GSULAR OXYGE?€ TOXICiTY': THE EFFECT OF HYPERBAREC OXYGEN ON THE {N VéTRO ELECTRORETSNOGRAM OF SELECTEE} TELEOSTS. AMPHEBIANS AND MAMMALS Thesis for the Degree of M. S. . MECHEGAN STATE UNIVERSITY EOHN LAMBERT UBELS 1978 LIBRARY ' "M”"M'LMWHWmummmunm“i ABSTRACT OCULAR OXYGEN TOXICITY: THE EFFECT OF HYPERBARIC OXYGEN ON THE IN VITRO ELECTRORETINOGRAM OF SELECTED TELEOSTS, AMPHIBIANS AND MAMMALS By John L. Ubels Electroretinograms were recorded in vitro from frog, rat, goldfish, and trout eye cup preparations during ex- posure to 100% 02 at 3800 torr in order to study the possible toxic effects of hyperbaric oxygen on the neural elements and supporting tissues of the retina. The frog, rat, and goldfish ERG's were initially enhanced by hyper- baric oxygen. All components of the frog ERG were significantly attenuated by exposure to hyperbaric oxygen for six hr, as compared to the control ERG recorded under 02 at 740 torr. The b- and c-waves declined more than the a-wave indicat- ing a toxic effect of 02 on the neurons lying postsynaptic to the photoreceptors and on the pigmented epithelium, respectively. John L. Ubels Hyperbaric oxygen is highly toxic to the rat retina, causing abolishment of the ERG after 90—100 min exposure. The greatest effect is on the a—wave and slow PIII com- ponent indicating a toxic effect of 02 on the photorecep- tors. The decline of the b-wave parallels that of the a-wave. The decline of the ERG in oxygen toxicity may be due to a toxic effect of 02 on the processes responsible for maintaining transmembrane potentials in the retina. No significant change in the goldfish ERG was seen during six hr exposure to hyperbaric oxygen. Normal a- and c-waves were also recorded from the trout retina under the same conditions. The trout b—wave was attenuated, however, this decline was not significantly different from a similar decline of the control b-wave. The teleost retina is normally exposed to a P02 above N00 torr which is generated by the counter-current oxygen multiplier of the choroidal rete. Previous studies have shown the teleost retinal metabolism and enzyme activity are not inhibited by hyper- baric oxygen. The results of the present study give further evidence that the teleost retina is well adapted to high P02 and that no toxic effect is seen from exposure to P02 well above that normally encountered. This resistance to oxygen toxicity may involve the presence of protective compounds or specially adapted isozymes. OCULAR OXYGEN TOXICITY: THE EFFECT OF HYPERBARIC OXYGEN ON THE IN VITRO ELECTRORETINOGRAM OF SELECTED TELEOSTS, AMPHIBIANS AND MAMMALS By John Lambert Ubels A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Physiology 1976 DEDICATION This thesis is dedicated to my parents, Mr. and Mrs. John Ubels, in gratitude for the distinctively Christian education with which they provided me throughout elemen- tary school, high school, and college. 11 ACKNOWLEDGMENTS The author expresses his sincere thanks to Dr. J. R. Hoffert for his advice and assistance throughout this study, and especially for his aid in the preparation of this thesis.. Dr. Hoffert's assiStance in the photogra- phic reproduction of the figures is also appreciated. I am also very grateful to Mrs. Esther Brenke and Dana Zakalik for their technical assistance. Thanks also to Mrs. Brenke and to my fiance Jan DeVries for typing the rough drafts of this thesis. My committee members, Dr. P. O. Fromm and Dr. R. A. Bernard are also to be acknowledged. This study was supported in part by Grant No. EYOOOO9 from the National Eye Institute and N.I.H. Training Grant No. HLO5873. iii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . vii INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW . . . . . . . . . . . . . . . . . 4 The Clinical Uses of Hyperbaric nygen . . . . . N In Vivo Oxygen Toxicity . . . . . . . . . . . . . 6 In Vitro Oxygen Toxicity: Effects on Cellular Metabolism and Enzyme Systems . . . . . . . . . lO Ocular Oxygen Toxicity . . . . . . . . . . 13 The Resistance of Teleost Ocular Tissues to Oxy- gen Toxicity . . . . . . . . . . . . . . . . . 17 The Electroretinogram . . . . . . . . . . . . . . 20 EXPERIMENTAL RATIONALE . . . . . . . . . . . . . . . 23 MATERIALS AND METHODS . . . . . . . .-. . . . . . . 24 Experimental Animals . . . . . Tissue Preparation . . . . . . . . . . . . 25 ERG Recording Circuitry . . . . . . . . . . . . 25 Oxygen Toxicity Experiments . . . . . 32 Recording of ERG Using Recirculation Blocks . . . 32 RESULTS 0 O O O O O O O O O O O O O O O O O O O O O 3 5 The In Vitro ERG—General Observations . . . . . The Effect of Dark Adaptation on the ERG . . . Effect of Hyperbaric Oxygen on the Frog ERG . . Effect of Hyperbaric Oxygen on the Rat ERG . . Effect of Hyperbaric Oxygen on the Goldfish ERG . 58 Effect of Hyperbaric Oxygen on the Trout ERG . DISCUSSION . . . . . . . . . . . . . . . . . . . . . 75 The Toxicity of Oxygen to the Frog and Rat Retinas 77 The Frog ERG . . . . . . . . . . . . . . . . . 77 The Rat ERG . . . . . . 78 Retinal Cells Effected by Oxygen as Indicated by ERG Changes . . . . . . . . . . . . . . . . . . 80 iv Mechanisms of Retinal Oxygen Toxicity . The Resistance of the Teleost Retina to Oxygen Toxicity . . . Oxygen Supply to the Teleost Retina from the Choroidal Rete . . . . . . Mechanisms of Resistance to Oxygen Toxicity SUMMARY . . . . . . . . . . . RECOMMENDATIONS FOR FURTHER STUDY . . . . . . . LITERATURE CITED . . . . . . . . . . . . . . . . APPENDIX I: Composition of Cold Blooded Ringers APPENDIX II: Chamber Comparisons . . . . . . . Page 83 84 87 89 91 92 98 99 Table 1. LIST OF TABLES Changes in the ERG of the isolated frog retina following exposure to pure 02 at 7H0 torr and 3800 torr at 22°C . . . . . . . . . . . Changes in the rat ERG during exposure to oxygen at 7N0 torr and 3800 torr . . . . . . Effect of exposure to pure 02 at 7H0 torr and 3800 torr on the ERG of the isolated goldfish retina . . . . . . ._. . . . . . . . . . . . Changes in the trout ERG during exposure to oxygen at 740 torr and 3800 torr . . . . . . . Comparison of trout ERG amplitudes recorded at 15°C from the hyperbaric and control chambers under 02 at 7N0 torr . . . . . . . . . . . vi Page A8 59 6A 72 99 Figure l. 10. ll. 12. 13. l“. 150 LIST OF FIGURES In vitro ERG recording chamber and electrodes Schematic of in vitro ERG recording circuitry Schematic of a typical ERG showing how wave amplitudes and latency were recorded . . . . Effect of dark adaptation on the frog ERG . . Effect of dark adaptation on the goldfish ERG Effect of dark adaptation on the trout ERG . . Changes in the frog ERG during exposure to 02 at 7M0 torr and 3800 torr at 22°C . . . . . . Differences of amplitudes of frog ERG's during 02 eXposure at 7H0 torr and 3800 torr at 22°C Frog ERG's recorded in vitro during exposure to oxygen at 3800 torr . . . . . . . . . . . . Changes in the rat ERG during exposure to 02 at 7N0 torr and 3800 torr at 40°C . . . . Rat ERG's recorded in vitro . . . . at 740 The goldfish ERG during exposure to 02 torr and 3800 torr at 22°C . . . . . . Differences of amplitudes of goldfish ERG's during exposure to 02 at 7N0 torr and 3800 torr at 22°C . . . . . . . . . . . . . . . . . Goldfish ERG's recorded in vitro during expo- sure to 02 at 3800 torr . . . . . . . . . . Changes in the trout ERG during exposure to 02 at 7N0 torr and 3800 torr at 15°C . . . . . . vii Page 27 29 37 MO #2 an “7 51 53 55 57 61 63 66 69 Figure Page 16. Differences of amplitudes of trout ERG's re- corded during exposure to 02 at 7M0 torr and 3800 torr at 15°C . . . . . . . . . . . . . . 71 17. Trout ERG's recorded in vitro . . . . . . . . 7“ viii INTRODUCTION Oxygen tensions (P02) in excess of that found in air are known to have toxic effects on biological systems. One organ which has been shown to be particularly suscep— tible to oxygen toxicity is the vertebrate eye. Attention was focused on the sensitivity of the eye to high P by the O discovery in the early 1950's that the disease retroIental fibroplasia in premature infants was caused by the pure O2 atmospheres to which these infants were exposed (Patz, Hoeck and De La Cruz, 1952). Since the major effect of oxygen in retrolental fibroplasia is on the retinal blood vessels, most research has centered on the effect of high P02 on ocular circulation. These studies have shown that high P02 causes retinal vasoconstriction (Dollery g£_al., 196A). Few studies have been done on the effect of high oxygen tensions on ocular metabolism, however it has been shown that exposure to hyperbaric oxygen causes a decrease in the metabolic rate of the mammalian retina (Baeyens, Hoffert and Fromm, 1973). The teleost eye appears to be an exception to general rule of high susceptibility of ocular tissues to oxygen toxicity. In most teleosts a portion of the choriocapil- laris is modified into a rete mirabile which operates as a counter-current multiplier for oxygen. Ocular oxygen 2 tensions in excess of #00 torr are generated by this system (Wittenberg and Wittenberg, 1962; Fairbanks, Hoffert and Fromm, 1969). This suggests that the teleost has adapted to high ocular P02 and is resistant to the toxic effects of oxygen. Baeyens §£_§1, (1973) have shown that retinal metabolism in the trout is in fact enhanced by exposure to hyperbaric oxygen. This apparent resistance makes the teleost an ideal subject for the study of oxygen toxicity in that our understanding of oxygen toxicity may be aided by elucidation of its mechanism of reSistance to such toxicity. Except for the previously mentioned studies on reti- nal metabolism little work has been done on the direct effect of high oxygen tension on the function of the neural elements and supporting tissues of the retina. The elec- troretinogram (ERG) is generally accepted as a monitor of the functional integrity of the retina and thus any toxic effect of oxygen on the retina should be evidenced by changed in the ERG. Noell (1962) showed that the in vivo rabbit ERG is severely attenuated by eXposure to hyperbaric oxygen. Since this study was done in vivo, however, it is not possible to separate cardiovascular effects of high P02 from direct toxic effects on the visual cells. The work presented here is a comparative study of the effect of hyperbaric oxygen on the in vitro ERG of an amphibian, a mammal, and two teleosts. Possible cardiovascular effects of increased oxygen tension were eliminated by recording the ERG from the isolated, non-perfused retina. The questions: 1. purpose of this study was to answer the following How is the in vitro ERG affected.by the eXposure of the retina to hyperbaric oxygen? Are there species differences in the response of the ERG to hyperbaric oxygen, especially in view of the fact that teleost ocular tissues are normally exposed to high P02? If toxicity of oxygen to the retina is demon- strated, how is each component of the ERG affected and how might this be related to the various cell types in the retina? LITERATURE REVIEW The Clinical Uses of Hyperbaric Oxygen In diseases caused or complicated by hypoxia, treat- ment by hyperbaric oxygen has often been indicated in order to increase oxygen delivery to the affected tissues. 