Biochemical Responses to Hypoxia in the Long-Jawed Mudsucker I: Metabolites Scott Stonington Hopkins Marine Station Issue in Marine Biology: 175H June 1999 Advisor: George Somero Permission is granted to Stanford University to use the abstract and citation of this paper Abstract Gillichthys mirabilis (the long-jawed mudsucker, Family Gobiidae), acclimated to 14°C, were exposed to different periods of hypoxia. I selected the hypoxia intensity based on critical oxygen concentration (Pai) data obtained from three fish, all of whom displayed Pait of 1.0 mg/L. I placed experimental fish in a consistent oxygen concentration of 0.8 mg/L, and sacrificed them after 0, 8, 24 and 72 hours of exposure by freeze-clamping in liquid nitrogen. I examined glycogen, lactate and ethanol concentrations in white muscle, liver, heart and brain. I found no ethanol in any tissue. Lactate accumulated significantly and similarly in all tissues throughout the time course. Glycogen changed significantly only in liver, which had very large initial stores (6.87 + 1.46 mg/g), and drastically depleted stores by 8 hours (0.59 + 0.22 mg/g) White muscle had consistently low glycogen (mean 0.27 + 0.01 mg/g). Brain showed some hints of glycogen depletion (p = 0.053) by 72 hours. I conclude that during hypoxia in Gillichthys mirabilis, glycogen is mobilized from stores in the liver and transported to meet the energy demands of brain, heart and white muscle, which conserve their own glycogen. Introduction Gillichthys mirabilis is a teleost fish of the family Gobiidae that has successfully adapted to a wide range of extreme environments. Possibly the most prominent of these extremes are the hypoxic and sulphurous sloughs of Western California - plains of partially toxic mud with tidal drainages that leave the substrate exposed to air and sun. To escape this exposure, Gillichthys mirabilis have been shown to hide in small burrows created by invertebrates in the slough's muddy substrate (Congleton, 1974). They thus spend significant time in small volumes of water that may experience extreme hypoxia and fluctuating temperatures. Indeed, as long ago as 1930, researchers became interested in Gillichthys mirabilis's ability to acclimate to temperature fluctuations (Sumner and Doudoroff, 1938), and more recently the gobiid has been used as a model for both heat and osmotic shock tolerance (Lin and Somero, 1995; Kultz and Somero, 1995; Kultz 1995; Kultz and Somero, 1994). The Gillichthys mirabilis habitat, however, is also a site of periodic extreme hypoxia, and they may thus prove valuable as a model for studying hypoxia tolerance. Hypoxia is an environmental stress pertinent to almost all aerobic organisms. It generally falls under two categories: environmental and physiological hypoxia. Environmental hypoxia occurs when an organism cannot obtain enough oxygen from its environment to maintain normal systemic energy consumption. It thus strikes at a systemic level, and the organism must address the problem in an integrated fashion among all tissues. In physiological hypoxia, despite abundant environmental oxygen, a tissue's energy demand is greater than the rate of oxygen supply to that tissue. This is usually a result of high activity, where one tissue (usually muscle) must work at such a rate that it cannot produce ATP quickly enough through aerobic pathways These two forms of hypoxia present very different problems to an organism. Moreover, hypoxia can span a large range of severity, from anoxia to common fluctuations in oxygen transport throughout the body. It is thus understandable that organisms might develop different strategies to different severities and forms of hypoxia. These strategies generally fall under three categories. First, organisms may reduce their physiological and behavioral activity to decrease the demand for ATP, and thus for oxygen (Hochachka et al., 1996). This usually involves a suppression of non-essential functions. Second, organisms may attempt to increase extraction of oxygen from the environment and delivery of oxygen to hypoxic tissues. This can include hyperventilation, vascularization of hypoxic tissues or alteration of blood properties such as hematocrit or hemoglobin concentrations. Third, and possibly most prominently, organisms may revert to anaerobic metabolic pathways to supply their remaining ATP needs. Consequently since anaerobic metabolism is far less efficient than aerobic metabolism, organisms require a greater quantity of metabolic substrate per ATP output, and thus may need to mobilize energy stores such as glycogen. Moreover, anaerobic metabolism produces