxia, and iv) to examine turtlecortical slices using both enzymology and calorimetry for evidence and capacity of a Pasteureffect.AdenylatesThe control concentrations reported here for all adenylates are similar to those reported inslices and whole brain in vivo (Kelly and Storey, 1988; Lutz, et a!., 1984). Kelly and Storey(1988) reported that ATP tended to rise in the whole brain with anoxia, but our results in thecortical slice are similar to those of Bickler which reported no increase (Bickler, 1992a).Significant differences with anoxia did occur with respect to [ADP], adenylate ratios, andenergy charge compared to the corresponding control time; however, these changes do notoccur until 240 mm. into the insult which is beyond the time course of the calorimetryexperiments.Normoxic Metabolic RatesThe higher metabolic rate for cortical slices (1.72 jmoles ATP/g/min; Fig. 12b),calculated from heat values of normoxic controls, was obtained at the beginning of each run,because heat dissipation tended to drop over the course of the experiment (Fig. 9a,b). Thereason for this decrease is unknown, but presumable is due to degeneration of unhealthycells. Interestingly, our initial value is very close to the turtle whole brain ATP utilization(assuming 36 ATP/glucose catabolized) of 1.9 J4moles ATP/g/min (Suarez, Ct al., 1989)based on deoxyglucose measures. The close agreement between the two independent studiessuggests that the initial ATP utilization is an accurate assessment. Glucose consumption(initial) in guinea- pig cortex slices (37 \u00C2\u00B0C) yields a value of 16 jmoles ATP/g/min whenglucose consumption values are corrected for glycogen breakdown and lactate production(Rolleston and Newsholme, 196Th). This estimate indicates a 9.3 fold difference in48metabolic rate between the turtle and mammalian slice preparation consistent with our earlierwhole brain deoxyglucose observation of a 12 fold difference (Suarez, et al., 1989).Additional support for the difference in metabolic rates between the turtle and the rat cortexcan be seen in the measurement of citrate synthase (Table 2). The rat cortex (37 \u00C2\u00B0C)expressed 3.5 times the activity compared to the turtle cortex (25 \u00C2\u00B0C) supporting a largeraerobic flux rate in the rat cortex.Anoxic Metabolic RatesThe nitrogen perfusion experiments indicate a reduction in heat dissipation in the turtlecortex during 02 deprivation. The resulting ATP utilization estimates indicate that the turtlecortex depresses metabolism between 30% (nitrogen) and 42% (phaimacological anoxia)during 02 limitations. Estimates from lactate accumulation studies with whole brain as wellas whole brain slices suggest a larger ( > 80%) metabolic depression (Kelly and Storey, 1988;Lutz, et al., 1984; Robin, et al., 1979). While our estimates do not support this degree ofdepression in the cortex, the data clearly show a significant metabolic reduction.The discrepancies between previous in situ studies and the current in vitro study may bepartially explained by the absence of spontaneous electrical activity in the bran slicepreparation. Current evidence suggests that one mechanism for turtle brain metabolicdepression in the in vitro is reduction of spontaneous electrical activity (Lutz, 1992). Brainslices consume approximately 50% less oxygen than the intact tissue (Lipton andWhittingham, 1984). This discrepancy has been attributed to the loss of spontaneouselectrical activity in the tissue slice (Mcllwain and Bachelard, 1971). This hypothesis issupported by the observations that tetrodotoxin (voltage-gated Na+ channel blocker) causesonly a 5% reduction in oxygen consumption in vitro(Okamoto and Quastel, 1972), butbarbiturate anesthetics (electrical activity depressants) do inhibit oxygen consumption by50% in the intact brain (Siesjo, 1978). The absence of significant amounts of spontaneouselectrical activity in the turtle cortical slice may cause a reduced metabolic depressioncompared to the in vivo. The lack of electrical activity in the slice combined with the large49metabolic depression observed in the present study would suggest that other metabolicprocesses are being depressed besides electrical activity.Anoxic GapIn some organisms, the anoxic heat dissipation does not match indirect measures ofmetabolism (heat dissipation can exceed measured end product formation by as much as50%) (Gnaiger, 1980; Shick, et al., 1983). Several reasons suggest a negligible \u00E2\u0080\u009Canoxic gap\u00E2\u0080\u009Dfor our turtle cortical slice preparation (Hardewig eta!., 1991; Shick, eta!., 1983): i)adenylate levels were fairly constant, ii) all experiments were short duration, iii) use ofcyanide + N2 minimized any oxidative metabolism, and iv) the 1:2 matching of glucoseconsumption and lactate production in whole brain slices of anoxic turtles indicating no otherglucose catabolic pathways during anoxia (Robin, Ct al., 1979). Based on glucoseconsumption and lactate production, turtle whole brain slices yield an anoxic ATP utilizationrate of 0.88 pmo!es ATP/g/min (Robin, Ct al., 1979) which is almost identical to the .82j4moles ATP/g/min reported in this study suggesting that our estimation of the anaerobiccortical slice metabolic rate is accurate assessment. It is important to note, however, that ifsuch a gap does exists in our preparation, this discrepancy would increase the differencebetween control and anoxic ATP utilization.GlycolysisIn the calculation of ATP turnover, we have assumed that the slices are fully aerobic orfully anaerobic. Aerobic brain tissue generally has an anaerobic component (Robin, et a!.,1979; Rolleston and Newsholme, 196Th). Additionally, as a tissue enters an anoxictransition, there is a shift from aerobic to anaerobic metabolism making metaboliccalculations difficult (Gnaiger and Kemp, 1990). However, for our tissue, because thetheoretical amount of heat dissipated per mole of ATP consumed is similar between aerobicand anoxic conditions (0.76 and 0.86, respectively; see methods), the error associated withassuming the tissue is totally aerobic or anaerobic is small ( 12%). Consequently, we50conclude that heat dissipation along the experimental curves (Fig 9a or b) approximateschanges in ATP turnover fairly closely.Under conditions of pharmacological anoxia, the insult is rapid. The mixture of N2 +cyanide minimizes any oxidative metabolism in the transition phase giving an accurateanaerobic metabolic picture. As a result, the percent depression by 20 minutes forpharmacological anoxia is significantly greater than for nitrogen perfusion (Fig 1 la).Because adenylates through pharmacological anoxia are reasonably constant (indicatingglycolytic ATP supply is matching cellular ATP demands; Fig 13) and because the totalinhibition of oxidative metabolism, the heat dissipation curve reflects both metabolic ratechanges and glycolytic changes. Hence, changes in heat flux during pharmacological anoxiarepresent changes in both the metabolic rate and the glycolytic rate. Four possible anoxictransition scenarios are illustrated in Fig. 14. A 12% drop in heat dissipation with the insultwould be indicative of no metabolic depression in the slice due to the difference betweenaerobic and anaerobic ATP/mW heat production (see methods), and also would be indicativeof a large Pasteur effect (Fig. 14a). Note that this 12% would be a theoretical maximum andassumes that the control heat measure is 100% oxidative. A more likely scenario is thegradual decline of heat dissipation, and thus, metabolic rate indicating a gradually decliningPasteur effect (Fig 14b). A biphasic transition into pharmacological anoxia would be anindicator of both a sustained, but depressed metabolic rate and Pasteur effect during thebiphasic plateau of the curve (Fig. 14c). A rapid transition into a large heat flux depressionwould be indicative of a large metabolic depression which is most likely accompanied by aslight or reversed Pasteur effect. Note that due to the catabolic heat production from glucose(140 kJ/mole, anoxia vs. 2820 kJ/mole, aerobic), heat flux would have to drop to 5% ofcontrol measures before a reversed Pasteur effect would occur. However, this measure againrepresents a theoretical value which assumes that the control heat production is 100%aerobic. Kelly and Storey have suggested that the turtle brain goes through a biphasictransition into anoxia (Kelly and Storey, 1988). In the first phase, the glycolytic activation51A C100CC,]12 % drop 100Time= NormoxiaBPasteur Effect= Slight Pasteur Effect= Reversed Pasteur Effectif heat dissipation decreasesbelow 5%1005B1005Figure 14. Four possible scenarios for the response of cortical brain slices topharmacological anoxia(see text for further details).52occurs to maintain ATP concentrations. As the brain reduces its metabolism, glycolysis canreturn to normal or reduced rate (reversed Pasteur effect). All pharmacological cases appearto conform to a combination of 3 of these scenarios. There is initially a veiy rapid drop inheat dissipation similar to Fig. 14d which plateaus for 10 minutes similar to Fig. 14c. Atthis plateau point, the average heat dissipation for all three cells was 77% of controlindicating a 15.5 fold Pasteur effect which gradually dropped similar to Fig. 14b. Based onthe final heat dissipation measures for all three pharmacological anoxia cases (49 % ofcontrol) the final Pasteur effect would be 10.3 fold at 120 minutes of anoxia. Thus, theanoxic data support an initial large Pasteur effect which is gradually declining over theinsult..Enzymatic AnalysisThe correlation of both hexokinase and lactate dehydrogenase with glycolytic flux rateshas been documented (Hochachka and Somero, 1984). One reason why maximal enzymeactivity can be compared usefully in vitro is because there is a potent conservation ofcatalytic turnover of substrate for homologous enzymes in vertebrates (Hochachka andSomero, 1984). The enzyme activities reported in Table 2 indicates that the turtle cortex at25 \u00C2\u00B0C has an equal activity as the rat cortex at 37\u00C2\u00B0C. When activities are corrected fortemperature (Q10 2), the turtle brain exhibits twice the glycolytic capacity which supportsa relatively enhanced glycolytic capacity in the turtle cortex compared to the rat cortex.Two additional mechanisms deserve mention with regards to glycolytic capacity in theturtle. Recent studies have indicated covalent modification of glycolytic enzymes duringanoxia through reversible incorporation of phosphate into proteins. This modification canproduce rapid changes in the expressed activity of cellular enzymes and possible influencethere association with certain cellular or subcellular components (Brooks and Storey, 1989).Glycogen phosphorylase, phosphofructokinase, and pyruvate kinase were shown to bemodulated during anoxia in the turtle such as to increase activity. Additionally, in brain,hexokinase, phosphofructokinase, and phosphoglycerate kinase were all shown to bind to53subcellular components(Duncan and Storey, 1992). The resulting positioning is believed tochannel ATP to critical areas such as ion pumps.ConclusionData presented give strong evidence for the maintenance of adenylates during anoxia inthe turtle brain. The low metabolic rate combined with the ability to further suppressmetabolism during anoxia undoubtedly plays a critical role in allowing the cell to maintainenergy balance during anoxia. However, at least on the short interim, the ability to increaseglycolytic flux (Pasteur effect) combined with increased glycolytic machinery appears also toplay an important role. The end result of all of these adaptations is that cellular ATP supplycan meet demands even when ATP production is substantially inhibited by an anoxicepisode. The reader is refereed to chapter 5 where a more complete discussion of theseissues are presented.54CHAPTER 4: A CRITICAL TEST OF CHANNEL ARRESTPrefaceIn the previous chapter, calorimetric measures supported both a low normoxic metabolicrate and the ability to further depress metabolism during anoxia in the turtle cortical slice.This chapter explores the concept of reduced ion leakage as a mechanism for metabolicdepression in the turtle cortex during both normoxia and anoxia. This chapter is excerpted(in part) from the papers C. J. Doll, P. W. Hochachka, and P. B. Reiner (Am. J. Physiol. 261:R1321-R1324, 1991) and C. J. Doll, P. W. Hochachka, and P. B. Reiner (Am. J. Physiol., InPress). The enzymatic measures were adapted from R. K. Suarez, C. J. Doll, A. E. Buie, T.0. West, 0. D. Funk, and P. W. Hochachka (Am. J. Physiol. 