02 of mammalian arterial blood is 100 torr. The hemoglobin is 97% saturated and the oxygen in physical The normal P solution is 0.3 ml/lOO ml. Breathing pure oxygen at normal pressure will increase dissolved oxygen to only 1.5 m1/100 ml, not a substantial increase in terms of oxygen delivery to tissues. However, if through the use of hyperbaric pressures the P02 is raised to 2000 torr the dissolved oxygen will be 6.0 ml/100 ml. This increase in P02 will increase the number of 02 molecules delivered to the tissues and will also increase extravascular penetra- tion of O2 increasing the volume of tissue supplied with 02 from a given capillary (Saltzman, 1967). Studies of the effect of hyperbaric oxygen exposure on cardiac patients have shown the PO is in fact increased 2 and that the O can be used since blood lactic acid levels 2 decrease indicating a greater dependence on aerobic meta- bolism. Hyperbaric oxygen is effective in the relief of pain associated with coronary thrombosis and angina pec- toris. 5 Hyperbaric oxygen has also been used in treatment of carbon monoxide poisoning. In this case oxygen administra- tion prevents hypoxia by increasing plasma dissolved oxygen. Malignant tumors have been treated by hyperbaric oxygen because oxygen has been found to enhance the effec- tiveness of X-irradiation. The reason for enhancement is unknown. The most effective clinical use of hyperbaric oxygen is in the treatment of gas gangrene. 'The response to treat- ment is very rapid. Mortality is decreased and limbs are often saved since the development of necrotic tissue is markedly reduced. This is probably due to direct effect of oxygen on the tissue. Elimination of bacterially pro- duced gas bubbles allows increased tissue perfusion by reduction of tissue pressure. The clinical uses of hyperbaric oxygen are covered in detail in a monograph by Meijne (1970). He stresses the point that there are many unanswered questions con- cerning the clinical use of hyperbaric oxygen. There is much controversy among clinicians about its value both on theoretical grounds and because of the risk of oxygen toxicity. It appears that in many cases the danger of toxicity outweighs the beneficial effects of hyperbaric oxygen. In Vivo Oxygen Toxicity Exposure of animals to oxygen tensions above that in air has profound deleterious effects on the pulmonary, cardiovascular, and central nervous systems. Changes in whole animal metabolism also result from exposure to high oxygen tension. Since the lung is in intimate contact with the atmos- phere it appears to be the organ primarily affected during breathing of pure oxygen. Lesions associated with toxicity include edema, interstitial fibrinization, alveolar cell hypertrophy, congestion and inflammation. Robinson 23:31. (1969) exposed several species of sub-human primates to oxygen at 1 atm. Mortality occurred among baboons from days ”-8 while Macaca irus monkeys survived as long as 13 days. The pulmonary lesions listed above were observed in these animals. Several monkeys were killed and examined during the experimental period and these examinations showed that the toxic effects grow worse with longer exposure to oxygen. Humans breathing oxygen at 2 atm complain of substernal pain on inspiration, sore throat and pulmonary congestion (Widdel §§_§1,, 197A). Studies on the effect of oxygen at high pressure on alveolar surfactant show that exposure of cats to 02 at 3000 torr increases surface tension almost three times. This change takes place before the occurrence of edema or other pulmonary lesions (Beckman and Houlihan, 1973). More recent work has related this increase in surface tension 7 to increased intra-alveolar cholesterol (Bergren and Beck- man, 1975). The major effect of hyperbaric oxygen on the cardio- vascular system is vasoconstriction. Wood §E_a1. (1972) found that exposure of rats to hyperbaric oxygen results in systemic hypertension which produces left vertricular failure. They suggested that much of the pulmonary edema seen in oxygen toxicity is due to cardiac failure. In anesthetized dogs, breathing oxygen at 3 atm causes brady- cardia and decreased iliac blood flow. An increase in blood pressure was added to these effects at 5 atm (Har— denbergh gt_a1,, 1973). Exposure of humans to hyperbaric oxygen results in decreased cerebral blood flow (Lambert- sen g£_a1., 1953; Anderson and Saltzman, 1965). Bridges (1966) observed cardiac arrhythmia and ventricular fibril- lation in anesthetized rabbits exposed to hyperbaric oxygen. High PO may also affect capillary permeability. Decreased totil plasma protein and haptoglobin in rats exposed to 02 at 760 torr for seven days was attributed to increased vascular permeability (Grunbaum and Leon, 1975). The central nervous system is highly susceptible to oxygen toxicity and this may well be the root of the toxic effect of oxygen CH1 other systems. Central nervous system toxicity causes convulsive seizures. These convulsions have been observed in many animals including man. Mice convulse after as little as 30 min under 0 at 3800 torr 2 (Hoffert, Baeyens and Fromm, 1975). A similar effect is seen in cats and rats (Beckman and Houlihan, 1973). CNS toxicity is seen in man after only a few minutes of oxygen ‘breathing at A atm (Saltzman, 1967). Saltzman states that toxicity can be avoided by limiting exposure to less than 2.5 atm, however more recently Krishnamurtu and Wadhawan (197”) reported seizures in two patients after repeated exposure to oxygen at 2.5 atm for treatment of peripheral arterial disease. They suggest a progressive increase in neuronal irritability by repeated exposure to hyperbaric oxygen. That CNS effects are not limited to mammals was shown by the work of D'aoust (1969) and Hoffert g£_a1. (1975) who reported losses of equilibrium and muscular control in marine teleosts, rockfish (Sebastodes miniatus) and pil- chards (Harengula cupeola) repsectively, exposed to hyper- baric oxygen. Histological alterations in the rat brain ultrastruc- ture after exposure of the animals to hyperbaric oxygen have been observed (von Schnakenberg and Nolte, 1970). These alterations were seen even when convulsions were suppressed by drugs, indicating a direct effect of 02 on cellular structure. The toxicity of oxygen to the CNS may well be the most important effect of hyperbaric oxygen since it is probably the cause of lesions associated with high oxygen in other systems. Oxygen toxicity to the CNS results in hyperactivity of the sympathoadrenal system and 9 this sympathetic activity has been associated with pulmo- nary lesions (Bean, Zee and Thom, 1966). Indeed, electri- cal stimulation of the sympathetics causes the same changes in surfactant as seen in oxygen toxicity (Beckman and Houlihan, 1973). Wood g§_§1. (1972) stated that pulmonary effects of high oxygen tension may be secondary to cardio- vascular effects, however, since the primary cardiovascular effect of sympathetic stimulation is vasoconstriction it is conceivable that cardiovascular effects of high P are 0 due to CNS toxicity. Some authors have overlooked this point and it is stressed here that in studies like those described above it is important to keep in mind the inter- actions between organ systems which take place in vivo. Given the present state of knowledge it cannot be stated with certainty which system, if any, can be considered the primary target for oxygen toxicity. As is evident by the work already described, most oxygen toxicity studies have been done using mammals. Hoffert g§_a1. (1975) have focused attention on oxygen toxicity in lower animals and have chosen an area of study largely ignored by previous investigators: the effect of hyperbaric oxygen on metabolic rate. Oxygen toxicity was defined as a decrease in metabolic rate after exposure to high P Shrimp (Peneus setriferus), false pilchards O O 2 (Harengula cupeola), fiddler crabs (Uca pugnax), and mice (Mus musculus) show significant decrease in metabolism after exposure to 02 at 3800 torr. Metabolic rate was lO enhanced in minnows (Pimephalesgpronmelas) and crayfish (Orconestes propinquus), an interesting difference, indi- cating that not all animals respond in the same way to high P02. These changes in whole animal metabolism are the results of changes at the level of cellular metabolism, a subject to be discussed in the next section of this review. In Vitro Oxygen Toxicity: Effects (n1 Cellular Metabolism and Enzyme Systems High oxygen tension has a direct effect on cellular metabolism, causing a decrease in oxygen consumption. Stadie, Riggs and Haugaard (1945) demonstrated a decrease in metabolism of rat brain, liver, kidney, lung, and