undesirable end-products, such as lactic acid, which can cause acidotic damage to tissues. Thus tissues with high metabolic costs may be in greater danger from hypoxia than tissues that operate at lower metabolic output. To protect against acidotic buildup and to mobilize energy stores, organisms may employ tissue¬ specific metabolic strategies, shuttling energy-substrates throughout the body to provide for various tissues' energy needs. The purpose of this investigation is to discern which of these strategies Gillichthys mirabilis employs to resist the periods of environmental hypoxia in its environment. A logical first place to turn, of course, is to the regulation of metabolites, since any change in metabolic strategy should alter both the use of energy stores and the quantity and form of waste products. Metabolic substrates and end-products can be used as markers for the "black box" of overall metabolic strategy. The most obvious of these markers are glycogen and lactate, because glycogen is the primary energy-storage molecule in vertebrates and the lactic-acid-producing metabolic pathway, glycolysis, is the most common of anaerobic strategies. There are of course, numerous exceptions to this. Goldfish, for instance, which experience very long periods of hypoxia, employ a metabolic pathway that produces ethanol, which is then excreted as waste (Johnston and Bernard, 1982). Organisms may also deplete energy stores other than glycogen, such as fatty acids or proteins. Alternatively, they may maintain a high constitutive anaerobic throughput, and thus display little change in anaerobic metabolism during hypoxia. Nonetheless glycogen and lactate are invaluable markers that can help elucidate the overall metabolic strategy that Gillichthys mirabilis use to survive their hypoxic slough. Materials and Methods I conducted all experiments in collaboration with a colleague, Joshua Troll, who used the same test population and tissues for assays on glycolytic enzyme activity. Gillichthys mirabilis were collected from a lagoon at the University of California Santa Barbara and acclimated to 14°C for several months. On three separate occasions after their acclimation, we placed individual fish in a sealed tank coupled to an oxygen electrode to measure the fish's rate of oxygen uptake versus available oxygen concentration. These data are shown in figure 1 and were used to determine the Pert of our stock of fish (the oxygen concentration below which an organism's oxygen consumption rate begins to decline). We found the Pent to be approximately 1.0 mg/L, and designed the remainder of our experiments appropriately. We placed four sets of seven fish (3.5g-19.25g) in aerated Tupper-Ware"" containers in a tank liked to a nitrogen-gas hypoxia setup. Containers were used to facilitate sacrificing the fish without causing any substantial handling stress which might alter the fish's metabolic state. Äfter one day in this new setting, we sacrificed one of the four sets of fish and reduced the oxygen concentration in the tank to 0.8 mg/L, 80% of their Perit. The remaining three sets of fish were sacrificed after 8, 24 and 72 hours of hypoxia. Data for this exposure are shown in figure 1. We sacraficed the fish by freeze-clamping them in liquid nitrogen to immediately preserve their tissues and prevent both post-mortem enzyme activity and metabolite breakdown. We transferred the frozen fish to storage at -80°C until removing them for dissection. During the dissections, we kept the fish on dry ice, allowing us to perform separate dissections and return the fish to storage without disturbing their endogenous enzyme integrity by freezing and thawing their tissue. This technique allowed us to remove white muscle at one sitting (-400mg), and liver (100mg), heart (-40mg) and brain (-20mg) at the next. We parceled these tissues into two sets, one for endpoint-assays (this paper), another for enzyme activity assays (see data from Joshua Troll, this volume). I placed tissues for endpoint assays immediately in 0.6 N perchloric acid (5 ul/mg tissue) to arrest endogenous enzyme activity. I then homogenized these mixtures with an Ultraturax?. For the four tissues mentioned above, I assayed for glycogen according to the method of Keppler and Decker (1969). Two aliquots of homogenate from each tissue of each fish were neutralized to protect assay enzymes from deproteinization. To one of these, I added 250 ul of 11.0 units/ml amyloglucosidase solution in 0.2 M acetate buffer (pH 4.8) to hydrolyze the tissue's glycogen into glucose. The other aliquot (with 250 ul acetate) served as a blank of