257: R1083-R1088, 1989).IntroductionIf energy status is to be sustained in the turtle brain during anoxia, supply must meetdemand. Three basic mechanisms in principle could aid in the maintenance of energybalance during anoxia assuming the tissue is not substrate limited: (i) enhanced glycolyticcapability to aid in the supply of ATP to the cell, (ii) low normoxic metabolic rate to lessenthe initial ATP demand of the brain (Chapter 3), (iii) metabolic depression during anoxia tofurther accentuate a low normoxic metabolic rate.With respect to (i) and (ii) above, we have demonstrated that the turtle brain expressestwice the glycolytic capability of the rat brain while consuming only 1/9 to 1/12 the glucoseunder control (normoxic) conditions (Suarez, et al., 1989; Chapter 3). Additionally, earlierindirect estimates implied that the metabolic rate of the turtle brain appears low compared toother ectotherms (McDougal, et al., 1968). With regards to (iii), significant anoxic metabolicdepression has been estimated in both whole brain preparations (Kelly and Storey, 1988;Lutz, Ct al., 1984), as well as in slices (Robin, et al., 1979; Chapter 3).Two of the three protective mechanisms outlined above concern a reduced cellularmetabolic rate. However, the question remains as to what ATP requiring processes are beingsuppressed in both normoxia and anoxia to account for lowered ATP turnover. It has been55hypothesized that one mechanism which could reduce energy expenditure in turtle brainwould be to decrease ATP-dependent ion pump activity by reducing ion leakage (Hochachka,1986).All cell membranes leak ions. Leakage is the result of both intracellular and extracellularions flowing down their electrochemical gradients. Several cellular processes contribute tothis phenomena in neurons including activation of voltage- and ligand- gated channels,neurotransmitter release and uptake, co- and counter- transport systems, and leakage channels(voltage independent ion channels) (Hille, 1992). Maintenance of a homeostatic intracellularenvironment requires the redistribution of these ions through the use of energy demandingpumping systems such as the Na+K+ATPase which may consume as much as 50% of thecell\u00E2\u0080\u0099s resting metabolic rate (McBride and Milligan, 1985). One mechanism by which theturtle brain could reduce its metabolism would be to reduce ion leakage (Hochachka, 1986;Lutz, et al., 1985), which could conceivable occur through the reduction of any of the aboveleakage processes.The concept of regulating cell metabolism through regulation of ATP-dependent ionpump activity is not new (Whittam and Blond, 1964) and has been forwarded as a possiblemechanism for the action of thyroid hormone in controlling thermogenesis (Ismail-Beigi andEdelman, 1970). This idea was later expanded as a mechanism contributing to the origin ofendothermy (Edelman, 1976). The hypothesis predicts that endotherms have leakiermembranes, and thus increased ATP dependent ion pump activity serving as an indirectmechanism of heat production. A similar concept involving regulation of ion pumping by theCa+2-ATPase is believed to account for thermogenesis in marlin heater tissue, a modifiedmuscle which displays an expanded sarcoplasmic reticulum but has little contractile protein(Block, 1987). Taken together, this evidence provides strong support for the idea thatregulation of leakiness, and hence, ion pump activity, could regulate ATP turnover in thecell.56This paper tests two predictions of such a \u00E2\u0080\u0098channel arrest\u00E2\u0080\u0099 hypothesis (Hochachka, 1986;Hochachka, 1987). First, do turtle cortical pyramidal neurons display lower passive ionleakage (conductance) compared to the rat cortical pyramidal neurons during normoxia, andsecond, are leakage channels in the turtle pyramidal neuron further down regulated duringboth long and short term anoxia as a potential survival strategy.MethodsGeneralMethods for data acquisition, tissue preparation, and fluid composition for intracellularrecording techniques were identical to those described in chapter 2, pgs. 3-4. Notableexceptions of patch clamping techniques and data analysis in general are described below.The reader is referred to Appendix A and B for a more general discussion of these techniquesand terminology.Patch Clamp Tissue PreparationYoung Wistar rats (25-40 g) were anesthetized with halothane decapitated and the brainrapidly removed and immersed in precooled oxygenated aCSF. After a few minutes ofprecooling, a block containing frontal - parietal cortex was dissected free, glued withcyanoacrylate to a mounting block and sliced (400 iM thickness) on a vibratome. Sliceswere stored at 22\u00C2\u00B0C for at least 60 mi prior to use.Patch Clamping TechniquesMethods for the formation of whole cell seals in slices of both turtles and rats have beendescribed in detail elsewhere (Blanton, et al., 1989). A general discussion can be found inappendix A. In brief, whole cell patch recordings were carried out using sutter 1.5 O.D. X1.10 I.D. boro silicate non filament glass. Electrodes were pulled on Sutter model P-15horizontal puller. Patch solution contained (in mM) 15 NaCI, 110 KOH (pHed to 7.4,titrated with Methanesulphonic Acid), 10 Na Hepes, 11 EGTA (dissolved in 29 mM KOH),1 CaCl2, 2Mg-ATP, 0.3 GTP, final pH = 7.4, for the rat. Turtle patch solution was the rat57patch solution diluted by 10%. The osmolarity of the patch solution was 320 - 330 mosmol!Kg rat; 290 - 300 mosmol / Kg turtle.Data were acquired using an Axoclamp 2A amplifier connected to an Axolab 1100interface, which also served to generate current commands using the Pclamp suite ofprograms. Data were also independently digitized at 49 kHz and stored on videotape for off-line analysis. Criteria for using a patched cell included the GQ seal, positive going actionpotentials, and electrode impedance 45 MQ (most were less than 20 MQ) at the completionof the experiment. Although some cells were held for up to 3 hours, changes in cellconductance generally occurred within 30 to 45 minutes after breaking into the cell. Allmeasurements were made within 10 minutes of obtaining the whole cell configuration.Data analysisWhole cell conductance (Gw) and time constant (Tc) were calculated using depolarizingcurrent pulses of 500 mS duration (turtle) or 200 mS duration (rat) sufficient to elicit 6.2 \u00C2\u00B10\u00E2\u0080\u00A29(SE) mV change in membrane potential from the resting potential of the cell. All measureswere done on cells in a quiescent state to minimize leakage due to electrical activity. In brief,Gw was calculated from Ohm\u00E2\u0080\u0099s law:VmI/Gwwhere Vm is the change in membrane potential and I is the current. Tc was calculated byfitting the membrane charging curve to the equation:Y=A0+A1where Y is the voltage at any given time t, Ao is the offset, and Al is the maximum voltage,using the Clampfit feature of Pclamp. Specific membrane conductance (Gm) was calculatedfrom the equation:TcCm/Gmwhere Cm is the capacitance of the membrane per unit area which is assumed to be 1 tF /cm2. All computer fits of membrane charging curves showed a 0.9900 least squares residual(R) value or they were rejected.58The term Qio is used to refer to the effect of a 10 \u00C2\u00B0C temperature change on the valuesbeing measured, and for the purpose of this paper, it is defined as:Qio = Value (x + 10\u00C2\u00B0C) / Value (x \u00C2\u00B0C)where x is the variable being tested.Na-K-ATPase ActivityMethods for tissue dissection were identical to those described in Chapter 3 for theenzymatic analysis. Once the brain was dissected free, the cortex was removed and placed in9 volumes of 50 mM Tris-Ci (pH 7.4 at 4\u00C2\u00B0C) which contained 0.1 mMphenylmethylsulfonyifluoride. Tissue was homogenized 3 times, 10 sec. each interval with30 sec. in between each interval. Crude homogenates were then assayed for Na+K+ATPaseactivity without further processing.In brief, K dependent p nitrophenylphosphate hydrolysis is an expression of Na-KATPase activity (Swann and Albers, 1975). This indicator is used to circumvent problemsassociated with competition of ATP utilization characteristic of crude tissue homogenatescontaining mitochondria. Final incubation mixture contained 0.05 M Tris-HC1, pH 7.5, 5mM MgCl 10 mM Tris-p-nitrophenylphosphate, about 20 pg of brain homogenate.Reactions were conducted in duplicate in a final volume of 0.2 ml at 25\u00C2\u00B0C. Assays weredone in the presence of either 25 mM KC1 or 1 mM ouabain (control). Reactions wereterminated by the addition of 0.8 ml of 0.1 N NaOH. Final concentration (p-nitrophenol)after incubation period (10 mm.) was determined spectrophotometncally by absorption at4lOnm.ResultsData presented (intracellular and patch clamp methods) are based upon recordings of (n =38) turtle and (n = 20) rat pyramidal neurons resulting in a total of 58 cells being tested.Each patch clamp group represents an n of 10 recordings and the intracellular grouprepresents an n of 8 repeatedly measured cells. No neurons were repeatedly used for any59ABCFigure 15. Examples of membrane charging curves for turtle and rat pyramidal neurons.Turtle neurons (25 \u00C2\u00B0C)are represented in (A), intracellular recording techniques and (B),patch clamp techniques. A rat pyramidal neruon (25 \u00C2\u00B0C)is represented in (C) using patchclamping techniques. Refer to methods for a complete discussion of the fitting proceduresand analysis.TurtleA01.5 MV-64.51A1T= -8.41_______6OmSR= 141.81= 0.9986Turtle__60 mSR = 0.993860160-145- 1130-115- I___-I..\u00E2\u0080\u00A2 Te(mS)100- \u00E2\u0080\u00A2 =M.P.(-mV)y =G (pS/cm2)72-\u00E2\u0080\u00A2 =G(flS):60 I12-0 60 120Time of Anoxia (mm.)Figure 16. Conductance and membrane potential changes with anoxic exposure. Whole cellconductante (Gw) was calculated from Fig. 17 while specific membrane conductance wascalculated from the time constant (Tc) of the membrane charging cureve. Membranepotential (M.P.) was also recorded for the duration of the experiments. See text for furtherfigure details. Data illustrated are means of n = 8 pyramidal cortical neurons from turtle.Data are illustrated with SE bars.61-20-25-35V (mV)Figure 17. Current (1)-voltage (V) plotfor 8 repeatedly measured turtle pyramidal neuronsover a time course of 120 mm. of anoxia (0.1 to -0.5 by 0.05 nA steps). Numbers inparentheses are n of cells for that point set. Data are illustrated with SE bars. The axies crossat the cell\u00E2\u0080\u0099s resting potential (R.P.)I(nA)-0.5 -0.4 -0.3R.P.(0,0)-0.2 -0.1Normoxia0.1p 60 minutes anoxia120 minutes anoxia-5A-10-15(8)(8)(8)(8)(8)(7) (7)(7)(7)-30-4062Table 3. Turtle cortical pyramidal cell patch clamp valuesNormoxia Normoxia Anoxiaafl 15 \u00C2\u00B0C \u00C2\u00B1 SE fl 25 \u00C2\u00B0C \u00C2\u00B1 SE fl 25 \u00C2\u00B0C \u00C2\u00B1 SEParameterM.P. (mV) 10-75.0 \u00C2\u00B1 1\u00E2\u0080\u00A20d 10 -72.8 \u00C2\u00B1 05d 9 .740 \u00C2\u00B1 25dGi (nS) 10 1.84 \u00C2\u00B1 021bc 10 2.73 \u00C2\u00B1 0.36C 10 2.35 \u00C2\u00B1 0.21CTc (mS) 10 423.5 \u00C2\u00B1 28\u00E2\u0080\u00A24be 10 200.7 \u00C2\u00B1 19\u00E2\u0080\u00A23C 10 209.3 \u00C2\u00B1 23.5cGm (jS / cm2)e 10 2.46 \u00C2\u00B1 017be 10 5.56 \u00C2\u00B1 0.67c 10 5.37 \u00C2\u00B1 o.61ca Turtle cortical slices were held in an anoxic chamber at 22\u00C2\u00B0C for 6-9 hrs. prior torecordingb Significantly different (P 0.05; Newman- Keuls test) from the control (25 \u00C2\u00B0C ) measure.C Significantly different (P 0.05; Newman - Keuls test) from rat (35 and 25\u00C2\u00B0C) corticalvalues (Table 2).d Significantly different (P 0.05; Tukey HSD test) from rat (35 and 25\u00C2\u00B0C) cortical values(Table 2).C Assuming a 1 jsF / cm2 capacitance.Refer to text or Table of Abbreviations for an explanation of table abbreviations.63Table 4. Rat cortical pyramidal cell valuesn 25 \u00C2\u00B0C \u00C2\u00B1 SE fl 35\u00C2\u00B0C \u00C2\u00B1 SEParameterM.P. (mV) 8-57.5 \u00C2\u00B1 29d 10 -55.4 \u00C2\u00B1Gw (nS) 10 4.08 \u00C2\u00B1 057C 10 7.56 \u00C2\u00B1 1.00T (mS) 10 46.5 \u00C2\u00B1 4.4\u00E2\u0080\u0099 10 25.0 \u00C2\u00B1 2.0kGm (J4S I cm2 10 23.5 \u00C2\u00B1 2\u00E2\u0080\u00A24c 10 42.4 \u00C2\u00B1 3.5a Assuming a 1 pcF / cm2 capacitanceb Significantly different (P 0.05; independent t test) from the 25\u00C2\u00B0C valueC Significantly different (P 0.05; Newman - Keuls test) from turtle (15 and 25\u00C2\u00B0C)cortical values (Table 1)d Significantly different (P 0.05; Tukey HSD test) from turtle (15 and 25\u00C2\u00B0C) corticalvalues (Table 1)Please refer to Table of Abbreviations or text for an explanation of table abbreviations64Table 5. Effect of temperature on membrane ion leakageTurtle Rat15-25\u00C2\u00B0C 25-35\u00C2\u00B0CMeasured Qio2.3 1.81.5 1.9Average 1.9 1.9Values obtained from Tables 3 and 4Refer to Table of Abbreviations or text for an explanation of table abbreviations6525-A. 25\u00C2\u00B0C rat! 25\u00C2\u00B0C turtleB.35\u00C2\u00B0Crat/25\u00C2\u00B0Cturtle20. C. 37\u00C2\u00B0C rat /25\u00C2\u00B0C turtleD. 