muscle tissue during six hours exposure to oxygen at 8 atm. More recently it has been shown that oxygen consumption by brain and hepatic tissue of trout, frogs, and dogs is inhibited by exposure to O et al., 1973)- 2 at lU70 torr for 2“ hr (Baeyens Numerous studies have shown that the toxic effect of oxygen is at the biochemical level. The greatest effect of high P02 is on the pathways of carbohydrate metabolism, e.g. the brain, are most susceptible to oxygen toxicity. Oxygen at 760 torr inhibits glucose and pyruvate oxidation (Haugaard, 1968). Glycolysis is inhibited directly by inactivation of the enzyme glyceraldehyde-3-P-dehydrogenase (Horn and Haugaard, 1966; Haugaard, 1968). Another glyco- lytic enzyme inhibited by high oxygen tension is lactate dehydrogenase (LDH). Baeyens, Hoffert and Fromm (197A) ll exposed homogenates of brain and liver tissue from frogs, trout and dogs to hyperbaric oxygen for 24 hr and found that LDH activity was decreased. They hypothesized that the effect of oxygen was on the active and/or allosteric site of the enzyme. Increased Km values indicated a decreased affinity of the enzyme for substrate. Further work by Baeyens (1975) gave similar results for mouse brain. As far back as 19u6 it was hypothesized that oxygen inactivated enzymes by oxidation of essential sulfhydryl groups and formation of disulfide bridges (Haugaard, 19A6). Since that time evidence supporting this idea has been presented (Haugaard, 1968). Strong support is offered by the fact that oxygen toxicity can be prevented or reversed by reduced glutathione (Horn and Haugaard, 1966; Baeyens, 1975). Glutathione is a strong reducing agent and may maintain the SH groups of enzymes in their reduced state. The proposed mechanism of oxygen toxicity is: 2 En-SH oxidation > EnS-SEn + 2(H) active :4 reduction inactive (Baeyens, 1972). Addition of reduced glutathione (G) to an enzyme system (En) could reverse the toxic effect of oxygen by the reaction: EnS-S-En + 2GSH ———————e 2 EnSH + GS-SG (._— Reduced glutathione could protect against oxygen toxicity since it would be more easily oxidized than the enzyme SH 12 groups. Enzymes other than glycolytic enzymes, such as the SH group containing pyridoxal enzymes and ferredoxins are also inhibited by oxygen. High oxygen can also cause extensive damage by peroxidation of lipids. These sub- jects have been extensively reviewed by Haugaard (1968). If carbohydrate metabolism is inhibited one would expect reduced production of high energy compounds. This is in fact the case as shown by Horn and Haugaard (1966) who observed that ATP production is inhibited when rat heart homogenate is incubated with 100% oxygen. ATP con- centration in rat heart muscle was decrease 15.7% by exposure to hyperbaric oxygen (Wood et_§l., 1972). Another effect of hyperbaric oxygen exposure on cellular metabolism is of interest in view of its convul- sive effects. It has been shown that exposure of rats and mice to hyperbaric oxygen causes decreased levels of gamma-amino-butyric acid (GABA) in the brain (Wood, Watson and Murray, 1969; Radomski and Watson, 1973). GABA is an inhibitory neurotransmitter and anti-convulsant (Goodman and Gilman, 1975). The effect of decreasing its level in the brain is obvious. Decreases in GABA are detectable before the onset of convulsions (Radomski and Wood, 1973) which could explain the increased sympathetic activity responsible for other lesions associated with oxygen toxi- city, i.e. in the lung. The changes in GABA metabolism in the brain are due either to inactivation by oxygen of glu- tamic acid decarboxylase which synthesizes GABA or activation 13 of GABA alpha-oxyglutarate transaminase which breaks it down (Radomski and Watson, 1973). Since the toxic effect of oxygen at the cellular level in most cases is enzyme deactivation the former hypothesis seems more plausible. Ocular Oxygen Toxicity The emphasis of the present study is on ocular oxygen toxicity. Therefore descriptions of the effects of high PO2 on ocular tissues have not been included in the pre- vious sections of this review so that attention might be focused on ocular oxygen toxicity at this point. In the early 19A0's a disease which became known as retrolental fibroplasia appeared in premature infants and reached epidemic proportions by the early 1950's. This disease was characterized by irreversible constriction of retinal arteries, cessation of development of new arteries, and proliferation of fibrous tissue in the retina, result- ing in blindness. Research efforts using both experimen- tal animals and clinical studies revealed that retrolental fibroplasia was caused by the pure oxygen atmosphere to which premature infants were exposed (Patz et_al., 1952; Ashton, Ward and Serpell, 1953; Patz, 1965). Since that time much research on ocular oxygen toxicity has centered on the circulatory and histologic changes in the eye asso- ciated with high PO . As in other systems high PO causes 2 2 reversible vasoconstriction in the mature retina. Dollery et a1. (196M) reported greater constriction of retinal 1A arteries at 2 atm than at 1 atm during oxygen breathing by humans. The percentage reduction in diameter was greater in smaller vessels than in larger ones. Anderson and Saltzman (1965) concurred with the findings of Dollery e§_§l,(l96u) and in addition reported that retinal veins constrict more than retinal arteries during oxygen breath- ing. Both groups suggest that this vasoconstriction may have protective value. Other workers however disagree with this idea, saying that the vasoconstriction seen with oxygen breathing should not restrict retinal oxygenation (Nichols and Lambertsen, 1969). They also suggest that the retinal vasoconstriction observed in oxygen breathing may at least initially be due to the hypocapnia which accom- panies oxygen breathing. A moderate decrease in blood flow to the retina has been observed in monkeys exposed to hyperbaric oxygen (Eperon, Johnson and David, 1975). Exposure to high oxygen tensions may also affect the permea- bility characteristics of the choriocapillaris. Exposure of dogs to 100% oxygen at 1 atm for “8-50 hr is known to cause retinal and uveal detachment due to leakage of ede- matous fluid from the choriocapillaris into the subretinal space (Yanoff, Miller and Waldhausen, 1970). Histological studies have also shown changes in ocu- lar tissue caused by high P02. The changes differ from circulatory effects in that they are in most cases irre- versible and may take place at relatively low P Lucas 0 O 2 and Trowell (1958) found that rat retinas in tissue culture l5 became necrotic under 60% oxygen at normal pressure while retinas under room air survived. Histological examina- tions of retinas from rabbits exposed to 100% 02 for “8 hr and 80% O for 100 hr at ambient pressure show degenera- 2 tion of 70% of the visual cell (rods and cones) population (Noell, 1955; 1962). Exposure of the animals to oxygen at 1.66 atm resulted in the death of 100% of the visual cells in 2“ hr. Examination of rabbit retinas four days after exposure to 3 atm oxygen for 1L; hr shows that the outer nuclear layer is reduced in thickness and that pycno- tic nuclei are present. A few days later the visual cells have disappeared by autolysis, however, the ganglion and biOpolar cells appear normal (Noell, 1962). Bresnick (1970) essentially repeated Noell's work and showed by electron microscopy that ultrastructural changes appear in the visual cells before the changes observed by Noell. Corneal and lens tissues are also affected by hyper- baric oxygen. Thinning of the corneal epithelium and dar- kening of its cytoplasm were observed in guinea pigs after exposure to oxygen at 3 and 5 atm. Pycnosis and loss of lens epithelium nuclei also occurs (Nichols e§_al., 1972). At the time of Nichols and Lambertsen's (1969) exten- sive review of ocular oxygen toxicity little work had been done on the effect of high oxygen tension on ocular metabolism. They state, The rates of inactivation of retinal enzymes, and the possible influences of concurrent decreases in nutrient flow and exhaustion of available energy 16 stores should provide important tOpics for study of oxygen effects upon vision. Since that time several studies of oxygen toxicity and retinal metabolism have been done in the laboratory of J. R. Hoffert at Michigan State University. Oxygen at 1H70 torr for 2“ hr had no effect on the metabolic rate of frog retinas (Baeyens et_§l., 1973), however, retinal LDH activity was decreased by exposure to 02 at 740 torr (Bae- yens et_§l., 197“). There is a marked decrease in both metabolic rate and LDH activity of dog retinas exposed to hyperbaric oxygen (Baeyens et_§l,, 1973; 1974). Studies of enzyme kinetics indicate that the toxic effect of oxy- gen is at the active site of the LDH molecule causing a decreased affinity of enzyme for substrate. In contrast, teleost retinal metabolism and LDH activity were enhanced by hyperbaric oxygen. The effect of hyperbaric oxygen on the retina may also be studied by observing changes in the electroretinogram (ERG) which is generally accepted as being a good indica- tor of the functional integrity of the retina. To this author's knowledge only Noell (1955; 1962) and Bridges (1966) have reported studies of this type. Noell (1955) recorded the ERG from rabbits in vivo under normobaric con- ditions and found that the ERG was significantly atten- uated after 20-36 hr under 100% oxygen, and after 50-90 hr under 80% oxygen. He reported that full recovery of the ERG was possible if exposure was not prolonged. At ambient pressure and 100% oxygen ERG attenuation occurred before 17 histological manifestations of visual cell death (Noell, 1962). Noell (1962) also studied the effect on the ERG of exposure to pure oxygen at 3-7 atm. He gives data only for the b-wave which at 7 atm disappeared within A2 min. Rate of decline changed directly proportionately