free glucose against which glycogen data could be compared. After digestion, I centrifuged the samples and used three 20 ul aliquots of supernatant for triplicate assays. I added these samples to 100 ul of reaction mixture (1 mM ATP; 0.9 mM B-NADP ; 5 ug glucose-6-phosphate dehydrogenase/ml.), and measured their change in absorbance before and after reaction with hexokinase. Absorbances were measured at 340 nm on a SpectraMax 340c 96-well spectrophotometer using SOFTMAX¬ PRO software, all manufactured by Molecular Devices. The absorbances from triplicate wells were averaged to obtain a value for a single tissue, and actual glucose concentration values were derived using the numerical analysis detailed in Keppler and Decker (1975) I assayed for l-lactic acid and ethanol using the methods and industrial kits provided by Boehringer Mannheim (Cat. Nos. 139084 and 176290). For each assay, I added three 20 ul aliquots of neutralized homogenate solution to wells containing 100 ul reaction solution (provided by the Boehringer Mannheim kits). To develop a standard curve for absorption change versus lactate or ethanol concentration, I also added aliquots of lactate and ethanol standard solution (of 9.7 mg/L, 29.1 mg/L, and 48.5 mg/L for lactate; 4.5 mg/L, 13.5 mg/L, 22.5 mg/L for ethanol). I then measured the absorbance of these solutions before and after reaction with lactate dehydrogenase and alcohol dehydrogenase, respectively. The method for these measurements is described above in the glycogen assay. -Results Oxygen consumption rate data as a function of oxygen concentration for three fish are shown in figure 1. This graph provides information both on the amount of oxygen that Gillichthys mirabilis consume as well as the critical oxygen concentration below which they begin to lose their ability to draw oxygen from their surroundings (Perit). Pent is thus measured as the oxygen concentration below which oxygen consumption rate consistently declines. The Perit for all three experimental fish was near 1.0 mg/L, and was independent of body mass. No ethanol was discernable in any of the four tissues. Ethanol is a volatile compound, and if present in small quantities in the tissues, may not survive in substantial quantities throughout the homogenizations and dilutions in the assay. The data, however, are clearly negative and rule out the existence of large ethanol concentrations in any of the tissues. Figures 3-6 show the effects of hypoxia on both glycogen and lactate concentration in each tissue. In all four tissues, lactate rose significantly after 8 hours of hypoxia (figure 7) (pS.001), and in liver and heart it rose again by 72 hours (pS.001). Each tissue’s increase in lactate is generally correlated with initial lactate levels, which were consistently highest in the brain and lowest in the liver (pS.005). Significant changes in glycogen concentration appeared only in liver (pS.001), which had a comparatively large initial store of glycogen that was almost completely depleted by 8 hours. The liver’s mean initial glycogen store (6.87 mg/g) was four times the store in the next highest tissue, heart (1.68 mg/g) and twenty-four times that in white muscle (0.29 mg/g). This low white muscle glycogen was consistent throughout the time-course. The observed decrease in brain glycogen was nearly significant (pS.053) at seventy-two hours, hinting at a pattern, but not significant enough to draw any conclusions. Discussion The 1.0 mg O/L Pait that I observed in G. mirabilis is not surprising given gobiid fishes capacity for hypoxia tolerance. Critical oxygen concentrations of 0.6 mg/L, for example, have been observed in gobies of the genus Typhlogobius. Interestingly, however, the Pent values observed in my study are on the low end of previously measured Pert in G. mirabilis, which range from 1.0-1.5 mg/L (Congleton, 1974). This may indicate that the G. mirabilis from my experimental population are more tolerant of hypoxia than most populations of the species. In a broader sense, the Pent values are informative in comparison to values from other organisms and to the oxygen concentrations found in most marine habitats. The oxygen concentration in surface seawater ranges from 8.1 mg/L to 9.5 mg/L (Yang et al., 1992). The Pet of yellowfin tuna, a highly aerobic pelagic fish, is 4-5 mg/L (Barbara Block, personal correspondence). Even Scorpaena guttata, a benthic fish that lives at 180m where oxygen concentration is one third of surface sea water, has a Pent of 3.0 mg/L (Yang et al., 1992). One of S. guttata's confamilials (Sebastolobus alascanus), however, lives in the oxygen