37\u00C2\u00B0C rat / 15\u00C2\u00B0C turtle015010-5.I\u00E2\u0080\u0094 I \u00E2\u0080\u0094 IA B C DTemperature ComparisonFigure 18. Conductance ratios (Gm) for rat cortical pyramidal neurons vs. turtle corticalpyramidal neurons. Values are obtained from Tables 3 and 4 except for the rat 37\u00C2\u00B0C valuewhich is calculated fro the rate 35 \u00C2\u00B0C assuming a Qio of 1.9 (Table 5).66Table 6. Na-K-ATPase activity in Turtle and Rat CortexTurtle Rat Magnitudej4moles/g!min J4moles!g/min DifferenceTemperature25\u00C2\u00B0C 3.52 \u00C2\u00B1 0.12 8.03 \u00C2\u00B1 0.5437\u00C2\u00B0C 19.3C 55ba Rat 25\u00C2\u00B0C! Turtle 25 \u00C2\u00B0Cb Rat 37 \u00C2\u00B0C / Turtle 25 \u00C2\u00B0CC Assuming a Q 10 of 267patch clamp group. Examples of raw data traces (of which both Gm and G were derived)are illustrated by Fig. 15a,b,c.Intracellular RecordingWe tested whether ion channels were down regulated with anoxia ( 120 mm.) usingintracellular recording methods. This method allowed the continuous monitoring of Gm , Tcand Gw. Results demonstrated that the turtle pyramidal neuron did not significantly changein any measured parameter (Te, Gm, Gw, M.P.) with anoxia (P> 0.05; repeated measuresdesign over time, tested for linear or quadratic fit). Two separate measures of conductancewere performed, Gw and Gm (Fig. 16). Gw indicates the conductance of the whole cell. Ifthe area of the cell were known, conductance per unit area could be calculated (Gm). Surfacearea of the cell is unknown, but 0m can be derived through a simple equation (see methods).The calculation is based on a 1 J4F/cm2capacitance of the membrane.A third observation of anoxic changes in resistance is obtained through the currentvoltage plot (Fig. 17). The slope of the line represents voltage - dependent changes in R(whole cell resistance; 1/R = Gw). As can be seen in Fig. 17, the voltage dependence ofGw does not significantly change over the course of 120 mm. of anoxia (P> 0.05, repeatedmeasures design over time, tested for linear or quadratic fits). Thus, neither steady state norvoltage- dependent ow change during anoxia.Patch Clamp ResultsIntracellular recording methods were used for short term anoxia studies because itallowed the continuous impalement of a cell, and thus, continuous monitoring of 0m Or Ow.However, based on previous experiments, we have found that continuous impalement willnot yield accurate results after several hours due to eventual electrode clogging ordegeneration of the cell. Thus, the use of whole cell patch clamp methods to comparepopulations was used. The relevant electrophysiological properties of these populations aredetailed in Tables 3 and 4. To assess differences in ion leakage between turtle and ratcortical neurons, Gm was calculated for both species at 25 \u00C2\u00B0C. Conductance values indicate68a 4.2 fold higher Gm in the rat neuron compared to the turtle. Comparison of conductance atmore physiological temperatures (35 \u00C2\u00B0C, rat; 15\u00C2\u00B0C, turtle) indicated a 17 fold increase inG for rat pyramidal neurons. Although not directly measured, Gm for the rat neurons atphysiological temperature (37 \u00C2\u00B0C) was inferred from the measured conductance Qio of 1.9(25-35\u00C2\u00B0C, Table 5). The calculated Gm at 37\u00C2\u00B0C for the rat neuron was 53.58 uS / cm2indicating a 22 fold more conductive membrane for the rat pyramidal neurons compared toturtle pyramidal neurons (15 \u00C2\u00B0C). The conductance ratios between turtle and rat pyramidalneurons are summarized in Fig. 18.To assess whether ion leakage further decreases with prolonged anoxia, turtle slices wereincubated in anoxic aCSF for 6-9 hours. Two separate and independent measures ofconductance were done (Gw and GJ for reasons discussed above. The results reported inTable 3 indicate no significant change in Ow or Gm with 6-9 hi\u00E2\u0080\u0099s. of anoxia supporting andextending intracellular recording results.Current evidence indicates that anoxic survival of the turtle is highly temperaturedependent (Ultsch, 1985). One question which remains unanswered is whether turtle neuronsexpress a large Qio value for ion conductance at lower temperatures. TableS summarizes theQio values for conductance in turtle and rat cortical neurons. Surprisingly, an average Qioof 1.9 was measured for both turtle (15 - 25\u00C2\u00B0C) and rat (25-35 OC) These results indicatethat Qio for leakage channels may be conserved across various temperature regimes as wellas species.Interestingly, temperature did not significantly change membrane potential in either therat or turtle cortical cell populations even though conductance of these cells is changed(Table 1 and 2). Additionally, one might suspect a change in membrane potential if ionchannels were down regulated during anoxia. However, there were no significant changes inmembrane potential with anoxia, in turtle cortical neurons supporting the findings of nochange in either Gm or G with anoxia.Na-K-ATPase Activity69In order to differently address the concept of lower resting ion leakage in the turtle braincompared to the rat during normoxia, Na+K+ATPase maximal activity was measured.Because this method uses an artificial substrate (p-nitrophenylphosphate), the results shouldbe interpreted not as absolute values, but values relative to each other. Table 6 illustrates theNa+K+ATPase activity in cortical homogenates of both the turtle and the rat. The resultssuggested that there is about a 2.3 fold lower activity in the turtle brain compared to the rat at25\u00C2\u00B0C. When corrected for temperature (Qio of 2) for the rat brain (37 OC), the dataindicated a difference of 5.5 fold compared to the turtle cortex (25 \u00C2\u00B0C). These results furthersupport the patch clamp recordings of lower ion leakage, and hence, ion pumpingrequirements in the turtle cortex.DiscussionThe channel arrest hypothesis is based on the premise that as the conductance of a cell(G increases so must the rate of ion pumping if ion homeostasis and membrane potentialare to be maintained. This prediction is experimentally supported (Pastuszko et al., 1981;Scott and Nicholls, 1980). The concept can be applied not only to anoxia, but also tonormoxia. If one cell is more conductive than another during normoxia, then ion pumpingdue to this increased conductance may be increased to maintain ion homeostasis (Hochachkaand Guppy, 1987). Based on Gm (Fig. 18) and Na-K-ATPase activity (Table 6), theseresults support reduced passive ion leakage during normoxia as one mechanism utilized bythe turtle neuron to reduce ATP expenditure. The enzymatic Vmax measures, althoughsupporting a reduced ion leakage in the turtle brain, only support 60% of the observed 4.2fold difference from Gm ratios suggesting that further down regulation of ion pumpingactivity may occur in vivo in the turtle brain (Fig. 18).Conductance ratios reported here (25 \u00C2\u00B0C) between the turtle and the rat are slightly higherthan K+ conductance ratios reported for similar sized nondiving reptiles (Amphibolorusvitticeps) vs. the rat (conductance done at 37\u00C2\u00B0C) using tracer methods (4.2 vs. 3.6respectively) (Else and Hulbert, 1987). Additionally, oxygen consumption ratios in the70presence of ouabain under the same temperature conditions (37 \u00C2\u00B0C) suggests ion pumpingdifferences of 4 fold between the rat and A. vitticeps (Else and Hulbert, 1987) compared tothe 2.3 fold difference reported here in maximal activities of Na-K-ATPase (25 \u00C2\u00B0C). Boththe K+ conductance studies and the ouabain studies compare well with the \u00C2\u00B0m ratiosreported here for the turtle vs. rat cortex at the same temperature (25 \u00C2\u00B0C; 4.2) suggesting thatthe maximal activities of Na+K+ATPase may not accurately reflect absolute ion pumpingdifferences because of in vivo Na+K+ATPase regulation (Rossier et at., 1987). However,the maximal activities of this enzyme do support large pumping capacity differences betweenthe two species.It is tempting to estimate the energy (ATP) savings achievable based on the lower Gm forthe turtle cortex; however, this is not yet possible. First, the specific passive ionconductance contributing to this savings are not yet known. Second, the exact energyexpenditure due to ion leakage independent of electrical activity is not well understood.Studies on canine brains in vivo suggest that ion pumping due to passive ion leakageconsumed more than 40 % of resting ATP turnover (Astrup et a!., 1981) suggesting asubstantial energy savings. Further speculation on the metabolic energy savings will bediscussed in Chapter 5.The calculations for Gm are based on two assumptions. First, both biological membranesdisplay a 1 .tF / cm2 capacitance. Careful studies of both ectotherms and enclothemis from avariety of cell types yield this value (Brown et at., 1981) suggesting that Cm may be abiophysical constant (Hille, 1992). Since Cm is a function of the membrane composition,and since membrane structure and composition is conserved, it is reasonable that Cm wouldalso be conserved. Secondly, the assumption of a Qio of 1.9 for the rat brain between 35 -37 \u00C2\u00B0C is made. We have measured the Qio for membrane conductance in the turtle neuronbetween the temperatures of 15-25\u00C2\u00B0C and for the rat neuron between 25-35\u00C2\u00B0C (Table 3),and have extrapolated the latter values to 37 \u00C2\u00B0C.71The Qio for both tissues is approximately the same (1.9) (Table 3) which is similar to thewhole body metabolic Qio for both species for these temperature ranges (Funk and Milsom,1987). The conservation of the conductance Qio between animals and temperatures suggeststhat the Qio for ion channels may be conserved across channel types as well as animals.Single channel studies from a variety of channel types from both ectotherms and endothermsreport Qio values ranging from 1.0 to 2.5 with most values between 1.3 - 1.6, close to theaqueous diffusion Qio of 1.3 (Hille, 1992). However, conservation of the conductance Qiomay hold true only for leakage channels. The Qio for this study is an average between Gmand 0w The average is used since these measures although calculated independently of eachother should change both qualitatively and quantitatively in parallel with temperature. Thisparallelism appears to be true for the rat cortical neuron but quantitatively deviates for theturtle neuron. Several reasons could possibly explain this deviation since these two measuresare calculated independently of each other. Changes in capacitance, cell size, soma todendritic conductance and bleb formation (Milton and Calswell, 1990) may all influence Gwdifferently from Gm with respect to Qio.In addition to large inherent leakage differences between rat and turtle membranes, thechannel arrest hypothesis predicts a decrease in Gw and G with anoxia. Results usingintracellular recording methods over short term anoxia indicated no change in Gm or 0w(Fig. 16, 17). Data from Chih suggests that K leakage in viva was suppressed by anoxia(Chih eta!., 1989). Recent cortical slice data indicated that [Ca]i accumulation during longterm anoxia (180 mins) is slower compared to short term anoxia (5 mins.) after the additionof iodoacetate (glycolytic inhibitor) (Bickler, 1992a). However, specific leakage processeswere not measured in either of these papers, and thus, the lowered leakage observed could becaused from a reduction in any leakage process including electrical activity. Due to theabove observations as well as limitations of the microelectrode technique, we measuredwhole cell conductance using patch clamping techniques. The advancement of whole cellpatch clamp methods in slices provides a way of easily comparing populations of cells while72ensuring that membrane electrode seals are not significantly contributing to the apparent cellconductance due to the GQ seal formation between electrode and cell membrane. The valuesobtained for Tc and Gw were greater than those obtained with intracellular recordingtechniques. Thus, ion leakage around the microelectrode may significantly contribute to thevalues of Tc and Gw, masking small conductance changes. However, the results reportedhere for long term anoxia (6- 9 hrs.) supported intracellular recording techniques.Both turtle and rat pyramidal neurons have inwardly rectifying K+ channel (Connors andKreigstein, 1986; McCormick, Ct al., 1985). Thus, one method for changing cell conductanceis to change membrane potential. As a result, channel arrest could be achieved by simplydepolarizing the cell. However, no noticeable change in membrane potential occurred duringanoxia in turtle neurons (Table 1) supporting earlier results with