with P02. The latency of the effect was 20 min at 7 atm and 100 min at 3 atm. The change in the ERG was often seen before the systemic effects of hyperbaric oxygen such as convulsions occurred. Bridges (1966) essentially repeated the work of Noell. He used hyperbaric oxygen ranging from 2.5 to 7 atm. Attenuation of the a-wave as well as the b- wave was reported in this study. The time to disappear- ance of the ERG was an inverse function of pressure with the a-wave lasting longer than the b-wave. Initial enhance- ment of the ERG was also reported. As noted earlier these studies were done in vivo and thus direct effects of oxygen on the retinal cells cannot be completely separated from possible circulatory and other systemic effects. The Resistance of Teleost Ocular Tissues to Oxygen Toxicity The teleost retina is normally enveloped by oxygen tensions in excess of MOO torr (Wittenberg and Wittenberg, 1962; Fairbanks et_§l,, 1969). Extended exposure to such high oxygen tensions would be toxic to most tissues. These high oxygen tensions are generated by the counter- current oxygen multiplier known as the choroidal rete mirabile. The concentrating potential of the rete may be 18 appreciated by the fact that the normal P0 of rainbow 2 trout (Salmo gairdneri) arterial blood is only 21 torr (Fairbanks, 1970). Electroretinographic studies have shown that the trout retina is dependent on these high oxygen tensions for normal function (Fonner, 1973; Fonner, Hoffert and Fromm, 1973). A mechanism for oxygen concentration has been pro- posed (Fairbanks e§_§l., 1974) which depends on the enzyme carbonic anhydrase (CA). Erythrocyte CA in the chorioca- pillaris hydrates C02, producing hydrogen ions and releas- ing 0 into the plasma by the Bohr and Root effects. In 2 the retina bicarbonate neutralizes lactic acid and CA dehydrates H200 forming CO which diffuses into the chor— 3 2 iocapillaris again raising the P of the venous blood by 0 the Bohr and Root effects. Oxygefi diffuses from the venous side of the rete into the arterial side and is carried to the retina. In the venous side of the rete CA, which is probably located in the walls of the vessels, hydrates C02 thereby preventing short circuiting of the system by dif- fusion of CO2 into the arterioles. The use of the CA inhibitors, acetazolamide and CL-11,366, showed that this short circuiting does in fact occur when CA is inhibited (Fairbanks e§_§1., 1974; Fonner et_§l., 1973). Recent histochemical studies have shown that high levels of CA are present in the retial vessels (Eldred, 1975). The presence of high PO at the retina suggests that . 2 the ocular tissues of teleosts may be resistant to oxygen l9 toxicity. Several studies on the effect of hyperbaric oxygen on metabolism of LDH activity in the teleost retina have indicated that this is the case. Baeyens et_§1. (1973) showed that trout retinal meta- bolism is enhanced by exposure to hyperbaric oxygen. Incubation of retinal tissue under 100% 02 at 154 torr, 400 torr and 1470 torr for 24 hr causes a linear increase in metabolic rate with rising P Metabolism of lens, 0 . cornea and retina of the white giunt (Haemulon plumieri) is enhanced by exposure to 02 at 3040 torr (Hoffert, Baeyens and Fromm, 1973). In general the enzyme LDH is inhibited by molecular oxygen. It has been shown however that trout retinal LDH activity is not affected by exposure to oxygen at 1 atm (Baeyens and Hoffert, 1972) indicating that there is a mechanism which prevents oxidation of the enzyme. Trout retina LDH activity is in fact enhanced by exposure to oxygen at 1470 torr for 24 hr (Baeyens e£_§l., 1974). This evidence indicates that the teleost has deve- lOped a mechanism of resistance to ocular oxygen toxicity. The dependence of the teleost retina on high P is proba- O 2 bly related to the fact that it has an avascular retina and therefore the whole retina depends on 0 diffusion from 2 the choroid. It should be noted that not all teleost tissues and enzymes are resistant to oxygen toxicity (Baeyens et al., 1973; 1974). 20 The Electroretinogram The electroretinogram (ERG) has been used as a moni- tor of retinal function since its discovery by Holmgrem in 1865 (Armington, 1974). The ERG is a complex response of the retina to a photostimulus which can be recorded by an active electrode on the cornea or in the vitreous humor and a reference electrode behind the eye. The response is relatively slow and is made up of a negative deflection known as the a-wave (by convention negative is a downward deflection in electroretinography) a positive b-wave, and in some animals a slow deflection known as the c-wave which may be positive or negative depending on the species. The ERG is the result of the summation of the responses of the various cell types of the retina to a photostimulus (Brown, 1968). Granit (1933) attributed the ERG to the summation of components which he labeled PI, PII, and PIII based on the order of their disappearance under ether anesthesia. Roughly, PIII corresponds to the a-wave, PII to the b-wave, and PI to the c-wave. The fact that the ERG is a simultaneous response from several cell types gives it good potential as an experi- mental tool, however, this same complexity made data ana- lysis difficult until some knowledge of the cellular origins of the ERG was gained. This research was aided by the develOpment of microelectrodes and intraretinal and intracellular recording techniques. The vertebrate receptor cell responds to light by hyperpolarization (Baylor 21 and Fourtes, 1970; Tomita, 1976). These hyperpolariza- tions correspond to the a-wave (PIII). Experiments on signal transmission along the rods have provided evidence that the photoreceptor response accounts for a major por- tion of the a-wave (PIII) (Penn and Hagins, 1969). Various authors have shown by intraretinal record- ing that the b-wave originates in the inner-nuclear layer (Brown, 1968). Intracellular recording shows that the only cells which respond to photostimulation with poten- tials corresponding to the b-wave are the Mfiller (glial) cells (Miller and Dowling, 1970). The Miller cells are not involved in signal transmission in the retina but act as potassium electrodes responding to increases in K+ ion concentration in the extracellular fluid. The bipolar cells may make a major contribution to the rise in extra- cellular potassium (Armington, 1974) however other cells are probably involved also. Tomita (1976) points out that although the Mfiller cells do make a significant contribution to the b-wave the question of whether it is exclusiVely of Miller cell origin is still unanswered. The pigmented epithelium, which lies behind the receptor cells and is not actually part of the retina, has been implicated as the source of the c—wave (PI). Schmidt and Steinberg (1971) using intracellular techniques recorded light induced responses from pigmented epithelium cells which were identical to the c-wave. The responses were shown to be rod dependent. 22 The ERG has been reviewed extensively by Brown (1968) and Armington (1974). Most recently, Tomita (1976) has published an excellent review of the electrophysiology of the retinal cells. EXPERIMENTAL RATIONALE Experiments were designed to study the effect of hyperbaric oxygen on the in vitro ERG. Since the ERG is a good monitor of retinal function, any effect of O on 2 the retina should be reflected by changes in the ERG. All experimental preparations were exposed to 100% 02 at 3800 torr. The control preparations were placed.under 02 at 740 torr because a satisfactory response could not be maintained under room air. Oxygen toxicity was defined as a decrease in the amplitude of ERG recorded from eyes exposed to hyperbaric oxygen as compared to control. These oxygen toxicity studies were performed using the eyes from teleosts (trout and goldfish), amphibians (frogs), and mammals (rats). Teleosts were chosen because the retina is normally exposed to oxygen tensions in excess of 400 torr and therefore some resistance to oxygen toxicity was expected. The amphibian was chosen because it is a poikilotherm whose eye is not normally exposed to high tensions. The mammal was chosen since it has been shown to be particularly susceptible to oxygen toxicity and also so that a possible relationship between body temperature and susceptibility to oxygen toxicity might be investigated. 23 MATERIALS AND METHODS Experimental Animals Southern grass frogs (Rana pipiens), 15-20 cm long were obtained from Mogul-Ed Corp. (Oshkosh, Wis.) and were kept at room temperature (21 1 1°C) in a moist aquarium. The frogs were maintained under natural photoperiod. Goldfish (Carassius auratus) (7-9 cm) were pur- chased from a local pet dealer. They also were maintained under natural photoperiod at room temperature in dechlor- inated tap water in an Instant Ocean aquarium (Aquarium Systems Inc., Wickliffe, Ohio). Rainbow trout (Salmo gairdneri) (100-200 g) from Midwest Fish Farms (Harrison, Mich.) were maintained in fiberglass tanks at 12 : 1°C. Dechlorinated tap water flowed continuously through the tanks and was aerated by compressed air filtered through activated charcoal. The photoperiod was 16 hr light, 8 hr dark. Rats (Long-Evans strain) (200 g) were obtained from Charles River Breeding Laboratories (Wilmington, Mass.). PhotOperiod was 12 hr light, 12 hr dark. Long-Evans rats were chosen, rather than the more common albino Sprague- Dawley strain, since their eyes are normally pigmented and it was therefore assumed the ERG recorded from these eyes 24 25 would be more comparable to the responses from the other animals used in these experiments. Tissue Preparation Frogs were killed by decapitation, fish by cervical section, and the rats by cervical dislocation. The ERG was recorded from an eye cup which was pre- pared by the following method: A small incision was made at the junction of the cornea and sclera of an enucleated eye. An iris scissors was inserted into this incision and the cornea and iris were removed by cutting around the circumference of the eye. The lens was then removed with a forceps. In order to reduce the distance for diffusion of oxygen, the aqueous humor and as much of the vitreous humor as possible were removed by blotting with tissue pa- per. Trout vitreous humor is a semiegel and therefore it was necessary to remove it by use of a forceps and iris scissors. This process was completed for both eyes within three minutes. ERG Recording Circuitry The eye cup preparation was placed in a Lucite record- ing chamber which was partially filled with non—nutrient Ringer solution (Appendix I). The eye was positioned over a hole in a shelf mounted in this chamber so that the back of the eye made contact with the Ringer solution (Figs. 1 and 2). 