minimum zone and has a Pert equivalent to that of G. mirabilis (Yang et al., 1992). Thus G. mirabilis has developed a physiology which facilitates life under hypoxic stress of similar intensity to that of the oxygen minimum zone. The glycogen levels that I found in G. mirabilis, especially those in the liver, shed important light on the G. mirabilis strategy for hypoxia tolerance. Glycogen levels changed significantly only in the liver, where large initial stores underwent a twelve-fold depletion by eight hours of hypoxia. Lactate, however, accumulated in all four organs, indicating that glycolysis proceeded according to each tissue's energy demand. Moreover, lactate levels were the lowest in liver, indicating that the glucose formed by breakdown of glycogen was not being consumed there, but was transported from the liver to other tissues. Lactate levels were highest in the brain and heart, neither of which depleted their own glycogen, indicating that they were receiving energy input from another source. These data argue that, in response to hypoxia, G. mirabilis may mobilize energy stores in the liver and shuttle glucose to support the energy demands of other tissues. These fish may thus be integrating each tissue's energy resources and demands in response to their systemic hypoxic stress. This strategy contrasts with anaerobic systems in other organisms. Although goldfish, for example, deplete their liver glycogen and maintain white muscle stores in response to anoxia (Thillart et al., 1980), they also deplete between 75 and 95% of their brain glycogen within two hours of exposure to anoxia (Schmidt and Wegner, 1988). This strategy is primarily in response to anoxia, however, and it is possible that G. mirabilis conserve their brain glycogen until the last moment or mobilize it under more severe hypoxia. Moreover, G. mirabilis does not tolerate anoxia to the same degree as goldfish, and anoxia-tolerance may entail a mechanism distinct from hypoxia-tolerance and thus be irrelevant in comparison to G. mirabilis. Another interesting contrast to other organisms' physiology is the size of G. mirabilis's glycogen stores. The initial stores of glycogen in the liver (6.87 + 1.46 mg/g) are larger than in most vertebrates. Human livers typically have 0.1-1.0 mg/g glycogen (Johnson and Fusaro, 1966; Lee and Whelan, 1966). Kepler and Decker (1969), who developed the glycogen assay employed in this study, found rat liver glycogen at 0.55 +0.1 mg/g. Glycogen levels in some other hypoxia-tolerant vertebrates, however, are much larger than those observed in this study. Glycogen makes up 30% of the crucian carp liver (300 mg/g), and the freshwater turtle has liver glycogen 25 times the quantities observed here (Lutz and Nilsson, 1997). Moreover, although the glycogen levels found here in the G. mirabilis brain (1.15 +0.11 mg/g) are 3-fold larger than in rat, mouse and rainbow trout, they are nonetheless 2-3-fold smaller than in the crucian carp, goldfish, and freshwater turtle (Lutz and Nilsson, 1994). This difference between G. mirabilis and the carp and turtle may be due to the types of hypoxia that each organism encounters in its natural environment. Carp and freshwater turtles are accustomed to longer durations of hypoxia (weeks to months), and may thus need to mobilize greater glycogen stores between opportunities for gluconeogenesis. Gillichthys mirabilis, on the other hand, have the tides to usher in fresh volumes of oxygenated water every twelve hours or less, so smaller glycogen stores may be sufficient to sustain them during hypoxic period at low tide. Different time-courses of hypoxia may also explain the failure to detect ethanol in G. mirabilis tissues. An ethanol-producing metabolic pathway appears in carp and goldfish to prevent acidotic buildup from lactic acied during very long periods of hypoxia (Johnston and Bernard, 1983). The absence of ethanol-producing pathways in G. mirabilis may indicate, in part, that acid accumulation is less of a problem over twelve-hour periods than over longer exposures. Gillichthys mirabilis, both in metabolic pathway strategies and in glycogen concentrations, appear to lie on a gradient dictated by the severity of hypoxia found in the natural environment. Certain features of the G. mirabilis biochemistry, however, are in contrast to this apparent trend. Glycogen in white muscle, for instance, was remarkably lower than in most vertebrates, including both anoxia-tolerant and anoxia-sensitive species. It remained at a constant mean of O.3 mg/g, an order of magnitude smaller than in goldfish (Thillart et al., 1980) and two