intracellular recordingtechniques and short term anoxia (Chapter 1).The energy conservation mechanism predicted by the channel arrest hypothesis is notdirectly due to ion channels closing, but by a decrease in ATP-dependent ion pump activityresulting from down regulation of ion channels. We have tested one leakage process (leakchannels). However, ion leakage can result from a variety of other processes. Currentevidence supports both a reduction in ion leakage (Bickler, 1992a; Chih, et al., 1989) and inelectrical activity (Feng, et al., 1988; Sick, et al., 1982) in the turtle brain with anoxia. Ourresults support that the concept that the turtle brain maybe spending less energy onmaintaining ion homeostasis in the normoxic state compared to the rat, but that further downregulation with anoxia is not measurable. Accumulating evidence supports down regulationof electrical activity as a more likely scenario for metabolic depression in the turtle brain.73CHAPTER 5: A THEORETICAL APPROACH TO ANOXIA TOLERANCE- ACONCLUSIONPrefaceThe goals for this final chapter are three fold. First, this chapter serves to unite all of thepreceding material into a unified theory of anoxia tolerance. Second, this chapter will serveas a mini review of the preceding material especially that material which relates to thisunified theory. Finally, it is hoped, this chapter will serve to generate curiosity and insightwhich may lead to a better fundamental understanding of the events surrounding both anoxiatolerance and intolerance. Parts of this chapter were excerpted from C. J. Doll (In SurvivingHypoxia, CRC Press, pgs. 389-400, 1993).IntroductionThe second law of thermodynamics states that natural processes move towards a state ofgreater disorder (increasing entropy, AS > 0). Living organisms remain in an unchangingstate of entropy by consuming energy. In the case of a neuron, glucose is consumed andenergy is provided to maintain cellular homeostasis via ion pumping, molecular repair,synthesis etc., thus, AS 0 for the organism. However, if the cell cannot provide enoughenergy to maintain integrity, AS will increase. The results may be catastrophic (Chapter 2).Thus, the emphasis of this thesis (the importance in maintaining energy balance for the cell)is really a restatement of the second law of thermodynamics. In this final chapter, twoquestions will be explored in detail. First, how is the turtle cortical neuron able to maintainmembrane integrity during anoxia, and second, why is the rat cortical neuron (a cell whichperforms similar functions as the turtle neuron) more vulnerable to these same insults?An Interpretive ModelThe results of this thesis in part can be summed up by Fig. 19. This figure represents aunified theory of why the turtle brain is able to survive anoxia and in part why themammalian brain cannot. It is divided into two parts: the turtle brain (left side) and themammalian brain (right side).74I ANOXIA/ISCHEMIAICytosolic [ATPJ Falls(Turtle Brain)Ipp_p_PASTEUR EFFECT!LOW METABOLIC RATE!METABOLIC ARREST DEFENSEMECHANISMS INITIATEDPASTEUR EFFECT!HIGH MEABOUC RATEJNO ARREST MECHANISMSMAINTENANCE OFCYTOSOLIC [ATPI4.ION GRADIENTSMAINTAINEDANOXIC SURVIVALUNTIL ENERGY DEPLETED4REOXYGENATION& SURVIVAL(+2& Na,KATPase FAILURE4,CYTOSOLIC [(+2]INCREASES4,Ka IIOPEN4,CYTOSOLIC EFFLUXOF K+CYTOSOUC [+2]ATP DEPENDENTK CHANNELS OPEN4CYTOSOLICKEFFLUX4CYTOSOLIC [ATP]DECREASES FURTHERION GRADIENTSDIMINISH+PHOSPHOLIPASESACTIVATED+MEMBRANEHYDROLYSIS4,CELL DEATHIFigure 19. Possible scenario for the degeneration and survival of the turtle and rat neuron.See text for further details.(Mammalian Brain)FURTHER DECREASEIN CYTOSOLIC [ATP}I MASSIVE RELEASE IOF I ATP DEPENDENTINEuR0TRANsMrrTERs INONSELECTIVEh4%..-INCREASES FURTHERCATION CHANNELSOPEN++2 NONSELECHVE4)CATION CHANNELS OPVoltage-gated IonChannels Open75Mammalian BrainAs discussed in Chapters 1 and 2, a major difference between the turtle and rat brain is intheir ability to maintain [ATP] dunng the anoxic/ischemic insult. I have hypothesized thatthe declining [ATP] is the trigger which directly or indirectly causes the anoxic/ischemiccascade of events observed in the mammalian brain. There are currently several lines ofindirect evidence supporting this position. First the celPs ability to maintain resistance andmembrane potential is inversely related to the degree of metabolic inhibition (Reiner, et al.,1990; Chapter 2). Second, temperature plays a direct role in protecting ion gradientssuggesting that the lower the metabolic rate (the slower ATP is depleted) the longer iongradients are maintained (Moms et at., 1991; Okada eta!., 1988a; Chapter 2). Third,manipulation of the intracellular or extracellular environment to increase energy richcompounds intracellularly, prolongs the maintenance of ion homeostasis (Caldwell, et al.,1960; Hansen, 1978; Kass and Lipton, 1982; Upton and Whittingham, 1982; Okada, et al.,1988a). Fourth, there is a direct relationship between loss of electrical activity and [ATP](Lipton and Whittingham, 1982; Yamamoto and Kurokawa, 1970), as well, bathingmammalian brain slices with creatine to increase intracellular [CrPJ (thus, defending [ATP]j)prolongs electrical transmission (Lipton and Whittingham, 1982). Although the fourth pointis controversial, with some studies showing changes in synaptic activity before changes in[ATP]i are detectable (Duffy eta!., 1972; Schmahl eta!., 1966; Siesjo and Nilsson, 1971),studies in the hippocampus suggest that whole hippocampal slice [ATP] may show littleattenuation from control, but when the molecular layer (an area which contains the synapsesof the pathway) is examined separately, changes in [ATPJ1 are measurable prior to changes inelectrical activity (Lipton and Whittingham, 1984 for review). These data would suggeststudies which do not detect a substantial fall in [ATPJ before electrical changes may not bemeasuring ATP in appropriate compartmentalized areas (synapses, dendrites, and axontenninals).76Initially, with an anoxiclischemic insult, there is a large Pasteur effect observed for themammalian brain (discussed in Chapter 3). Unfortunately, due to the high metabolic rate (9-10 fold higher than the turtle brain at 25 OC; Chapter 3) and the comparatively low amount ofglycolytic enzymes (Chapter 3), the mammalian brain is unable to have cellular ATPproduction meet demand resulting in a loss of [ATPJi (as discussed in Chapter 2). Fig. 19depicts the importance of immediate defense mechanisms (see below) to maintain [ATP]1. Inthe mammalian brain an unsuccessful attempt is made to maintain energy balance despite alarge Pasteur effect. Unsuccessful, the mammalian brain enters a cascade of events whichleads ultimately to neuronal death (Fig. 19).(For the next few sections I will refer the reader to review Chapter 2 Introduction where amore complete description and references are given for the discussed material. Additionally,I have referenced Chapter 2 material where the applicable data may be found.)Both extracellular observations and the hippocampal slice studies (as discussed inChapter 2), as well as the results from Chapter 2 support the hypothesis that the first visiblereaction to anoxialischemia is a hyperpolanzation and an increase in [K+]0. Two possiblepathways have been hypothesized for this observed increase in [K]0 (Krnjevic, 1993;Chapter 2): (i) opening of KATp channels, or (ii) opening of Kca channels. With regards to(ii), the activation of these channels by ATP would be indirect. As [ATP]i decreases,[Ca2]j homeostasis mechanisms (ion pumps) fail resulting in an increase in [Ca2]j which,in turn, causes activation K channels (Fig. 19).Talbutamide (KATp channel blocker) does not inhibit the initial anoxic hyperpolarizationin the hippocampal CA 1 region (Leblond and Krnjevic, 1989), and several lines of evidencenow support Kca channels involvement in the hyperpolarization in the CA 1 region(Krnjevic, 1993, also see Chapter 2). Interestingly, studies in the CA3 hippocampal regionwhere a higher concentration of glibenclamide receptors are found (thus, indicating a greaterconcentration of KATp channels) demonstrate that the slow phase of the initialhyperpolarization is blocked by glibenclamide (Mourre et al., 1989); these results imply that77a combination of events cause the anoxic hyperpolarization (as supported by Fig. 19) and thatevents may vary depending on brain region.As [Ca2]j increases [ATP]1decreases, three mechanisms for further depolarization andconductance increase are hypothesized. The mechanisms may be synergistic as indicated bycrossover arrows in Fig. 19. As previously discussed (Chapter 2), one possible hypothesisfor the massive and sudden diminishment of the ion gradients is neurotoxicity. Thishypothesis predicts a massive release of excitatory neurotransmitters like glutamate whichmay cause the sudden diminishment of ion gradients similar to that observed in Fig. 5. Thishypothesis has gained support, in part, from observations that excessive application ofglutamate does give a similar response to that observed for ischemia and pharmacologicalischemia (Choi, 1988a; Rothman, 1985; Wood and Reiner, 1990), and from the observationthat there is a sudden release of neurotransmitters corresponding to the ischemic insult (seeChapter 2 for a review). The mechanism for this release is consistent with the hypothesisthrough an increase in [Ca+2]j. However, two other channel activations may individually orsynergistically aid in the anoxic/ischemic neuronal collapse: (i) Ca2 - activated non -selective cation channels (ii) ATP - dependent non - selective cation channels. Both channeltypes have been identified in neuronal tissue (Ashford, et al., 1988; Partridge and Swandulla,1988), and therefore, could play a role in the anoxic I ischemic depolarization. Recentstudies suggest that preincubation with glibenclamide (KATp channel antagonists)significantly reduces the amplitude and rate of anoxic [K+]0 in the hypoglossal nucleus inadult rats. Preliminary observations in hippocampus suggest that glutamate antagonists alsoreduce the rate of depolarization during pharmacological anoxia in the rat hippocampus, butas in the glibenclamide study above, the neurons still depolarize (Wood and Reiner, 1990)supporting a combination of events being involved in the anoxiclischemic depolarization.Verification of this hypothesis awaits further detailed studies. Once membrane potential fallsbelow threshold, the activation of voltage-gated ion channels (Na+, Ca+2and K+) can thenparticipate in the observed ion gradient diminishment.78The specific cause of cell death is not well understood. In Fig. 19, I have suggestedmembrane hydrolysis by the activation of phospholipases as the most likely scenario butpermanent damage could also be caused by free radicals (including nitric oxide), pH, cellblebbing, or cytoskeletal destruction (Choi, 1988b; Haddad and Jiang, 1993; Hochachka,1987 for reviews). Although the actual cause of cellular death is poorly understood, mostevents up to and including the large conductance and ionic changes leading to the membranedepolarization are reversible, and permanent cellular damage appears to occur after the lossof intracellular ion homeostasis (Hansen et at., 1982; Kirino et al., 1985; Leblond andKrnjevic, 1989; Siemkowicz and Hansen, 1981). It is important to note that an understandingof neuronal functional collapse is considered central for this thesis because it is functionalcollapse rather than neuronal death per se that is ultimately responsible for anoxic death ofthe whole organism.Turtle CNS ResponseThe left half of Fig. 19 is concerned with the turtle brain. There are two importantaspects to this figure. First, the ability of the turtle cortical slice to maintain [ATPJi withanoxia, and second, an arrow linking the turtle to the mammalian side in the event that[ATP]i is not maintained. However, the turtle cortical slice is capable of maintaining [ATP]through a combination of 3 central mechanisms (discussed and supported by studies inChapter 3): (i) an enhanced glycolytic capacity, (ii) a low normoxic metabolic rate, (iii) theability to further depress metabolism with anoxia. Support for the linkage of the turtle side tothe mammalian side is based on the intracellular observations that the turtle corticalpyramidal neuron could not significantly withstand pharmacological ischemia any better thanthe rat pyramidal neuron (Fig. 3) with the turtle brain indicating similar conductance andmembrane potential changes with pharmacological ischemia (Fig. 5). Additional support forthis linkage comes from in vivo studies which demonstrate a very rapid rise in [K+]0 in theturtle brain similar to that observed in the rat brain with ischemia and iodoacetic acid (Sick,et al., 1982; Xia