26 .Mmosmno oapmnkmdzn on» ma pcsonwxomn on» CH .mmwpfinn pmwm an whoa on» so Amnemzo who map 0» pwuomcsoo ma paw coauSHOm newcfim was mononpomam HmEOHmo map mpaon unwfih no message oapmmfia one .moponpomam new nonsmso wsapnoomp 0mm onpfi> EH .H ossmflm H mmDUHm 28 Figure 2. Schematic of in vitro ERG recording circuitry. Two conductor shielded cable lead from the pair of calomel electrodes to the P16 preamplifier and was connected to ground. For further details see the text. 29 in Vitro ERG Recording Circuitry . PRO ync. .‘ in DC I out . one. an e: :' Cf 13;. Colonial Eliotrodo r ' i | | ' .ll‘- ‘anr l|-‘ | I L I ' ° Bride. | | i ' 7" | R’ «'3 i I l \ I ' ea : : Ringer's I L-......_._- .......__.__._._...| Copper Screen FIGURE 2 30 The active electrode, which was placed in a holder positioned over the eye cup, was lowered into the remain- ing vitreous humor. The electrodes used in these experiments were hand drawn from 2 mm I.D. soft glass tubing (0.2 m1 Prothrombin Pipette, Dade, Miami, Fla.). The tip diameters of these electrodes were approximately 0.5 mm. The elec- trodes were filled with 4 g% agar in non-nutrient Ringer solution. These electrodes were connected to agar bridges made from PE 280 tubing also filled with 4 g% agar in Ringer solution. The electrodes and bridges were stored in Ringer's to prevent drying. The active electrode was connected to Grid 1 (G1) of a Grass P16 AC-DC preamplifier (Grass Instruments, Quincy, Mass.) by an agar bridge and a Ringer's filled calomel electrode (S-30080-l7 miniature reference electrode, Sargent-Welch Scientific 00., Detroit, Mich.) (Figs. 1 and 2). The reference electrode was an agar bridge immersed in the Ringers solution which bathed the back of the eye cup and was connected to Grid 2 (62) of the preamplifier by a Ringers-calomel electrode. The experimental preparations and electrodes were placed in a light tight electrically shielded cage connected to common ground. This cage was thermally insulated and temperature could be regulated between 15°C and 40°C. Due to the slow response of the c—wave of the ERG it was necessary to use a direct-coupled output from the P16 31 preamplifier. The rise time was set at 10 msec (1/2 amp high frequency = 30 Hz). In order to minimize the pick up of extraneous signals the electrodes were wired in the standard push-pull configuration. A double-pole, double- throw switch between the preamplifier and the preparations allowed selective recording from either the control or experimental preparation. Photostimulation was by means of a Grass PS2 photo- stimulator (Grass Instruments, Quincy, Mass.). The photo- stimulator was set at intensity 8 inall experiments. The light source was placed about 70 cm above the experimental preparations. The light was positioned so that both the control and experimental preparations were simultaneously and evenly illuminated. For a discussion of illumination by this instrument see Fonner (1973). ERG's were displayed on channel 1 of a storage oscil- loscope (Tektronix Instruments, Model 5031, Beaverton, Ore.) which was set in the direct—coupled mode. The oscil- loscope was synchronized with the PS2 photostimulator and the stimulus was recorded by a photocell and displayed on channel 2 of the oscilloscOpe. The instrumentation was calibrated by an internal calibration signal in the P16 preamplifier. Measurements of the amplitudes of the component waves of the ERG were made directly on the oscilloscope screen and selected responses were photographed with a 35 mm camera. The nega- tives were projected onto paper and traced by hand for reproduction in this dissertation. 32 Oxygen Toxicity Experiments The in vitro ERG could not be maintained under room air therefore, it was necessary to place the control pre- paration in a sealed Lucite box under 100% 02 at 740 torr. The experimental preparation was placed in a Bethle- hem Environmental Chamber (Model H-70-A, Bethlehem, Pa.) which was purged with pure O2 and pressurized to 3800 torr (5 atm absolute). The chamber was fitted with a glass window which allowed photostimulation of the preparation. In order to facilitate statistical analysis of the data the right and left eyes of the experimental animal were select- ed randomly for either the control or experimental prepar- ation and recordings were made alternately from the preparations. Test flashes were begun after sufficient time for dark adaptation. In earlier experiments on frogs the eyes were stimulated every hour for 6-7 hr. In later experi- ments on fish stimuli were given every 1/2 hr for 6 hr. The response of the rat retina to hyperbaric oxygen was so rapid that it was necessary to record the ERG every 5 min. Frog and goldfish experiments were run at room temperature (20-22°C). Trout experiments were run at 15°C and rats at 40°C. Recording of the ERG Using Recirculation Blocks During this study it was questioned whether or not the eye cup was a satisfactory preparation for recording the trout ERG. Because of the relatively large size of the 33 eye and because the vitreous humor is difficult to remove it was questionable whether sufficient oxygen could diffuse to the retina. As an alternative method of recording the ERG a pair of modified Lucite Ussing blocks was used. In preparation for recording, the eye was dark adapted, enucleated, and dissected in a petri dish with Ringer solution under red light. The cornea and lens were removed and a disc of sclera including the optic nerve was removed from the back of the eye. The retina was then teased away from the remaining ring of sclera including the optic nerve was removed from the back of the eye. The retina was then teased away from the remaining ring of sclera. The pig- mented epithelium which pulls away from the retina during dark adaptation was peeled off of the retina and as much of the vitreous humor as possible was removed. A millipore filter (Millipore Corp., Bedford, Mass.) was placed on the vitreal side of the retina for support and this prepar- ation was clamped between the reservoirs of the recircula- tion blocks. The blocks were filled with media and connected by Ringer's-4 g% agar bridges to the calomel cells as previously described. The recirculation blocks have several advantages for ERG recording: (1) the blocks may be filled with differ- ent kinds of media and various substances and drugs may be added to the media. In this study several modifications of Ringer solution were used without success, however, an 34 ERG could be recorded when the blocks were filled with Tissue Culture Media 199 (Earles Base) (Grand Island Bio- logical Co. constantly siphon the ERG may be P02 at the , Grand Island, N.Y.). (2) Since the media is recirculated and aerated by an air lift effect of various gases and gas mixtures on the studied. (3) It is possible to control the retina. Although several trout ERG's were recorded using the recirculation blocks, no oxygen toxicity experiments were done due to difficulty with the operation of the air lift siphons in the hyperbaric chamber. RESULTS The In Vitro ERG—-General Observations The ERG's recorded in this study were typical of both in vivo and in vitro examples found in the literature. The a-wave and b-wave were recorded in all cases and a c-wave usually appeared upon dark adaptation. Because of high intra-species variation in wave amplitude, especially in the case of the frog it was necessary to express the graphic data as percent of initial amplitude, i.e. the first recording made after the eye was considered to be dark adapted. Wave amplitudes were measured as shown in Figure 3. An ERG could be recorded from frog and fish eyes for at least six hr if the preparation was under oxygen at 740 torr. Rat eyes could be maintained for only 2-3 hr under oxygen at 740 torr. It was found that if the chamber was purged with air the rat ERG was abolished within three min. If the chamber was again purged with oxygen the ERG recovered within five min. Actual dark adapted amplitudes of the ERG components are given in Tables 1, 2, 3, and 4. Except in the case of the trout, initial enhancement of the ERG was observed under hyperbaric oxygen. 35 Figure 3. 36 Schematic of a typical ERG showing how wave amplitudes and latency were measured. a is a-wave amplitude b is b-wave amplitude c is c-wave amplitude Lb is b—wave latency S is photostimulus 37 Schematic Representation of the ERG b c + - 'i —————— ""— ° 4 _r-'-b+ FIGURE 3 38 Differences in latency and waveform of the ERG from species to species were observed. The frog retina gave the slowest response with a b-wave latency of 500 msec. The b-wave latency was about 200 msec in the other species used. In frog ERG the potential returns nearly to base line before the beginning of the c-wave. This is not the case with the goldfish and trout. No c-wave was recorded from several of the control trout retinas. The c-wave was not present in the rat ERG. The response from the rat eye was unique however, in that a negative wave followed the b-wave. This late negativity is probably part of PIII (Weidner, 1976). The Effect of Dark Adaptation on the ERG The ERG is known to change during dark adaptation, therefore for each species used in this study it was necessary to determine the time required for dark adapta- tion of the isolated retina (Armington, 1974). Eyes were prepared for recording under white light and an ERG was recorded immediately after the experimental chamber was darkened. Subsequent recordings were made every five min until no further change in ERG was seen. The dark adapta- tion studies on frogs, goldfish and trout are shown in Figure 4, 5, and 6. Based on these results, data collection in oxygen toxicity experiments was begun after sufficient time for dark adaptation so as not to confuse changes in the ERG due to dark adaptation with changes caused by oxygen toxicity. Figure 4. 