orders of magnitude smaller than in humans (Hultman, 1967). Such low white muscle glycogen may be due to the limited capacity of G. mirabilis for intense "burst" locomotion (personal observation). A second contrast to other hypoxia-tolerant species is the consistent accumulation of lactate in G. mirabilis. Normoxia-acclimated goldfish show no significant lactate accumulation under hypoxia in white muscle or red muscle, but show three-fold increases in liver (Thillart et al., 1980). In G. mirabilis, on the other hand, lactate accumulated in all tissues, with brain lactate consistently highest and liver lactate consistently lowest. High lactate concentrations in the brain indicate a high mass-specific metabolic rate, which is typical of vertebrate brain tissue, but the low lactate concentrations in the liver are surprising because vertebrate livers normally have high mass- specific metabolic rates (Hulbert and Else, 1981). This difference may indicate that the liver enters metabolic arrest during hypoxia, suppressing certain functions that normally consume so much energy. Indeed, goldfish have been shown to arrest protein synthesis in the liver during anoxia (Smith et al., 1996). Alternatively, the low lactate levels in liver could be due to the liver's role in removing lactate from the blood stream. The liver could produce lactate and subsequently process it, thereby containing low lactate levels but nonetheless conducting glycolysis and gluconeogenesis. Such processing, however, requires energy that may not be available during hypoxia, and it would seem counter-productive to burn glycogen and subsequently spend energy rebuilding storage substrates while still under hypoxic stress. This study, therefore, illustrates at least two strategies that G. mirabilis may employ in response to hypoxic stress. First, I conclude that under hypoxic conditions, G. mirabilis rely on glycolysis for ATP production, and supply the necessary substrates for glycolytic flux by depleting liver glycogen stores and shuttling them as glucose to other tissues. Second, G. mirabilis may metabolically suppress the liver so that it consumes less ATP and produces less lactate than other tissues. These strategies are a clear example of a complex systemic response to environmental hypoxia, one that integrates the needs and responses of different tissues into a unified strategy for the whole organism's survival. -Works Cited Congleton, J.L. (1974). The respiratory response to asphyxia of Typhlogobius californiensis (Teleostei: Gobiidae) and some related gobies. Biol. Bull. 146: 186-205. Hochachka, P.W., Buck, L.T., Doll, C.J. and Land, S.C. (1996). Unifying theory of hypoxia tolerance: molecular/metabolic defense and rescue mechanisms for surviving oxygen lack. Proc. Natl. Acad. Sci. USA 93: 9493-9498. Hulbert, A.J. & Else, P.L. (1981). A comparison of the "mammal machine" and the "reptile machine": energy use and thyroid activity. Amer. J. Physiol. 241: R350-R356. Hultman, E. (1967). Muscle glycogen in man determined in needle biopsy specimens: method and normal values. Scand. J. Clin. Lab. Invest. 19, Suppl. 94: 209-217. Johnson, J.A. & Fusaro, R.M. (1966). The quantitative enzymic determination of animal liver glycogen. Anal. Biochem. 15: 140-149. 1966; Johnston, I.A. & Bernard, L.M. (1983). Utilization of the ethanol pathway in carp following exposure to anoxia. J. Exp. Biol. 104: 73-78. Kepler, D. & Decker, K. (1969). Studies on the mechanism of galactosamine-I-phosphate and its inhibition of UDP-glucose pyrophosphorylase. Eur. J. Biochem. 10: 216-225. Kepler, D. & Decker, K. (1975). In Methods of Enzymatic Analysis, ed. Bergmeyer, H.U. Weinheim : Verlag Chemie, 1127-1131. Kültz, D. (1995). Distinct immediate effects of osmotic shock and heat shock on protein metabolism in Gillichthys mirabilis gill cells. Mol. Biol. Cell. 6(SUPPL.): 221A. Kültz, D. & Somero, G.N. (1994). Ion transport in gills of the euryhaline fish Gillichthys mirabilis is facilitated by a phosphocreatine circuit. Amer. J. Physiol. 268(4): R1003¬ R1012. Kültz, D. & Somero, G.N. (1995). Osmotic and thermal effects on in situ ATPase activity in permeabilized gill epithelial cells of the fish Gillichthys mirabilis. J. Exp. Biol. 198(9). 1883-1894. Lee, E.Y.C. & Whelan, W.J. (1966). Enzymic methods for the microdetermination of glycogen and amylopectin, and their unit-chain lengths. Arch. Biochem. Biophys. 