eta!., 1992). There is very little information regarding channels, channel79types or densities in the turtle brain; most emphasis is on voltage dependent channels whichappear to occur at a lower density in the turtle brain compared to the rat brain (Edwards et al.,1989; P\u00C3\u00A9rez-Pinz\u00C3\u00B3n, et al., 1992c; Suarez, et al., 1989; Xia and Haddad, 1993). Identificationof KATp channels in the turtle brain has been confirmed (Jiang, et al., 1992; Xia and Haddad,1993) with the highest concentration occurring in the turtle cortex. The presence of thesechannels is consistent with the hypothesis that the mechanisms of pharmacological ischemiadepolarization occurs through similar mechanisms in both species.Metabolic RateThis thesis has presented the hypothesis that a low metabolic rate is critical for anoxiatolerance. Interestingly, both the turtle and the rat neuron have similar physiologicalfunctions (electrical activity, ion pumping, biosynthesis, cellular maintenance, etc.). Chapter4 was an initial attempt to understand mechanisms which may be down regulated in the turtlecortex during both normoxia and anoxia to conserve ATP utilization. In this section, a\u00E2\u0080\u00A2theoretical examination of metabolic rate will be explored in an attempt to partially explainthe observed brain metabolic rate discrepancies between the turtle and the rat.Chapter 3 demonstrated a 9 to 10 fold difference in metabolic rate between guinea pigand the turtle cortical slice preparation (Table 7, no. 6 vs. 7). Additionally, deoxyglucosestudies have indicated approximately the same difference (Table 7, average of 1 and 2 vs. no.3). Fig. 20 illustrates an explanation for the metabolic rate differences between the corticalslice preparations of these two species. Because this argument is theoretical, the units on they axis are arbitrary, and thus, reflect the 10 fold metabolic difference between the tissues (Fig20a,d).Because the metabolic rate measures have been performed at physiological temperatures,metabolism must be corrected for temperature. Two studies which examined the effect oftemperature on glucose utilization in the mammalian cortex (Table 7, no. 4 and 5) reported aQio of 2.7 and 4.32 yielding an average Qio of 3.5. Thus in (B) of Fig. 20 the metabolic80Table 7. Glucose consumption measuresTemperature Glucose*Animal Preparation range (\u00C2\u00B0C) pmoles/g/min Qio Reference1 Rat Whole Brain 37 O.45a (Hawkins et at., 1974)(in vivo)2 Rat Whole Brain 37 060a (Crane et at., 1978)(in vivo)3 Turtle Whole Brain 25 o.053a____(Suarez, et al., 1989)(in vivo)4 Cortex 37.4 1.15 (MeCulloch et at., 1982)(in vivo)31.8 075a 2.75 Cortex 35.8 O.93a (Ito et a!., 1990)(in vivo)27.7 O.27a 4.326Guniea Pig Cortical 38 O.42 (Rolleston andSlice Newsholme, 196Th)7Turtie Cortical 25 0048C Chapter 3Slicea Technique used was radio labeled glucose or glucose anologue.b Technique used was direct measurement of glucose consumption.C Technique used was calorimetry and then converted to glucose consumption assuming2820 kJ I mole glucose (Gnaiger and Kemp, 1990).d Corrected for lactate production and glycogen catabolism.* Numbers superscripted beside animals are for in-text reference8110_______________________10 A. Rat cortical slice metabolic ratea) (37\u00C2\u00B0C)a)B. Rat cortical slice metabolic rate(25 \u00C2\u00B0C) assuming a Qio of 3.5\u00E2\u0080\u0094 C. Rat slice metabolic rate (25 \u00C2\u00B0C)0with turtle neuronal leakage- I). Turtle cortical slice metabolic rateI (25\u00C2\u00B0C)C.)\u00E2\u0080\u0094 6.0Cunits4.Q2.406 N0C) \u00E2\u0080\u0094__1.0 1.0o 4.0__ __Units 0.6 I Turt1e10\u00E2\u0080\u00A2-A B C DFigure 20. Adjusted metabolic rate for the mammalian cortical slice.82rate of the rat cortical slice has been metabolically adjusted for this temperature based on aQ 10 of 3.5 yielding a value of 2.4 units (Fig. 20b).For the rat cortical slice 40% (4 of the 10 units) of the metabolic rate has been allotted toion pumping (Fig 20a). Two independent studies have suggested this percentage of energymetabolism for ion pumping independent of electrical activity in vitro (Whittam, 1962) andin vivo (Astrup, et a!., 1981). In Chapter 4, the change in conductance with temperature wasmeasured for the rat cortical neuron (Q1o(35- 25\u00C2\u00B0C) = 1.9; Table 5). When the rat neuron iscorrected for the 12\u00C2\u00B0C temperature change, the adjusted energy allotment for ion pumping is1.8 units of the total 2.4 units (Fig. 20b). Because the Qio for total energy metabolism (3.5)is larger than the Qio for ion leakage (1.9), the percentage of energy consumed by ionpumping has increased from 40% to 75% of the total allotted energy utilization of the slice,independent of electrical activity (Fig. 20c).In Chapter 4, the difference in conductance between the turtle and the rat pyramidalneuron at the same temperature (25 \u00C2\u00B0C) was measured. The turtle neuron expressed 4.2times less conductance than the rat pyramidal neuron (Fig. 18). Thus, the 1.8 units which hasbeen allotted to the rat neuron for ion pumping at 25\u00C2\u00B0C can be divided by 4.2 (= 0.4 units).Thus, both the turtle and the rat now have the same energy allotment for ion pumping. Whenthe 0.4 unit for ion pumping is added to the 0.6 unit (other metabolic processes), the finalmetabolic rate of the rat cortical slice is identical to the turtle cortical slice (1.0 unit). Sinceboth the total metabolic rate (1.0 unit) and the ion pumping energy allotment (0.4 unit) arethe same for both species, then the metabolic rate allotted to other metabolic processes mustbe the same for both species (0.6 units). Thus, when the rat cortical slice is adjusted to havethe same degree of ion pumping as the turtle cortical slice at 25\u00C2\u00B0C, the metabolic rate for thetwo tissues is identical.These results support four concepts for the turtle and the rat neuron in the absence ofelectrical activity: (i) the rat neuron, when corrected for both temperature and ion leakageconsumes the same amount of ATP as the turtle neuron, (ii) the metabolic rate difference (at8325\u00C2\u00B0C) between the turtle and the rat cortical slice (2.4 vs. 1.0, Fig 20b vs. d) is due to theamount of ion pumping rather than differences in other metabolic processes; and (iii) theturtle neuron is spending approximately 0.4 of the 1.0 unit on ion pumping (40%) and 0.6 ofthe 1.0 unit (60%) on other metabolic processes which is identical to the estimated metabolicrate allotted to the rat cortical slice at its physiological temperature (37 \u00C2\u00B0C).Ectothermy vs. EndothermyAlthough the emphasis for anoxia tolerance in this thesis has centered on metabolism andits regulation, a low metabolic rate is not the only component to anoxic survival as has beendiscussed throughout this section. All three of the components, (1) low metabolic rate, (ii) theability to further suppress metabolism with anoxia, and (iii) a large glycolytic capacity, mustwork synergistically. The requirement for such a integrative or collaborative effect may, inpart, explain the wide tolerance to anoxia observed among ectotherms, and why being anectotherm does not necessarily imply being anoxia tolerant. Tolerance among ectothermicvertebrates is highly variable with some amphibians (Penney, 1987; Sick and Kreisman,1981), and fish (Doudoroff and Shumway, 1970; Randall, 1982; Ultsch, 1989 forreviews)having similar responses as the mammal to anoxia while others such as carp, turtle,alligators, and snakes (Bennett and Dawson, 1969; Ultsch, 1989 for reviews) possess variableanoxic windows. Even among turtle species anoxia tolerance varies (Ultsch, 1985).Ectotherms have lower metabolic rates than similar sized mammals (Schmidt-Nielsen, 1984)and presumable lower brain metabolic rates (Mink, Ct al., 1981), but a low metabolic rateonly in combination with metabolic and glycolytic regulation will provide anoxic protection.Thus, ectothermic species not expressing anoxia tolerance may not fully exhibit one of thethree requisite adaptations. Confirmation of this hypothesis awaits further studies of brainmetabolism and glycolytic regulation in other ectotherms (anoxia intolerant vs. anoxiatolerant). One study which did examine brain metabolic rates in different ectothermsindicated that the turtle brain (Pseudemys scripta) expressed half the metabolic rate comparedto the fish (Carassius auratus) and frog (Rana pipiens) (McDougal, Ct al., 1968).84Unfortunately, this thesis is written at a time when very little is understood about energybudgets of neurons. Although an attempt has been made at measuring one mechanism formetabolic depression (reduced ion leakage; Chapter 4) we have not identified any metabolicarrest processes in the cortical slice. Experiments with turtle hepatocytes indicate a markedreduction in ion leakage with anoxia (Buck et al., In Press). As discussed in Chapter 4,current evidence supports a marked reduction in electrical activity in the turtle brain.However, studies have demonstrated that 02 consumption is reduced by only 5 % with theinhibition of voltage dependent Na channels in cortical slices (Okamoto and Quastel, 1972).We have observed a substantial metabolic depression (30-40 %) in the anoxic turtle corticalslice (see chapter 3). Thus, other metabolic process must be depressed in the turtle slice toaccount for the measured metabolic depression. As a final note, this study focused on the invitro preparation. Changes which occur as a result of the intact tissue have not beendetermined. It is possible that leakage channel down regulation may occur in vivo. Recentstudies have demonstrated a reduced Ca+2 influx in response to glutamate when the in vitrocortex of both turtles and rats was exposed to anoxic turtle plasma in the aCSF (Bickler andGallego, 1993). Therefore, additional studies may be necessary to determine whetherleakage channels are down regulated in response to anoxia in the turtle cortex.85LITERATURE CITEDAshford, M. L., N. C. Sturgess, N. J. Gardner, and C. N. Hales. Adenosine-5-triphosphate-sensitive ion channels in neonatal rat cultured central neurons. Pfluegers Arch. 412: 297-304, 1988.Astrup, J., P. M. Sorensen, and H. R. Sorensen. Oxygen and glucose consumption related toNa-K transport in canine brain. Stroke 12: 726-730, 1981.Aw, T. Y., and D. P. Jones. Cyanide toxicity in hepatocytes under aerobic and anaerobicconditions. Am. J. Physiol. 257: C435-C441, 1989.Beal, M. F. Mechanisms of excitotoxicity in neurologic diseases. FASEB J. 6: 3338-3344,1992.Ben-An, Y. Galanine and glibenclamide modulate the anoxic release of glutamate in rat CA3hippocampal neurons. Eur. J. Neurosci 2: 62-68, 1990.Bennett, A. F., and W. R. Dawson. Metabolism In: Biology of the reptilia, edited by C. Gans.London: Academic Press, 1969, 127-223.Benveniste, H., J. Drejer, A. Schousboe, and N. H. Diemer. Elevation of the extracellularconcentrations of glutamate and aspartate in rat hippocampus during transient cerebralischemia monitored by intracerebral microdialysis. J. Neurochem. 43: 1369-1374, 1984.Berta, A., C. E. Ray, and A. R. Wyss. Skeleton of the oldest known pinniped Enaliactosmealsi. Science Wash. D. C. 244: 60-63, 1989.Beyersdorf, F., A. Unger, A. Wildhirt, U. Kretzer, N. Deutschlander, S. Kruger, G. Matheis,A. Hanselmann, G. Zimmer, and P. Satter. Studies of reperfusion injury in skeletal muscle:preserved cellular viability after extended periods of warm ischemia. .1. Cardiovasc. Surg.32: 664-676, 1991.Bickler, P. E. Cerebral anoxia tolerance in turtles: regulation of intracellular calcium. Am. J.Physiol. 263: R1298-R1302, 1992a.Bickler, P. E. Effects of temperature and anoxia on regional cerebral blood flow in turtles.Am. J. Physiol. 262: R538-R541, 1992b.Bickler, P. E., and S. M. Gallego. Inhibition of brain calcium channels by plasma proteinsfrom anoxic turtles. Am. J. Physiol. 265: R277-R281, 1993.Blanton, M. G., J. J. Lo, and A. R. Kreigstein. Whole cell recording from neurons in slices ofreptilian and mammalian cerebral cortex. J. Neurosci. Meth. 30: 203-210, 1989.Block, B. A. The billfish brain and eye heater a new look at non-shivering thermogenesis.News Physiol. Sci. 2: 208-213, 1987.Borgstrom, L., K. Norberg, and B. K. SiesjO. Glucose consumption in rat cerebral cortex innormoxia, hypoxia and hypercapnia. Acta Physiol. Scand. 96: 569-574, 1976.86Boyle, R. New pneumatical experiments about respiration. Trans. R. Soc. Lond. B Biol. 5:2011-2031, 1670.Brooks, S. P. J., and K. B. Storey. Regulation of glycolytic enzymes during anoxia in theturtle Pseudemys scripta. Am. J. Physiol. 257: R278-R283, 1989.Brown, T. H., D. H. Perkel, J. C. Norris, and J. H. Peacock. Electronic stnicture and specificmembrane properties of mouse dorsal root ganglion neurons. J. Neurophysiol. 