39 The effect of dark adaptation on the frog ERG. Amplitudes are expressed as percent initial and means i SE for 4 experiments are plotted against time (min). Initial amplitude at beginning of dark adaptation, a-wave = 21.5 uV, b-wave = 250.0 uV. Recorded under 0 at 740 torr at 22°C. In subsequent studies dark adaptation was assumed to have occurred by 30 min. 9‘ Initial N O O 5 9 40 32 room o-wovc 5 (0 I526 252K) 354B T|m0(min) W W0" 2'5 :6 I5 26 2'5 3'0 33 4b Tlmc(mtn) FIGURE 4 Figure 5. 41 The effect of dark adaptation on the goldfish ERG. Amplitudes are expressed as percent ini- tial and means :ISE for 4 experiments are plotted against time (min). Initial amplitude at beginning of dark adaptation, a-wave 14.3 uV, b-wave 23.0 uV. Recorded under 0 at 740 torr at 22°C. In subsequent studies dark adaptation was assumed to have occurred by 30 min. initial % % Initial 3001 200‘ iOO< 42 300. 200% I004 FIGURE 5 o-wovo 5 no {5 26 23 50 Time (min) b—wave 5 to IE 2'0 25 so Time (min) Figure 6. 43 The effect of dark adaptation on the trout ERG. Amplitudes are expressed as percent initial and means i’SE for 4 experiments are plotted against time (min). Initial amplitude at beginning of dark adaptation, a-wave 14.5 uV, b-wave 16.3 uV. Recorded under 0 at 740 torr at 22°C. In sub- sequent studies aark adaptation is assumed to have occurred by 30 min. x Initial 600. 500‘ 400< 300 200‘ IOO 44 b-wavo O - HOV. {0 2b 36 4b 30 Time (min) FIGURE 6 45 Effect of Hyperbaric Oxygen on the Frog ERG Exposure of the frog retina to oxygen at 3800 torr caused a definite decline of the ERG as compared to con- trol. Wave amplitudes were expressed as percent initial amplitude and the results of all experiments were averaged. Lines were fitted to the data by linear regression (not including initial 100% reading). The reader is cautioned that this method may negate transient changes which take place during the time course of the experiment. Significant attenuation of the a—wave is seen in both the experimental and control preparations (Table 1). How- ever, the decline of the a-wave under hyperbaric oxygen begins at 2 hr, while the a-wave of the control preparation does not decline until 4 hr (Fig. 7). It should be noted that in both preparations an a-wave could always be recorded throughout the experiment and that the attenuation was not as great as that of the b—wave and c-wave of the retinas exposed to hyperbaric oxygen (Fig. 7; Table I). The b- and c-waves recorded from eyes under hyper- baric oxygen were significantly attenuated and sometimes abolished while normal responses could be recorded from the control preparations throughout the experimental period (Fig. 7). Note the sudden drop in b-wave amplitude between 4 and 5 hr. Hyperbaric oxygen causes the retinal cells to respond more uniformly than those under normo- baric conditions as indicated by the small standard errors on the experimental data. Figure 7. 46 Changes in the frog ERG during exposure to O at 740 torr and 3800 torr at 22°C. Points repr — sent means : SE of 9 experiments with wave ampli- tudes expressed as percent initial amplitude. Lines were fitted by linear regression. A. a-wave 740 torr. Slope = -5.63a, rxy = 0.351b, n = 57, initial amplitude 30.4 i 4.2 uV. a-wave 3800 torr. Slope = -9.39a, rxy 0.56b, n = 59, initial amplitude 25.4 i 2.9 uV. Slopes of plots A and B are not significantly different from each other at 5% level. b-wave 740 torr. Slope = 9.80a, rxy = 0.3329, n = 52, initial amplitude 287.0 1 78.0 uV. b-wave 3800 torr. Slope = -15.38a, rxy = 0.801b, n = 55, initial amplitude 575.8 i 90.1 uV. Slopes of plots 0 and D are sig- nificantly different from each other at 5% level. c-wave 740 torr. Slope = 6.330, rxy = 0.192d, n = an, initial amplitude 313.4 i 123.1 uV. c-wave 3800 torr. Slope = -15.08a, rxy = 0.790”, n = nu, initial amplitude 860.0 : 130.9 uV. Slopes of plots E and F are sig- nificantly different from each other at 5% level. a. SlOpe significantly different from 0 at 5% level. b. Correlation significantly at 5% level. c. Slope not significantly different from 0 at 5% level. d. Correlation not significant at 5% level. 1:7 O 740 Torr IZOi 2 .m‘ 02 ””70" L . 0 co. _ '0‘ 3 , I 0 . .3. 40. i 40. I a! . fl ' o . . . . . . I 0 . r z 3 4 5 a 7 ' 2 3 4i 5 ' fl 1mm Thur») ‘ ' 22°} it 02 740 Ton loo. 3 D 8 3 I40 ’ . loo 9? Timaiiw) 02 740 TO'V 2 ma _. . :9. 99* 3:20. 5 ’ 5 o I 9' ‘0‘ i am A . r . , , , a . I . . . I ' 2 3 ‘ 5 a 7 i 2 3 4 5 a 7 Tina (M) Tim. (It!) FIGURE 7 48 .HeepH am on eoeeaeeono oeoofioaemam * an H I a: m u U Am i no mm + one: am.moa H m.Hmoi o.sm H o.omH o.oma H o.ooo ooom meagre o.mHH H o.om + o.mo H o.omm H.mmH H o.mam one eemzuo .o.~o H H.Hmmu m.oH H o.oo H.oo H o.m~m ooom meagre m.aw H 0.5m I :.mm H 0.0mm o.w> .H 0.5mm on» m>m3rn mo.m H m.o I H.m H N.AH o.m H o.mm ooom meagre so.H H H.HH I H.m H o.om m.o H o.om one meagre A>av p A>av an m A>av an H m .oomm no and» ooom one use» ooh no 0 mass ou chamomxm wcfizoaaom scopes mosh pmpmaomfi on» mo wmm 0:» CH mmwcwno .H manna 49 Figures 8 and 9 and Table 1 give further evidence that the ERG is attenuated by hyperbaric oxygen. Effect of Hyperbaric Oxygen on the Rat ERG Changes in the rat ERG took place rapidly compared to the other animals and as pointed out earlier the prepara- tions did not last as long. Therefore recording was begun 15 min after darkening the chamber. Dark adaptation was complete by 35 min (Fig. 10). For this reason that data collected does not lend itself well to the same statistical analyses used for the other animals in this study. Definite differences, however, are seen between the responses of the preparations under normobaric oxygen and hyperbaric oxygen. While the control a-wave held its dark adapted ampli- tude for 2 hr the a-wave was greatly attenuated and often abolished after 95-105 min under hyperbaric oxygen (Fig. 10). Initially the b-wave recorded under hyperbaric oxy— gen was greatly enhanced over the control b-wave. After dark adaptation, however, both b-waves declined. The waveform of the control ERG (Fig. 11) was like the ERG recorded by Weidner (1976) during hypoxia, indicating that perhaps oxygen at 740 torr is not sufficient to maintain the rat eye cup. The wave-form of the ERG under hyperbaric oxygen has typical a—wave and b-wave components throughout most of the experiment, but the b—wave declines at a fas- ter rate than the control b-wave and is often abolished by 50 Figure 8. Difference of amplitudes of frog ERG's during 0 exposure at 740 and 3800 torr at 22°C. Differ- ences in wave amplitudes (control minus experi- mental) for 9 paired experiments were averaged and plotted against time (hours). Points represent Mean i SE. Lines were fitted by linear regression. A. a-waze. Slope = 3 609, rxy = 0.130b, n= 0 B. b-wazg. Slope = 26.u3°, rxy = 0.55ud, n: . C. c-wagi. Slope = 18.720, rxy = 0.445d, n= . a. SlOpe not significantly different from 0 at 5% level. b. Correlation not significant at 5% level. 0. Slope significantly different from 0 at 5% level. d. Correlation significant at 5% level. X Dif. Amp. = Mean difference in amplitude (control-experimental) 51 Tlmo (hr) '60 1 IZOI 804 X Dif. Amp. X Dif. Amp. Tim. (M) FIGURE 8 Figure 9. 52 Frog ERG's recorded in vitro during exposure to oxygen at 3800 torr. Recordings were made over a 6 hr period at 22°C. The response from the control preparation was comparable to those seen in part A and was maintained throughout the 6 hr period. Note change in amplitude calibration in parts C and D. 53 >1 OON S 5 0.5 hr 5 sec >1 OON hr >3 OOH 5 sec 5 hr >n OOH 5 sec 6 hr FIGURE 9 Figure 10. 54 Changes in the rat ERG during exposure to O at 740 torr and 3800 torr at 40°C. Points repge- sent means + SE for 14 control eyes and 11 experimental eyes. Curves were fitted by hand. A. a-wave 740 torr. Initial amplitude 16.1 i 4.9 uV. B. a-wave 3800 torr. Initial amplitude 15.0 + 1.9 uV. C. b-wave 740 torr. Initial amplitude 10.8 + 2.6 uV. D. b-wave 3800 torr. Initial amplitude 32.3 + 6.6 uV. E. PIII 740 torr. Initial amplitude 39.4 i 9.5 uV. F. PIII 3800 torr. SIOpe = -0. 91 (significantly different from 0 at 5% level), rx = 0.520 (significant at 5% level), n = 3’82, initial amplitude 73. 4 i 10.1 uV. $ IIIINOI % Initial Time (min) C 0274010" 400 :g 2200 a o‘fifvvvvvva—fi I5 as 55 75 95 us O2 740 To" mi .0. .o. I 0 'fh - - - - v - v - I5 35 55 75 95 "5 TIIMImIn) 200« I50« I00 ; O "a - Tim. (min) I5 35 55 73-9316: FIGURE 02 3.00 TO" I50! 2 IOOI 3 . u 50‘ 040%. . - . _ I5 55 55 75 95 Timoimin) 300« 3200. :uoo‘ O TIInaImin) 02 3500 To" .0011 . E 50« i I .0 204 0,;331-.,--fi i5 35 55 75 55 Timotmin) 10 Figure 11. 56 Rat ERG's recorded in vitro. A, B, C and D. Recorded during exposure to O at 3800 torr and 40°C. Note change in ampIitude calibra- tion. E and F. Control rat ERG's recorded in vitro under 0 at 740 torr and 40°C. See text for interpretation. 90 min 35 min 125 min 57 FIGURE 11 50 uV 200 msec 50 uV r—. 