116: 162-167. Lin, J.J. & Somero, G.N. (1995). Temperature-dependent changes in expression of thermostable and thermolabile isozymes of cytosolic malate dehydrogenase in the eurythermal goby fish Gillichthys mirabilis. Physiol. Zool. 68(1): 114-128. Lutz, P.L. & Nilsson, G.E. (1994). The Brain without Oxygen, Causes of Failure and Mechanisms for Survival. Austin, TX: R.G. Landes. Lutz, P.L. & Nilsson, G.E. (1997). Contrasting strategies for anoxic brain survival - glycolysis up or down. J. Exp. Biol. 200: 411-419. Schmidt, H. & Wegner, G. (1988). Glycogen phosphorylase in fish brain (Carassius auratus) during hypoxia. Trans. Biochem. Soc. 16: 621-22. Smith, R.W., Houlihan, D.F., Nilsson, G.E., & Brechin, J.G. (1996). Tissue-specific changes in protein synthesis rates in vivo during anoxia in crucian carp. Am. J. Physiol. 271: R897- R904. Sumner, F.B. & P. Doudoroff (1938). Some experiments upon temperature acclimatization and respiratory metabolism in fishes. Biol. Bull. 74: 403-428. Thillart, G.V.D., Kesbeke, F. & Waarde, A.V. (1980). Anaerobic energy-metabolism of goldfish, Carassius auratus (L.). J. Comp. Physiol. 136:45-52. Yang, T.H., Lai, N.C., Graham, J.B. & Somero, G.N (1992). Respiratory, blood and heart enzymatic adaptations of Sebastolobus alascanus (Scorpaenidae: Teleostei) to the oxygen minimum zone: a comparative study. Biol. Bull. 183: 490-499. Figure 1. Critical oxygen concentration (Pqu) for three Gillichthys mirabilis. Pait is taken as the oxygen concentration below which the fish begin to lose their ability to draw oxygen from their surroundings (here an upward slope). All three fish have a Peri of 1.0 mg/L. This value was used to determine the experimental oxygen concentration (0.8 mg/L). Figure 2. Lactate and glycogen concentrations in white muscle of fish exposed to 0, 8, 24, and 72 hours of 0.8 mg O/L hypoxia. Error bars represent onestandard deviation from the mean of each time-point. (*) mark significant changes from control (O hours). Significance was determined using a standard one-way ANOVA (p.001) with a Tukey's test (HSD = 0.54). Sample sizes are given in parenthesis on the x-axis. Figure 3. Lactate and glycogen concentrations in heart of fish exposed to 0, 8, 24, and 72 hours of 0.8 mg O/L hypoxia. Error bars represent one standard deviation from the mean. *) mark significant changes from control (0 hours). (**) mark significant changes from 8 hours. Significance was determined using a standard one-way ANOVA (pS.001) and a Tukey's test (HSD = 0.58). Sample sizes are given in parentheses on the X-axis. Figure 4. Lactate and glycogen concentrations in brain of fish exposed to 0, 8, 24, and 72 hours of 0.8 mg O/L hypoxia. Error bars represent one standard deviation from the mean. (*) mark significant changes from control (0 hours). (**) mark significant changes from 8 hours. Significance was determined using a standard one-way ANOVA (pS.001) and a Tukey's test (HSD = 0.52). Sample sizes are given in parentheses on the X-axis. Figure 5. Lactate and glycogen concentrations in liver of fish exposed to 0, 8, 24, and 72 hours of 0.8 mg O/L hypoxia. Error bars represent one standard deviation from the mean (*) mark significant changes from control (0 hours). (**) mark significant changes from 8 hours. Significance was determined using a standard one-way ANOVA (p..001 for lactate; p5.0001 for glycogen) and a Tukey's test (HSD - 0.31 for lactate; HSD- 2.87 for glycogen). Sample sizes are given in parentheses on the x-axis. Figure 6. Glycogen concentrations in liver, heart, white muscle, and brain of fish exposed to 0, 8, 24, and 72 hours of 0.8 mg O/L hypoxia. Error bars represent one standard deviation from the mean. Glycogen concentration changes significantly only in liver. Figure 7. Lactate concentrations in liver, heart, white muscle and brain of fish exposed to 0, 8, 24, and 72 hours of 0.8 mgO/L hypoxia. Error bars represent one standard deviation from the mean. Lactate accumulates significantly in all tissues throughout the time¬ course. 100 40 20 Figure 1. —e— Fish 1 (4.92g) —0— Fish 2 (17.479) —v— Fish 3 (25.089) . Vtet 00 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 O, Concentration (mg/L) 1.8 - 1.5 0.9 0.6 0.3 0.0 + Glycogen — Lactate O Hrs (7) Figure 2. 8 Hrs (7) 24 Hrs (6) 72 Hrs (6) Hypoxia Duration 4.0 3.5 3.0 - - 25 2.0 1.5 1.0 0.5 0.0- Glycogen □ Lactate O Hrs (7) Figure 3. 8 Hrs (7) 24 Hrs (6) Hypoxia Duration 72 Hrs 2.0 1.0 0.5 0.0 - — Glycogen — Lactate O Hrs (7) Figure 4. 24 Hrs (6) 8 Hrs (7) Hypoxia Duration 72 Hrs (6) O Hrs (7) Figure 5. Glycoger Lactate 8 Hrs (7) 24 Hrs (6) 72 Hrs (6) Hypoxia Duration 10 0 - O Hrs (7) Figure 6. I I 8 Hrs (7) 24 Hrs (6) Hypoxia Duration Liver — Brain Heart □ White Muscle ME 72 Hrs (6) 0 + Liver White Muscle Heart — Brain O Hrs (7) Figure 7. 8 Hrs (7) 24 Hrs (6) Hypoxia Duration 72 Hrs (6)