45: 1-15,1981.Buck, L. T., and P. W. Hochachka. An assessment of the importance of succinate as ananaerobic endproduct in the diving turtle (Chrysemys picta bellii). .1. Exp. Biol. , In Press.Buck, L. T., P. W. Hochachka, and E. Gniager. Microcalorimetric measurement of reversiblemetabolic suppression induced by anoxia in isolated hepatocytes. J. Exp. Biol. , In Press.Caidwell, P. C., A. L. Hodgkin, R. D. Keynes, and T. I. Shaw. The effects of injectingenergy-rich phosphate compounds on the active transport of ions in the active transport ofions in the giant axons of Loligo. J. Physiol. Lond. 152: 561-590, 1960.Castellini, M. A., G. N. Somero, and G. L. Kooyman. Glycolytic enzyme activities in tissuesof marine and terrestrial mammals. Physiol. Zoo! 54: 242-252, 1980.Chih, C. P., M. Rosenthal, and T. J. Sick. Ion leakage is reduced during anoxia in turtle brain:a potential survival strategy. Am. .1. Physiol 257: R1562-R 1564, 1989.Choi, D. W. Calcium-mediated neurotoxicity: relationship to specific channel types and rolein ischemic damage. TINS 11: 465-469, 1988a.Choi, D. W. Glutamate neurotoxicity and diseases of the nervous system. Neuron 1: 623-634,1988b.Clark, G. D., and S. M. Rothman. Blockage of excitatoly amino acid receptors protectsanoxic hippocampal slices. Neuroscience 21: 665-?, 1987.Clark, V. M., and A. T. Miller Jr. Studies on anaerobic metabolism in the fresh-water turtle(Pseudemys Scripta Elegans). Comp. Biochem. Physiol. 44A: 55-62, 1973.Connors, B. W., and A. R. Kreigstein. Cellular physiology of the turtle visual cortex:distinctive properties of pyramidal and stellate neurons. J. Neurosci. 6: 164-177, 1986.Crane, P. D., L. D. Braun, E. M Cornford, J. E. Cremer, J. M. Glass, and W. H. Oldendorf.Dose dependent reduction of glucose utilization by pentobarbital in rat brain. Stroke 9: 12-18, 1978.Cserr, H. F., M. DePasquale, and D. C. Jackson. Brain and cerebrospinal fluid ioncomposition after long-term anoxia in diving turtles. Am. J. Physiol. 255: R338-R343,1988.Cummins, T. R., D. F. Donnelly, and G. G. Haddad. Effect of metabolic inhibition on theexcitability of isolated hippocampal CA 1 neurons: developmental aspects. .1. Neurophysiol.66: 1471-?, 1991.87Daves, D. G. Distribution of systemic blood flow during anoxia in the turtle, Chrysemysscripta. Resp. Physiol. 78: 383-390, 1989.Daw, J. C., D. P. Wenger, and R. M. Berne. Relationship between cardiac glycogen andtolerance to anoxia in the western painted turtle, Chrysemyspicta beliii. Comp. Biochem.Physiol. 22: 69-73, 1967.DeLong, R. L. Diving patterns of northern elephant seal bulls. Mar. Mamm. Sd 7: 369-384,1991.Dessauer, H. C. Blood chemistry of reptiles: physiological and evolutionary aspects In:Biology of the Reptilia, edited by C. Gans and T. S. Parsons. London: Acedemic Press,1970, 1-72.Dessauer, H. C., and W. Fox. Changes in ovarian follicle composition with plasma levels ofsnakes during estrus. Am. J. Physiol. 197: 360-366, 1959.Doudoroff, P., and D. L. Shumway. Dissolved oxygen requirements of freshwater fishes.F.A.O. Fisheries Technical Paper, No. 86: 1970.Drewes, L. R., and D. D. Gilboe. Glycolysis and the permeation of glucose and lactate in theisolated perfused dog brain during anoxia and postanoxic recovery. J. Biol. Chem. 218:2489-2496, 1973.Duffy, T. E., S. R. Nelson, and 0. H. Lowry. Cerebral carbohydrate metabolism during acutehypoxia recovery. J. Neurochem. 19: 959-977, 1972.Duncan, J. A., and K. B. Storey. Subcellular enzyme binding and the regulation of glycolysisin anoxic turtle brain. Am. J. Physiol. 262: R517-R523, 1992.Edelman, I. S. Transition from the poikilotherm to the homeotherm: possible role of sodiumtransport and thyroid hormone. Fed. Proc. 35: 2180-2184, 1976.Edwards, A. R., P. L. Lutz, and D. 0. Baden. Relationship between energy expenditure andion channel density in the turtle and rat brain. Am. J. Physiol. 257: R1354-R1358, 1989.Else, P. L., and A. J. Hulbert. Evolution of mammalian endothermic metabolism: \u00E2\u0080\u009Cleaky\u00E2\u0080\u009Dmembranes as a source of heat. Am. J. Physiol. 253: R1-R7, 1987.Elsner, R., and B. Gooden. Diving and Asphyxia. Cambridge: Cambridge University Press,1983,168.Feng, Z., M. Rosenthal, and T. J. Sick. Suppression of evoked potentials with continued iontransport during anoxia in turtle brain. Am. J. Physiol. 255: R478-R484, 1988.Feng, Z. -C., M. Rosenthal, and T. J. Sick. Extracellular pH and suppression of electricalactivity during anoxia in turtle and rat brain. Am. J. Physiol. 258: R205-R210, 1990.Feng, Z.-C., T. J. Sick, and M. Rosenthal. Orthodromic field potentials and recurrentinhibition dunn anoxia in turtle brain. Am. J. Physiol 255: R485-R491, 1988.Fry, F. E. J., and J. S. Hart. The relation of temperature to oxygen consumption in thegoldfish. Biol. Bull, 94: 66-77, 1948.88Fujiwara, N., H. Higashi, K. Shimoji, and M. Yoshimura. Effect of hypoxia on rathippocampal neurons in vitro. J. Physiol 384: 131-151, 1987.Funk, G. D., and W. K. Milsom. Changes in ventilation and breathing pattern produced bychanging body temperature and inspired CO2 concentration in turtles. Respir. Physiol 67:37-51, 1987.Furler, S. M., A. B. Jenkins, L. H. Storlien, and E. W. Kraegen. In vivo location of the rate-limiting step of hexose uptake in muscle and brain tissue of rats. Am. J. Physiol. 261: E337-E347, 1991.Gatten, R. E., Jr. Anaerobic Metabolism in freely diving painted turtles (Chrysemys picta). J.Exp. Zoo!. 216: 377-385, 1981.Gingerich, P. D., and D. E. Russell. Dentition of early eocene Pakicetus (Mammalia,Cetacea). Contrib. Mus. Paleontol. Univ. Mich. 28: 1-20, 1991.Glass, M. L., R. B. Boutilier, and N. Heisler. Ventilatory control of arterial Po2 in the turtleChrysemyspicta bellii: effects of temperature and hypoxia. J. Comp. Physiol. 151: 145-153, 1983.Globus, M. Y.-T., R. Busto, W. D. Dietrich, E. Martinez, I. Valdes, and M. D. Ginsberg.Effect of ischemia on the in vivo release of striatal dopamine, glutamate, and yaminobutyric acid studied by intracerebral microdialysis. J. Neurochem. 51: 1455-1456,1988.Gnaiger, E. Energetics of invertebrate anoxibiosis: direct calorimetry in aquatic oligochaetes.FEBSLett. 112: 239-242, 1980.Gnaiger, E., and R. B. Kemp. Anaerobic metabolism in aerobic mammalian cells:information from the ratio of calorimetric heat flux and respirometric oxygen flux. BiochimBiophys Acta 1016: 328-332, 1990.Haas, H. L., B. Schaerer, and M. Vosmansky. A simple perfusion chamber for the study ofnervous tissue slices in vitro. Nerurosci. Methods 1: 323-325, 1979.Haddad, G. G., and C. Jiang. Oxygen Deprivation in the CNS. Prog. Neurobiol. 40: 278-318,1993.Hammerstedt, R. H., and H. A. Lardy. The effect of substrate cycling ion the ATP yield ofsperm glycolysis. J. Biol. Chem. 258: 8759-8765, 1983.Hand, S. C., and E. Gniager. Anaerobic dormancy quantified in Artemia embryos: acalorimetric test of the control mechanism. Science Wash. D.C. 239: 1425-1427, 1988.Hansen, A. J. The extracellular potassium concentration in brain cortex following ischemia inhypo- and hyperglycemic rats. Acta Physiol. Scand. 402: 324-329, 1978.Hansen, A. J. Ion and membrane changes in brain anoxia In: Protection of tissues againsthypoxia, edited by A. Wauquier, M. Borgers and W. K. Amery. Amsterdam: ElsevierBiomedical Press, 1982, 199-209.89Hansen, A. J. Effect of anoxia on ion distribution in the brain. Physiol. Rev. 65: 101-148,1985.Hansen, A. J., J. Hounsgaard, and H. Jahnsen. Anoxia increases potassium conductance inhippocampal nerve cells. Acta Physiol. Scand. 115: 301-3 10, 1982.Hansen, A. J., and T. Zeuthen. Extracellular ion concentrations during spreading depressionand ischemia in the rat brain cortex. Acta Physiol. Scand. 113: 437-445, 1981.Hardewig, I., A. D. F. Addink, M. K. Grieshaber, H. 0. Portner, and G. van den Thillart.Metabolic rates at different oxygen levels determined by direct and indirect calorimetry inthe oxyconformer Sipunculus nudus. J. Exp. Biol. 157: 143-160, 1991.Hawkins, R. Cerebral Energy Metabolism In: Cerebral Energy Metabolism andEncephalopathy, edited by D. W. McCandless. New York: Plenum Press, 1985,3-23.Hawkins, R. A., A. L. Miller, J. E. Cremer, and R. L. Veech. Measurement of the rate ofglucose utilization by rat brain in vivo. J. Neurochem. 23: 917-923, 1974.Herbert, C. V., and D. C. Jackson. Temperature effects on the responses to prolongedsubmergence in the turtle Chiysemys picta bellii. I. Blood acid-acid base and ionic changesduring the following anoxic submergence. Physiol. Zoo!. 58: 655-669, 1985.Herreid, C. F., II. Hypoxia in invertebrates. Comp. Biochem. Physiol. 67A: 311-320, 1980.Higashi, H., S. Sugita, S. Nishi, and K. Shimoji. The effect of hypoxia on hippocampalneurons and its prevention by +2 antagonists In: Mechanisms of Cerebral Hypoxia andStroke, edited by G. Somjen. New York: Plenum Press, 1988,205-218.Hille, B. Ionic Channels ofExcitable Membranes. Sunderland: Sinauer Ass. Inc. 1992,607.Hindell, M. A., D. J. Slip, and H. R. Burton. The diving behavior of adult male and femalesouthern elephant seals Mitounga leonina (Pinnipedia: Phocidae). Aust. .1. Zool. 39: 595-619, 1991.Hitzig, B. M., M. P. Kneussl, V. Shih, R. D. Brandstgetter, and H. Kazemi. Brain amino acidconcentrations during diving and acid-base stress in turtles. J. App!. Physiol. 58: 1751-1754, 1985.Hochachka, P. W. Living Without Oxygen. Harvard University Press, 1980, 181.Hochachka, P. W. Defense strategies against hypoxia and hypothermia. Science Wash. D.C.231: 234-241, 1986.Hochachka, P. W. Metabolic Arrest. Intensive Care Med. 12: 127-133, 1987.Hochachka, P. W., and M. Guppy. Metabolic Arrest and the Control ofBiological Time.Cambridge: Harvard University Press, 1987,227.Hochachka, P. W., T. G. Owen, J. F. Allen, and G. C. Whittow. Multiple end products ofanaerobiosis in diving vertebrates. Comp. Biochem. Physiol. 50B: 17-22, 1975.90Hochachka, P. W., and G. N. Somero. Biochemical Adaptation. Princeton: PrincetonUniversity Press, 1984,538.Ismail-Beigi, F., and I. S. Edelman. Mechanism of thyroid calorigenesis: Role of activesodium transport. Proc. Nail. Acad. Sci. U.S.A. 67: 1071-1078, 1970.Ito, K., Y. Sawada, H. Ishizuka, Y. Sugiyama, H. Suzuki, T. Iga, and M. Hanano.Measurement of cerebral glucose utilization from brain uptake of[\u00E2\u0080\u00984C]-deoxyglucose and[3H]3-O-methylglucose in the mouse. Pharmacol. Methods 23: 129-140, 1990.Jackson, D. C. Metabolic depression and oxygen depletion in the diving turtle. J. App.Physiol. 24: 503-509, 1968.Jackson, D. C. Long-term submergence at 3 oC of the turtle, Chrysemys picta bellii, innormoxia and severely hypoxic water. II. Extracellular ionic responses to extreme lactateacidosis. J. Exp. Biol. 96: 29-43, 1982a.Jackson, D. C. Plasma ion balance of submerged anoxic turtles at 3 \u00C2\u00B0C: the role of calciumlactate formation. Respir. Physiol. 49: 159-174, 1982b.Jackson, D. C., and K. Schmidt-Neilsen. Heat production during diving in the freshwaterturtle, Pseudemys scnpta. J. Cell. Physiol. 67: 225-23 1, 1966.Jiang, C., Y. Xia, and G. G. Haddad. Role of ATP-sensitive K channels during anoxia:major differences between rat (newborn and adult) and turtle neurons. J. Physiol 448: 599-612, 1992.Jones, M. D., Jr., and R. J. Trystman. Cerebral oxygenation of the fetus, newborn, and adult.Seminars Perinatology 8: 205-2 16, 1984.Kass, I. S., and P. Lipton. Mechanisms involved in irreversible anoxic damage to the in vitrorat hippocampal slice. J. Physiol. 332: 459-472, 1982.Kauppinen, R. A., and D. G. Nicholls. Synaptosomal bioenergetics. The role of glycolysis,pyruvate oxidation and responses to hypoglycemia. Eur. J. Biochem. 158: 159-165, 1986.Kauppinen, R. A., and D. 0. Nichols. Failure to maintain glycolysis in anoxic nerveterminals. J. Neurochem. 47: 1864-1869, 1986.Keiver, K. M., and P. W. Hochachka. Catecholamine stimulation of hepatic glycogenolysisduring anoxia in the turtle Chrysemys picta. Am. J. Physiol. 261: R1341-R1345, 1991.Keiver, K. M., J. Weinberg, and P. W. Hochachka. The effect of anoxic submergence andrecovery on circulating levels of catecholamines and corticosterone in the turtle, Chrysemyspicta. Gen. Comp. Endocrinol. 85: 308-315, 1992.Kelly, D. A., and K. B. Storey. Organ-specific control of glycolysis in anoxic turtles. Am. J.Physiol. 255: R774-R779, 1988.Kerem, D., and R. EIsner. Cerebral tolerance to asphyxia! hypoxia in the dog. Am. J. Physiol225: 593-600, 1973a.91Kerem, D., and R. Eisner. Cerebral tolerance to asphyxial hypoxia in the harbor seal. Respir.Physiol. 19: , 1973b.Kirino, T., A. Tamura, and K. Sano. Selective vulnerability of the hippocampus to ischemiareversible and irreversible types of ischemic cell damage. Prog. Brain Res 63: 39-58, 1985.Kitner, D., J. H. Fitzpatrick Jr., J. A. Louie, and D. D. Gilboc. Cerebral oxygen and energymetabolism during and after 30 minutes of moderate hypoxia. Am. J. Physiol. 