200 msec 2O uV 200 msec 10 uV 200 msec 20 uV K) 00 msec 2O uV M 00 msec 58 85-105 min of oxygen exposure. Note that the decline of the b-wave closely parallels the decline of the a-wave (Fig. 10). The late negative wave of the in vitro rat ERG makes it possible to study the effect of hyperbaric oxygen on the slow PIII. Initial enhancement of PIII in the control preparation is followed by a decrease in amplitude. The response is well above the initial amplitude at 2 hr (Fig. 10; Table 2). PIII declines continuously during exposure of the eye to hyperbaric oxygen (Figs. 10 and 11; Table 2). Means of differences were not plotted due to a lack of a significant amount of paired data. Effect of Hyperbaric Oxygen on the Goldfish ERG The goldfish retina continues to function normally during exposure to hyperbaric oxygen. No significant attenuation of the ERG occurs during the 6 hr experimen- tal period (Figs. 12 and 13; Table 3). Normal a- and b- waves were also maintained in the control preparation, however, the c-wave declined (Fig. 12; Table 3). Plots of the mean differences in amplitudes (control- experimental) versus time show that there was no signifi- cant change in these differences for the a- and b-waves during the experimental period (Fig. 13). ERG recordings are shown in Figure 14. These results are especially significant when com- pared to the changes in the frog and rat ERG's under hyperbaric oxygen which both showed attenuation. 59 CHE mm I CHE mOH mm H see: neon ooom om HH e C anon one on nn II T) .Hm>mH Rm pm mocmmemHU pcmonchHm * *m.e H m.nnI m.o H m.oH m.o H n.mo H.oH H n.me ooom HHHm o.o H o.mn- n.o H m.on o.HH H o.nm m.o H n.om one HHHm *o.mn H m.oeI on.o H on.o w.mH H e.me o.o H m.mm ooom meeaIe *n.m H m.o I m.H H n.m H.m H e.HH o.m H o.oH one meozIo *o.m H m.oHI m.n H m.H H.m H o.om o.n H o.mH ooom memaIm H.n H e.m I O.: H m.mm e.e H m.mm m.n H H.OH one m>m3rm Aeav o A>av cHe moH A>av one mm A>av ens ma .snoo oomm one upon one we cmwmxo ou mHszQHm mcHHSU Omm pap one CH momcwso .m mHomB Figure 12. 60 The goldfish ERG during exposure to 02 at 740 torr and 3800 torr at 22°C. Points represent means 1 SE for 9 control eyes and 15 experimen- tal eyes. Wave amplitudes are expressed as percent of initial amplitude. Lines were fitted by linear regression. A. a-wave 740 torr. Slope = 2.073, rxy = 0.100b, n = 88, initial amplitude 10.2 i 3.0 uV. B. a-wave 3800 torr. SlOpe = -l.12a, rxy = 0.041b, n = 137, initial amplitude 26.6 i 3.7 uV. Slopes of plots A and B are not significantly different from each other at 5% level. C. b-wave 740 torr. SlOpe = -3.90a, rxy = 0.100b, n = 88, initial amplitude A2.9 : 12.0 av. D. b-wave 3800 torr. SlOpe = 0.623, rxy = 0.0009, n = 137, initial amplitude 128.7 i 21.6 uV. SlOpes of plots 0 and D are not significantly different from each other at 5% level. E. c-wave 740 torr. Slope = -11.98c, rxy = 0.5049, n = 80, initial amplitude 150.3 : 17.8 uV. F. c-wave 3,800 torr. Slope = 5.573, rxy = 0.1u89, n = 110, initial amplitude 150.3 + 29.4 uV. a. Slope not significantly different from 0 at 5% level. b. Correlation not significant at 5% level. 0. Slope significantly different from 0 at 5% level. d. Correlation significant at 5% level. i50«[ iZOI % initial 00}, 0g A 02 740 Tan Time (hr) C .90, 02 740 Tan 3: I40i '7: woolE a I Tima (hr) E 02 T40 Torr I Tim. (in) 61 02 5000 Tan Tuna (in) D '59 02 3900 Tan iZOI 80- % initial 40} a 0L % Initial FIGURE 12 02 5500 Tan Time (hr) Figure 13. 62 Differences of amplitudes of goldfish ERG's during exposure to O at 740 torr and 3800 torr at 22°C. Differenceg in wave amplitudes (con- trol minus experimental) for 9 paired experi- ments were averaged and plotted against time (hours). Points represent mean : SE. Lines were fitted by linear regression. A. a-wagg. Slope = 0.u5a, rxy = 0.037b, n= . B. b-wagg. Slope = 4.58a, rxy = 0.170C, n= 0 c. c-wave. Slope -l5.98d, r = 0.uooc, n = 44. xy a. Slope not significantly different from 0 at 5% level. b. Correlation not significant at 5% level. 0. Correlation significant at 5% level. d. Slope significantly different from 0 at 5% level. X Dif. Amp. = Mean difference in amplitude (control-experimental). I? . i“ ti Hill '60 XDMAma Tima (hr) (Amp XDI I hfi O i Dif. Amp 5 r0 0 flOOI .140 V I a v v u . i 2 3 45 £5 6 Time (hr) FIGURE 13 .Ho>oH mm pm mocohommHO pCMOHthmHm * an H I an o I o ooomAmH u av onefim u so mm H not: o.em H n.mm+ H.om.H n.mom n.om H m.omH ooom oeoeIo mo.oa H o.mmI n.o H.o.mm o.en H_m.em one oeszo O.om_H m.nHI n.om H m.nHH m.Hm H e.omH ooom o>m3In o.o H o.mHI o.o H o.om o.mH H o.mn one oemsIo m.m H m.n I o.m H m.Hm e.m H m.om ooom o>m3nm H.m H_H.H + o.m H o.HH o.m H m.oH one oeozIm A>av o A>ao an o Asao an H .ocnooa amneonom NO maze on chamOQNo mo pooumm .m oHnt .ooooaomn no omm one do enoo ooom one neon oeeoa 65 Figure 14. Goldfish ERG's recorded in vitro during expo- sure to O at 3800 torr. Recordings were made over a 6 6r period at 22°C. The control response was comparable and followed the same course during the experiment. 66 1.5 hr 2.5 hr 6.5 hr FIGURE 14 50 uV r—. 500 msec 50 uV O 0 msec 5 50 uV L. 500 msec 5O uV I. 500 msec 67 Effect of Hyperbaric Oxygen on the Trout ERG ERG's were recorded from the trout retina throughout the 6 hr experimental period under oxygen at 740 torr and 3800 torr at 15°C. The ERG recorded under room air (Fig. 17E) resembled those recorded by Fonner (1973) during hypoxia. As shown in Figures 15 and 16 and Table 4, there was no significant change in the a—wave during exposure to hyperbaric oxygen and the response was not different from that of the control preparation. The b-wave of both preparations was significantly attenuated after 6 hr (Fig. 15; Table 4). These responses were not significantly different from each other (Figs. 15 and 16). The c—wave was a highly variable response and its amplitude fluctuated during the experimental period (Fig. 13). It was not always recorded from the control prepara- tion and a difference of means were not plotted due to a lack of significant number of paired experiments. The c-wave did not reach peak amplitude until 1 hr of dark adaptation and did not change significantly after that time (Fig. 15 and Table 4). Trout ERG's recorded during hyperbaric oxygen expo- sure are shown in Figure 17. An ERG recorded using the modified Ussing blocks is also shown. This may be a more satisfactory preparation for study of the in vitro trout ERG in view of the decline of the b-wave recorded from the eye cup. Figure 15. Changes in the trout ERG during exposure to O at 740 torr and 3800 torr at 15°C. Points 68 2 represent mean 1 SE for 10 experiments with wave amplitudes expressed as percent initial amplitude. Lines were fitted by linear regres- sion. A. a-wave 740 torr. SlOpe = -4.76a, rxy = 0.200”, n = 102, initial amplitude 33.8 i 5.6 uV. a-wave 3800 torr. SlOpe = -3.39c, rxy = 0.0859, n = 101, initial amplitude 27.3 i 3.9 uV. Slopes of plots A and B are not significantly different from each other at 5% level. b-wave 740 torr. Slope = -ll.12a, rxy = 0.539b, n = 102, initial amplitude 62.5 .i 7.5 uV. b-wave 3800 torr. Slope = —12.523, rxy = 0.412b, n = 101, initial amplitude 60.1 i_l2.7 pV. Slopes of plots 0 and D are not significantly different from each other at 5% level. c-wave 740 torr. Slope = -2.2oc, rxy = 0.1739, n = 42, initial amplitude 22.0 i 3.7 uV- c-wave 3800 torr. Slope = -0.4l°, rxy = 0.00, n = 75, initial amplitude 26.5 i_6.0 pV. Slopes of plots E and F are not significantly different from each other at 5% level. a. Slope significantly different from 0 at 5% level. b. Correlation significant at 5% level. c. Slope not significantly different from 0 at 5% level. d. Correlation not significant at 5% level. 69 A B 0 500 I40. 2 0Tan 02 7‘0 TO" 1 MO .- EIOOI a 2 .2 filoo o 5 ’ o g f f - , , a 40} I 2 3 4 5 5 °¢_I - - v - - —H nmrm ' 2 3 " 5 0 Timainri 1 .20 ‘ 0 g .0 ' I go I I i _ ‘ w ‘ t ‘01 ° 3 2 s I 5 3 ° ' ' ' ' I 2 ' a Timainri 7mg" 5 . Oz 740 TN! 2 I (D NMXDT , l 5 I60. 3 Inc. It i ‘ |20 i i "0 ‘ .0} i001 a o g - fi g 3 _ 0% - _ I z 3 4 5 a I 2 a 4 i i TIMIM) TRIM) FIGURE 15 Figure 16. 70 Differences of amplitudes of trout ERG's recorded during exposure to O at 740 torr and 3800 torr at 15°C. Differencgs in amplitude (control minus experimental) for 10 paired experiments were averaged and plotted against time. Points represent mean : SE. Lines were fitted by linear regression. A. a-wave. Slope = 0.02a, rx = 0.00, n = 101. y _ a _ b B. b-wave. Slope - 0.55 , rxy - 0.1“? , n = 101. a. Slope not significantly different from 0 at 5% level. b. Correlation not significant at 5% level. 7 Dif. Amp. = Mean difference in amplitude (control-experimental). 71 I m 2 3 4 5 6 Tlmo (hr) v m 2 3 4 5 6 Time (hr) FIGURE 16 72 .mo.o om mosmpmoofio undefioficmfim * on H I as o u o mm H.cmmz m H.0H H m.m + m.> H z.mm o.m H m.mm comm m>szo m o.o H o.m I m.H H o.oH o.m H o.mm ooh m>szo o *H.oH H o.o:- H.» H o.oH o.mfl H H.oo ooom m>szo o *o.oH H m.omI o.o H o.om m.o .H m.mo on» m>m3uo o o.» H m.o I o.m H m.mm o.m H.o.mm ooom o>m3Im o m.m H o.o I o.H H o.om o.m H m.sm one m>m3Im c A>no o A>oo an o A>oo no a .ppop oomm cam ago» ozo um cmwmxo op whamooxo moanso wmm psopu on» CH mowcmnu .: magma 73 Figure 17. Trout ERG's recorded in vitro. A, B, C, and D. Recorded during 6 hr exposure to O at 3800 torr at 15°C. Control responses wgre compara- ble to experimentals. E. Trout ERG recorded in vitro under room air at 15°C. F. Trout ERG recorded in vitro using modified Ussing blocks and Medium 199. Aerated with pure 0 at 15°C. The pigmented epi- thelium has been removed. 