247: \u00C2\u00A3475-E482, 1984.Kooyman, G. L. Diverse Divers. Berlin: Springer-Verlag, 1989, 200.Kraig, R. P., and C. Nicholson. Extracellular ionic variations during spreading depression.Neurosci 3: 1045-1059, 1978.Krnjevic, K. Membrane current activation and inactivation during hypoxia in hippocampalneurons In: Cellular Defense Strategies to hypoxia, edited by P. W. Hochachka, P. Lutz,A. Wauquier, M. Rosenthal and G. van den Thillart. Boca Raton: CRC Press, 1993,Krnjevic, K., and J. Leblond. Changes in membrane currents of hippocampal neurons evokedby brief anoxia. J. Neurophysiol. 62: 15-30, 1989.Krnjevic, K., and Y. Xu. Mechanisms underlying anoxic hyperpolanzation of hippocampalneurons. Can. J. Physiol. Pharmacol. 68: 1609-?, 1990.Kmjevic, K., and Y. Z. Xu. Dantrolene suppresses the hyperpolarization or outward currentobserved during anoxia in hippocampal neurons. Can J. Physiol. Pharmacol. 67: 1602-1604, 1989.Ksiezak, H. J., and G. E. Gibson. Oxygen dependence of glucose and acetycholinemetabolism in slices and synaptosomes from rat brain. J. Neurochem. 37: 305-3 14, 1981.Leblond, J., and K. Krnjevic. Hypoxic changes in hippocampal neurons. .1. Neurophysiol. 62:1-15, 1989.Leivestad, H. The effect of prolonged submersion on the metabolism and the heart rate of thetoad (Bufo bufo). Univ. Bergen Arbok Naturvitenskap Rekke 5: 1-15, 1960.Lipton, P., and T.S. Whittingham. Reduced ATP concentration as a basis for synaptictransmission failure during hypoxia in the in vitro guinea -pig hippocampus. J. Physiol.325: 51-65, 1982.Lipton, P., and T. S. Whittingham. Energy metabolism and brain slice function In: BrainSlices, edited by R. Dingledine. New York: Plenum Press, 1984,Lowiy, 0. H., J. V. Passonneau, F. X. Hasselberg, and D. W. Schulz. Effect of ischemia onknown substrates and cofactors of the glycolytic pathway in brain. J. Biol. Chem. 239: 18-30, 1964.Lutz, P. L. Mechanisms for anoxic survival in the vertebrate brain. Annu. Rev. Physiol. 54:619-637, 1992.92Lutz, P. L., and P. M. McMahon. y-Aminobutyric acid concentrations are maintained inanoxic turtle brain. Am. J. Physiol. 249: R372-R374, 1985.Lutz, P. L., P. M. McMahon, M. Rosenthal, and T. J. Sick. Relationship between aerobic andanaerobic energy production in turtle brain in situ. Am. J. Physiol. 247: R740-R744, 1984.Lutz, P. L., M. Rosenthal, and T. J. Sick. Living without oxygen: turtle brain as a model ofanaerobic metabolism. Mo!. Physiol. 8: 411-425, 1985.Lynch, R. M., and R. J. Paul. Compartmentation of glycolytic and glycogenolyic metabolismin vascular smooth muscle. Science Wash. D.C. 222: 1344-1346, 1983.Macmillan, V., and B. K. SiesjO. Cerebral energy metabolism In: Scientific Foundations ofNeurology, edited by M. Critchley, J. O\u00E2\u0080\u0099leary and B. Jennett. London: William HeinemannMedical Books Ltd, 1972,21-32.Mangum, C., and W. Van Winkle. Response of aquatic invertebrates to declining oxygenconditions. Amer. Zoo!. 13: 529-541, 1973.McBride, B. W., and L. P. Milligan. Magnitude of ouabain-sensitive respiration of lambhepatocytes (Ovis aries). mt. J. Biochem. 17: 43-49, 1985.McCormick, D. A., B. W Connors, J. W. Lighthall, and D. A. Prince. Comparativeelectrophysiology of pyramidal and sparsely spiny stellate neurons of the neocortex. J.Neurophysiol. 54: 783-806, 1985.McCulloch, J., H. E. Savaki, J. Jehle, and L. Sokoloff. Local cerebral glucose utilization inhypothermic and hyperthermic rats. J. Neurochem. 39: 255-258, 1982.McDougal, D. B., Jr., J. Holowach, M. C. Howe, E. M. Jones, and C. A. Thomas. The effectsof anoxia upon energy sources and selected metabolic intermediates in the brains of fish,frog, and turtle. J. Neurochem. 15: 577-588, 1968.Mcllwain, H., and H. S. Bachelard. Biochemistry and the Central Nervous System.Edinburgh: Chruchill Livingston, 1971,616.Milton, R. L., and J. H. Calswell. How do patch clamp seals form. Pflugers Arch. 416: 758-765, 1990.Mink, J. W., R. J. Blumenschine, and D. B. Adams. Ratio of central nervous system to bodymetabolism in vertebrates: its constancy and functional basis. Am. J. Physiol 241: R203-R212, 1981.Morris, M. E., J. Leblond, N. Agopyan, and K. Krnjevic. Temperature dependence ofextracellular ionic changes evoked by anoxia in hippocampal slices. .1. Neurophysiol. 65:157-167, 1991.Mourre, C., Y. Ben An, H. Bernardi, M. Fosset, and M. Lazdunski. Antidiabeticsulfonylureas: localization of binding sites in the brain and effects on the hyperpolarizationinduced by anoxia in hippocampal slices. Brain Res. 486: 159-164, 1989.Musacchia, X. J. The viability of Chrysemys picta submerged at various temperatures.Physiol. Zoo!. 32: 47-50, 1959.93Nilsson, G., A. A. Alfaro, and P. L. Lutz. Changes in turtle brain neurotransmitters andrelated substances during anoxia. Am. J. Physiol. 259: R376-R384, 1990.Nilsson, G. E., and P. L. Lutz. Release of inhibitory neurotransmitters in response to anoxiain turtle brain. Am. J. Physiol. 261: R32-R37, 1991.Nilsson, G. E., P. L. Lutz, and T. L. Jackson. Neumtransmitters and anoxic survival of thebrain: a comparison of anoxia-tolerant and anoxia-intolerant vertebrates. Physiol. Zool. 64:638-652, 1991.Okada, Y., M. Tanimoto, and K. Yoneda. The protective effect of hypothermia onreversibility in the neuronal function of the hippocampal slice during long lasting anoxia.Neurosci. Lett. 84: 277-282, 1988a.Okada, Y., M. Tanimoto, and K. Yoneda. The protective effect of hypothermia onreversibility in the neuronal function of the hippocampal slice during long lasting anoxia.Neurosci. Lett. 84: 277-282, 1988b.Okamoto, K., and J. H. Quastel. Tetrodotoxin-sensitive uptake of ions and water by slices ofrat brain in vitro. Biochem. .1. 120: 37-47, 1972.Partridge, L. D., and D. Swandulla. Calcium-activated non-specific cation channels. TINS 11:69-72, 1988.Pasteur, L. Experiences et vues nouvelles sur la nature des fermentations. C.R. Acad. Sci 52:1260-1264, 1861.Pastuszko, A., D. F. Wilson, M. Erencinska, and I. A. Silver. Effects of in vitro hypoxia andlowered pH on potassium fluxes and energy metabolism in rat brain synaptosomes. J.Neurochem36: 116-123, 1981.Penney, D. G. Effects of prolonged diving anoxia on the turtle, Pseudemys scripta elegans.Comp. Biochem. Physiol. 47A: 933-941, 1974.Penney, D. G. Frogs and turtles: different ectotherm overwintering strategies. Comp.Biochem. Physiol. 86A: 609-615, 1987.P\u00C3\u00A9rez-PinzOn, M. A., C. Y. Chan, M. Rosenthal, and T. J. Sick. Membrane and synapticactivity during anoxia in the isolated turtle cerebellum. Am. J. Physiol. 263: R1057-R1063,1992a.P\u00C3\u00A9rez-Pinz\u00C3\u00B3n, M. A., M. Rosenthal, P. L. Lutz, and T. J. Sick. Anoxic survival of the isolatedcerebellum of the turtle Pseudemis scripta elegans. J. Comp. Physiol. B 162: 68-73, 1992b.P\u00C3\u00A9rez-PinzOn, M. A., M. Rosenthal, T. J. Sick, P. L. Lutz, J. Pablo, and D. Mash.Downregulation of sodium channels during anoxia: a putative survival strategy of turtlebrain. Am. J. Physiol. 262: R712-R715, 1992c.Processer, C. L., L. M. Barr, R. D. Pinc, and C. Y. Lauer. Acclimation of goldfish to lowconcentrations of oxygen. Physiol. Zoo!. 30: 137-141, 1957.Prosser, C. L. Adaptational Biology: Molecules to Organisms. New York: Wiley & Sons,1986, 1-784.94Randall, D. The control of respiration and circulation in fish during exercise and hypoxia. J.Exp. Biol. 100: 275-288, 1982.Reiner, P. B., A. G. Laycock, and C. J. Doll. A pharmacological model of ischemia inhippocampal slice. Neurosci. Lett. 119: 175-178, 1990.Ridge, J. W. Hypoxia and the energy charge of the cerebral adenylate pool. Biochem. J. 127:351-355, 1972.Robin, E. D., N. Lewiston, A. Newman, L. M. Simon, and J. Theodore. Bioenergetic patternof turtle brain and resistance to profound loss of mitochondnal ATP generation. Proc. Nail.Acad. Sci. USA 76:3922-3926, 1979.Robin, E. D., D. A. Robin, R. Ackerman, N. Lewiston, A. J. Hance, M. Caligiuri, and J.Theodore. Prolonged diving and recovery in the freshwater turtle, Pseudemys scripta-I.lunge and blood gasses, pH, lactate concentrations and \u00E2\u0080\u009Ccation gap\u00E2\u0080\u009D. Comp. Biochem.Physiol. 70A: 359-364, 1981.Rolleston, F. S., and E. A. Newsholme. Control of glycolysis in cerebral cortex slices.Biochem. J. 104: 524-533, 1967a.Rolleston, R. S., and E. A. Newsholme. Effects of fatty acids, ketone bodies, lactate andpyruvate on glucose utilization by guinea - pig cerebral cortex slices. Biochem. J. 104: 519-523, 1967b.Rossier, B. C., K. Geering, and J. P. Kraehenbuhl. Regulation of the sodium pump: how andwhy? TIBS 12: 483-487, 1987.Rothman, S. The neurotoxicity of excitatory amino acids is produced by passive chlorideinflux. J. Neurosci. 5: 1483-1489, 1985.Sinchez-Prieto, J., and P. GonzAlez. Occurrence of largeCa2+independent Release ofglutameate during anoxia in isolated nerve terminals (synaptosomes). J. Neurochem. 50:1322- 1324, 1988.Schmahl, F. W., E. Betz, E. Dettinger, and H. Hohorst. Energiestoffwechs der grosshirnnndeund elektroencephalogram bie sauerstoffmangel. Plug. Arch. Gesamte Physiol. 292: 46-59,1966.Schmidt-Nielsen, K. Animal physiology.- Adaptation and environment. Cambridge:Cambridge University Press, 1979, 1-560.Schmidt-Nielsen, K. Scaling, why is animal size so important. Cambridge: Cambridge Univ.Press, 1984,Schulte, P. M., C. D. Moyes, and P. W. Hochachka. Integrating metabolic pathways in post-exercise recovery of white muscle. J. Exp. Biol. 166: 181-195, 1992.Schurr, A.., C. A. West, and B. M. Rigor. Lactate-supported synaptic function in the rathippocampus slice preparation. Science Wash. D. C. 240: 1326-1328, 1988.95Scott, I. D., and D. G. Nicholls. Energy transduction in intact synaptosomes. Biochem. J.186: 21-33, 1980.Shick, J. M., A. De Zwaan, and A. M. T. Dc Bont. Anoxic metabolic rate in the musselMytilus edulis 1. estimated by simultaneous direct calorimetry and biochemical analysis.Physiol. Zool. 56: 56-63, 1983.Shoubndge, E. A., R. W. Bnggs, and G. K. Radda. 31P NMR saturation transfermeasurements of the steady state rates of creatine kinase and ATP synthetase in the ratbrain. FEBS Lett. 140: 288-292, 1982.Shoubridge, E. A., and P. W. Hochachka. The origin and significance of metabolic carbondioxide production in the anoxic goldfish. Mol. Physiol. 1: 315-338, 1981.Sick, T. J., E. P. Chasnoff, and M. Rosenthal. Potassium ion homeostasis and mitochondrialredox status of turtle brain during and after ischemia. Am. J. Physiol 248: R53 1-R540,1985.Sick, T. J., and N. R. Kreisman. Potassium ion homeostasis in amphibian brain: contributionof active transport and oxidative metabolism. .1. Neurophysiol. 45: 998-1012, 1981.Sick, T. J., M. Perez-Pinzon, P. L. Lutz, and M. Rosenthal. Maintaining coupled metabolismand membrane function in anoxic brain: a comparison between the turtle and rat In:Surviving Hypoxia, edited by P. W. Hochachka, P. L. Lutz, T. J. Sick and G. van denThillart. Boca Raton: CRC Press, 1993,351-364.Sick, T. J., M. Rosenthal, J. C. LaManna, and P. L. Lutz. Brain potassium ion homeostasisduring anoxia and metabolic inhibition in the turtles and rats. Am. J. Physiol. 243: R281-R288, 1982.Sick, T. J., E. L. Solow, and E. L. Roberts Jr. Extracellular potassium ion activity andelectrophysiology in the hippocampal slice: paradoxical recovery of synaptic transmissionduring anoxia. Brain Res. 418: 227-234, 1987.Siemkowicz, E., and A. J. Hansen. brain extracellular ion composition and EEG activityfollowing 10 minutes ischemia in normo- and hyperglycemic rats. Stroke 12: 236-240,1981.Siesjo, B. K. Brain Energy Metabolism. New York: Wiley, 1978,607.Siesjo, B. K., and L. Nilsson. The influence of arterial hypoxemia upon labile phosphates andupon extracellular and intracellular lactate and pyruvate concentrations in the rat brain.Scand. .1. Clin. invest. 27: 83-96, 1971.Suarez, R. K. Thinking with and without oxygen: energy metabolism in vertebrate brains.Can. .1. Zool. 66: 1041-1045, 1987.Suarez, R. K., C. J. Doll, A. E. Buie, T. G. West, G. D. Funk, and P. W. Hochachka. Turtlesand rats: a biochemical comparison of anoxia-tolerant and anoxia -sensitive brains. Am. J.Physiol. 257: R1083-R1088, 1989.Suzuki, R., T. Yamguchi, Y. Inaba, and H. G. Wagner. Microphysiology of selectivelyvulnerable neurons. Prog. Brain Res. 63: 59-68, 1985.96Swann, A. C., and R. W. Albers. Sodium+photassium-activated ATPase of mammalian brainregulation of phosphatase activity. Biochim et Biophys Acta 382: 437-456, 1975.Ultsch, G. The Viability of nearctic freshwater turtles submerged in anoxia and normoxia at 3and 10\u00C2\u00B0C. Comp. Biochem. Physiol. 81A: 607-611, 1985.Ultsch, G. R. Ecology and physiology of hibernation and overwintering among freshwaterfishes, turtles and snakes. Biol. Rev. 64: 435-5 16, 1989.Ultsch, G. R., and D. C. Jackson. Long-term submergence at 3 \u00C2\u00B0C of the turtle, Chrysemyspicta bellii, in normoxic and severely hypoxic water. I. Survival, gas exchange, and acid-base balance. J. Exp. Biol. 96: 11-28, 1982.van Waversveld, J., A. D. F. Addink, G. van den Thillart, and H. Smit. Direct calorimetry onfree swimming goldfish at different oxygen levels. .1. Therm. Anal. 33: 1019-1026, 1988.Wegener, G. Insect brain metabolism under normoxic and hypoxic conditions In: ArthropodBrain: Its Evolution, Development, Structrue and Functions, edited by A. P. Gupta. NewYork: John Wiley & Sons, Inc., 1987,369-397.Wegener, G., R. Michel, S. Kieffer, and M. Thuy. Different metabolic reactions to anoxia inthe central nervous system of lower vertebrates and adult insects and their possiblesignificance for postanoxic revival. Mol. Physiol 8: 653-655, 1987.Whalen, W. J., D. Buerk, and C. A. Thuning. Blood flow-limited oxygen consumption inresting cat skeletal muscle. Am. J. Physiol. 224: 763-768, 1973.Whittam, R. The dependence of the respiration of brain cortex on active cation transport.Biochem. J. 82: 205-212, 1962.Whittam, R., and D. M. Blond. Respriatory control by an adenosine tnphosphatase involvedin active transport in brain cortex. Biochem. J. 92: 147-157, 1964.Wood, E. H., and P. B. Reiner. Glutamate plays a minor role in the membrane events leadingto neumnal death in a pharmacological model of ischemia in vitro. Soc. Neurosci. Abstr.16: 274, 1990.Xia, Y., and G. G. Haddad. Neuroanatomical distribution and binding properties of saxitoxinsites in the rat and turtle CNS. .1. Compar. Neurol. 330: 381-404, 1993.Xia, Y., C. Jiang, and G. G. Haddad. Oxidative and glycolytic pathways in rat (newborn andadult) and turtle brain: role during anoxia. Am. J. Physiol. 262: R595-R603, 1992.Yamamoto, C., and M. Kurokawa. Synaptic potentials recorded in brain slices and theirmodification by changes in the level of tissue ATP. Exp. Brain Res. 10: 159-170, 1970.Yellen, G. Single Ca2+ -activated nonselective cation channels in neuroblastoma. Nature296: 357-359, 1982.97APPENDIX A: TECHNIQUESBecause a large part of this thesis focuses on electrophysiological techniques which arenot familiar to a large portion of the scientific community, this section is devoted todescribing these methods in a more general detail. However, I refer the reader to themethods of individual chapters for technique specifics.The two most common methods of electrophysiological recording are the sharp electrode(intracellular) recording technique and the patch clamp technique. The intracellularrecording technique uses a very fine piece of glass tubing which is pulled (simultaneouslyheating) to a sharp point .1 pM in diameter. The electrode is then filled with a currentcarrying solution usually KCI 2M and connected to a headstage giving it direct hookup toan amplifier. The electrode is then lowered on to the surface of the cell membrane.Penetration of the electrode through the membrane is accomplished via a current pulse. Oncethe electrode penetrates the cell membrane, a seal forms due to the attraction of the chargedlipids of the membrane and the glass (Fig. 21).Two subcategories of intracellular recording technique are current clamping and voltageclamping. Current clamping is a method in which the experimenter controls the currentwhich is injected into the cell and recording the resulting voltage deflection for that givenamount of current (I). Voltage clamping is a method in which the experimenter controls thevoltage potential (Vwj of a cell and records the resulting I. Both techniques are suited toparticular applications. For this thesis, current clamping methods are exclusively used.The second major technique is patch clamping (Fig. 21). The setup for patch clamping issimilar to intracellular recording techniques. Instead of breaking through the cell membrane,the electrode is lowered to the surface of the cell, where, through the use of negativepressure, a high resistance seal ( 1GQ) is created. This technique involves the use of amuch larger electrode tip (tip diameter 1 - 5 pM). Once the seal has formed, severalconfigurations can be attained (Fig. 22). Note that with intracellular recording, the electrodesolution does not immediately equilibrate with the intracellular contents, but with the whole98AmplifierBFigure 21. Electrophysiological recording techniques. (A) represents intracellular recordingtechniques using sharp electrodes. A method commonly used with both voltage and currentclamp techniques. (B) represents patch clamping. Note that the electrode does not penetratethe cell membrane, rather it is bonded to the outside membrane (see text for further details)KC1Voltage meterVIcell IGround GroundReferenceElectrodeAmplifierPatchSolutionVoltage meterGround GroundReferenceElectrodeICeill99Patch ElectrodePullSuctionPullPatch AttachedRight Side OutPatched AttachedInside OutFigure 22. The four configurations of patch clamping. All methods start with the on-cellconfiguration and are subsequently modified by pulling or sucking once attached to the cell.All configurations result in the formation of a GQ seal being formed between the electrodeand cell membrane.Initial Seal Whole Cell100ToGasBottlePlexiglassbox= Ringers Solution= 95% 02/5% CO2= 95% N/5% CO2Figure 23. The slice chamber recording set-up and perfusion system for both patch clampingand intracellular recording techniques.- To GasBottleRubberStopperHPLCSolutionBottlesGas ImpermeableHPLC TubingTo suction bottleSlice Chamber101cell patch clamp configuration, the electrode content equilibrates rapidly with the cellcytoplasm.Both electrophysiological recording techniques used in this study involved a flow throughslice chamber (Fig. 23). This chamber was simple a Plexiglas box in which the top had achannel cut. Fluid entered the chamber in the front via gravity. Excess fluid was suctionedoff at the opposite end drawing fluid across the slice. Slices were encapsulated in a mesh onboth the top and bottom which was weighted down with platinum wire. This configurationallowed aCSF to flow both above and below the slice ensuring viability of the tissue.Mixture bottles were switched via a low volume HPLC valve. Fluid flow across the slicewas approximately 1.5 mI/mm.102APPENDIX B: TERMONOI1OGYWith regards to the techniques used, there are several common terms which are used inthe electrophysiology field, but are not well understood by the non-electrophysiologists. Thisthesis focuses considerably on resistance and conductance of cells. Because electricalcurrents in solution are carried by ions in that solution, solution electricity is really the studyof how ions move in solution. Thus, when the term resistance or conductance is used inreference to a cell parameter, this is not in reference to how an electric current is moving perSe, but how ions are moving across the cell membrane. Perhaps the simplest term is wholecell resistance (Rw). This term refers to the resistance of the whole cell (impedance to theflow of ions across the cell membrane). It is generally determined by injecting a currentpulse across the cell membrane using either whole cell patch clamping or intracellularrecording techniques in the current clamp mode and measuring the resulting changemembrane potential (VJ. Since the current was preselected and the resulting Vm ismeasured, the resistance (Rw ) can be calculated by Ohms law:VmIXRwThis simple equation will indicate the resistance that the ions in a solution encounter whengoing from the current generating electrode to the bath ground. Since the electrode is in thecell (or on the cell membrane in the case of patch clamping), and the resistance through thesolution is negligible, the resulting resistance measure is due primarily to the cell membrane.Since the current can potentially leave the cell anywhere on the membrane, this term is anaverage over the whole cell. The reverse is true when considering whole cell conductance(Gw). That is:Gw1/RwConductance is a measure of how easily ions flow across the membrane. A cell with a highresistance will have a low conductance and vise versa. Once again, this is in reference to thewhole cell which is attached to the electrode.103This terminology is not to be confused with specific membrane resistance (Rij orspecific membrane conductance (Gm). This terminology refers to the resistance orconductance for a given area of membrane and is calculated totally independent of the wholecell measures discussed above. 0m is generally calculated from the membrane chargingcurve of the cell. As a current is injected into a cell, part of the current goes to charge thecapacitance of the cell membrane, and part of the current escapes through the ion channels.As the membrane reaches its maximum charging potential, more and more current escapesthrough ion channels. The resulting curve is termed a membrane charging curve (for anexample see Fig. 15, Chapter 4). The time at which it takes a membrane to become 63%charged of its maximum charging potential is termed the time constant (Te). This term byitself does not yield a considerable amount of information, but in conjunction with thefollowing equation, it yields the cell Gm:TcCm/Gmwhere Cm is equal to the specific capacitance of the membrane. Given that Cm is known,this equation then allows the calculation of Gm or Rm. Specific limitations to this equationwill be discussed in Chapter 3. Also, one could calculate the Gm and Rm by dividing Gw orRw by the total cell membrane surface area respectively. However, this is a considerablemore difficult task since each cell area would have to be measured.Specific membrane resistance/conductance and whole cell membrane resistance Iconductance may look similar, but the information which is provided is not interchangeable.Because Rw and G are averaged over the whole cell, they are dependent on cell size aswell as G and Rm. Thus, it is appropriate to compare the Rw and G from a single cellacross an insult, it is not correct to compare Rw and Gw to different cells or species. Incontrast, Gm and Rm can be used for both comparison from a single cell across an insult, orbetween different cells in a group, or across species since it does not have the limitations ofbeing dependent on cell size. However, both of these parameters are easily measured, and in104some cases the results appear redundant; however, because these measures are calculatedindependently of each other, the provide a very useful validation of results.Throughout this thesis, several terms are used in regards to various tissue insults. A basicunderstanding of this terminology will aid the reader in better understanding these insults.The terms anoxia/hypoxia can refer to several different states. Anoxia indicates a situation ofno 02, and hypoxia indicates low oxygen tension. However these terms can be complicatedby such additives as pharmacological anoxia, environmental anoxia, and physiologicalanoxia. Physiological anoxia refers to the blood P02 of zero. For example, since a seal cannot extract oxygen from the water it encounters environmental anoxia, yet it does notencounter physiological anoxia as discussed above. For the experiments in this thesis whichdeal with anoxia, they will be considered physiological anoxia since the tissue is bathed inaCSF equilibrated with 95% N2 15% C02. Anoxia is similar to pharmacological anoxiawhich denotes a situation in which anoxia is present, but also NaCN (cytochrome P-450inhibitor) to eliminate the possibility of any residual 02 metabolism.The third term ischemia denotes a situation in which the tissue receives no blood flow.In this case, the tissue is rapidly confronted with anoxia and once endogenous stores ofglucose and glycogen are used, a situation of no energy production occurs becauseexogenous substrate (glucose) is not delivered. The experiments dealing withelectrophysiology utilize a flow through slice chamber (as discussed previously). The flowthrough design of this chamber makes true ischemia impossible to mimic. However, thedevelopment of a pharmacological mimic of this insult (pharmacological ischemia) allows areasonable facsimile of this insult (Reiner, et al., 1990). Pharmacological ischemia subjectsthe cortical slice to a perfusing solution of pharmacological anoxia and iodoacetic acid(IAA). lodoacetic acid blocks glycolysis by inhibiting glyceraldehyde-3-phosphatedehydrogenase (glycolytic enzyme). Pharmacological ischemia is considered more severecompared to ischemia because of its rapid block of the glycolytic pathway.105"@en . "Thesis/Dissertation"@en . "1994-05"@en . "10.14288/1.0088158"@en . "eng"@en . "Zoology"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Mechanisms of anoxia tolerance in the turtle cortex"@en . "Text"@en . "http://hdl.handle.net/2429/6828"@en .