74 CI]: 4"" or 4" EV” "" FIGURE 17 20 uV 20 uV U1 0 O 20 uV r 500 20 uV 10 uV U1 0 O 20 uV r 500 msec msec msec msec msec msec DISCUSSION In this study the ERG was used as a monitor of reti- nal function in investigation of the effect of hyperbaric oxygen on the retinal cells. Attenuation of the ERG amplitude during exposure to hyperbaric oxygen, as compared to control, was considered to be indicative of oxygen toxicity. Maintenance of the ERG, i.e. non-attenuation of amplitude, or a response not different from control was seen as evidence of resistance to oxygen toxicity. The vasculature of the enucleated eye was not perfused, therefore any changes observed in the ERG were due to a direct effect of 02 on the retina and were not mediated by the cardiovascular, respiratory or nervous systems of the animal. The data was also interpreted in terms of the effect of 02 on individual retinal cell types based on our pre- sent understanding of the cellular origins of the ERG. Such interpretation should be made with caution in that the ERG is a composite of simultaneous activity in several different parts of the retina. This retinal activity sums algebraically and since the various potentials contribut- ing to the ERG are different in amplitude, polarity, and duration changes in one component may cause apparent changes in another component. These difficulties may be 75 76 overcome by surgical removal of the pigmented epithelium, the use of selective inhibitors, or the more tedious technique of intracellular recording. The ERG's recorded were typical of those recorded both in vivo and in vitro by other workers” There was no significant difference between the responses recorded from the control chamber and the hyperbaric chamber when both contained 02 at 7&0 torr (Appendix II). This indicates that differences in the ERG were not due to differences in the recording geometry, but to the differences in experi- mental conditions. The retina was not perfused (perfusion here defined as bathing the retina with a constant flow of oxygenated, nutrient saline) and the question may arise whether or not the eye cup is a good preparation for the type of study in which changes in the ERG over an extended period of time are observed. The simplicity of the preparation is a point in its favor and the fact that a satisfactory response could be recorded from the poikilotherm retinas for at least six hr under control conditions indicates that the eye cup was a satisfactory preparation. Perfusion with a nutrient saline probably would have maintained the rat retina for a longer period of time, since existing stores of nutrients might otherwise be rapidly depleted due to the high metabolic rate of the rat. Another problem which arises from the use of the eye cup is that of diffusion of oxygen through the vitreous 77 humor. Preliminary studies indicate that the rate of diffusion is very slow. This was not a problem in the goldfish, frog, and rat studies because almost all of the vitreous body could be removed. In one study the rat ERG was abolished within three min when 02 at 7ND torr was replaced with air. When the chamber was again purged with O the ERG was restored to original amplitude within five 2 min indicating that 02 was reaching the retina quite rapid— ly. The vitreous body of the trout eye is very difficult to remove and most of it remained in the eye cup during the recording session. The diffusion path for oxygen is relatively long and it is not known what the P02 is at the retina in the steady state under 0 at 3800 torr. 2 It has been noted that initial enhancement (except in the trout) of the ERG and low variability between pre- parations was seen under hyperbaric oxygen. The cells responsible for the ERG respond with graded potentials (Armington, 197”). Under control conditions in vitro, some of the retinal cells may be in a slightly anoxic state and respond with lower potentials due to reduced ATP levels. Under conditions of very high P all retinal cells may 02 respond maximally for a given stimulus intensity leading to increased amplitude and a lower variability. The Toxicity of Oxygen to the Frog and Rat Retinas The Frog ERG The attenuation of the frog ERG during exposure to hyperbaric oxygen (Fig. 9) indicates that high PO is toxic 2 78 to the retinal cells responsible for the generation of the ERG. Although the slopes of the lines fitted to the a-wave data are not significantly different from each other (Fig. 7), examination of the transient changes dur- ing the experiment shows that a-wave begins to decline after two hr under hyperbaric 02 while the control a-wave is maintained for four hr. The decline of the control a- wave is probably due to deterioration of the preparation due to depletion of nutrients since the eye cup is not perfused. Oxygen probably has its toXic effect on pro- cesses separate from those causing the decline of the control a—wave. The data for the b-wave and c-waves show a definite toxic effect of oxygen. These waves are a-wave dependent, that is, attenuation of the a-wave should be accompanied by attenuation of the b-wave and c-wave. In this study, the fact that the b- and c-waves decline much more than the a-wave indicates that oxygen must have a direct effect on the cells generating these components of the ERG. The Rat ERG The rat retina is highly susceptible to oxygen tox- icity. The abolishment of the a-wave (fast PIII) and the attenuation of the late negative wave (slow PIII) indicates that hyperbaric oxygen has a definitive effect on the PIII component of the rat ERG. This effect is much greater than that seen in the frog, a finding which is in keeping with high susceptibility to oxygen toxicity generally seen in 79 mammals. BaeyensiflLal. (1973; 1974) also found that the amphibian retina is less susceptible to oxygen toxicity than the mammalian retina. Mammalian metabolic pathways are more highly integrated than those of lower animals. The mammalian enzyme systems operate within narrower limits than those of poikilotherms and there is a greater degree of interaction between these systems so that a given al- teration in a pathway will have a greater effect on a mammal than on a lower animal. An example of this is that while cyanide is highly toxic to mammals it will not kill a trout because it can utilize alternative metabolic pathways (Hoffert, personal communication). The same principle appears to hold in this study of oxygen toxicity. The effect of hyperbaric oxygen on the rat b-wave is not well defined since the control b-wave is also atten— uated (Fig. 11). As stated earlier this decline of the control b-wave may be due to hypoxia (Weidner, 1976). The b-wave of the experimental preparation has a greater abso- lute decline than the control b-wave (Table 2) and since the decline parallels that of the a-wave the explanation of the b—wave attenuation in terms of oxygen toxicity is plausible. The lack of a c-wave in the rat ERG of this study is puzzling. Weidner (1976) recorded c-waves from Long-Evans rats in vivo, but not from albino rats. In vitro Long- Evans rat eye cup ERG's in this study resemble Weidner's recordings from albino rats in vivo. This problem was not 80 pursued experimentally but perhaps the large slow PIII com- ponent recorded in vitro negates the c-wave. Retinal Cells Affected by Oxygen as Indicated by ERG Changes In recent years much information has been gained about the origins of the component waves of the ERG. Using this knowledge it is possible to analyze the changes in the ERG during hyperbaric oxygen exposure in terms of the specific retinal cell types upon which oxygen has its toxic effect. As stated earlier this must be approached with caution since the ERG is a summation of the simultaneous activity of several cell types and changes in the activity of one cell group may cause apparent changes in the acti- vity of another group as indicated by the ERG. The a-wave (fast PIII) is the result of receptor cell activity (Dowling, 1970). Intracellular recordings from mud-puppy rods and cones show hyperpolarizations with latency and waveform of the leading edge which are very similar to the a-wave (Fain and Dowling, 1973). This sug- gests that the attenuation of the a-wave seen in this study is due to a toxic effect of oxygen on the photoreceptors. The slow component of PIII was also present in our record- ings from rat retinas (Fig. 11). Localization of this acti- vity has been difficult, however, Witkovsky, Dudek and Ripps (1975) have suggested that the slow PIII, as well as b-wave, originates in the Muller cells and that it is initiated directly by rod activity through changes in the 81 extracellular ionic environment. If this is so (it has not yet been confirmed by intracellular recording) the decline of the slow PIII of the rat ERG noted in the pre- sent study may indicate a toxic effect of 02 on the rods. Intraretinal recording shows that the b-wave origi— nates in the inner nuclear layer of the retina (Armington, 197M) and depolarizations corresponding to the b-wave can be recorded intracellularly from the Muller cells (Miller and Dowling, 1970). The Muller cells act as K+ electrodes and respond to increasing K+ in the extracellular fluid as the result of activity of neurons located postsynaptic to the receptors (primarily the bipolar cells) (Miller, 1973). Thus the decline of the b-wave under hyperbaric oxygen which was so evident in the frog data of this study may be due either to a toxic effect of oxygen on the neu- rons surrounding the Muller cells, or a direct effect on the Muller cells themselves inhibiting their response to K+ fluxes. As shown above, the slow PIII may also arise in the Mfiller cells. Since this component could often be recorded from both rat and frog retinas after abolishment of the b-wave the toxic effect of oxygen is probably on the postreceptor neurons rather than on the Muller cells them- selves. If slow PIII and the b—wave both arise in the Muller cells we have here a good example of the difficulty in interpretation of the ERG, especially since these poten- tials are opposite in polarity. 82 Extracellular recording has confirmed that the pig- mented epithelium is the source of the c-wave and that this response is rod dependent (Steinberg, Schmidt and Brown, 1970; Schmidt and Steinberg, 1971). Steinberg and Miller (1973) suggested that the c-wave is a diffusion potential and results from either decreases in K+ or increases in H003- around the apices of the epithelial cells. Oakley and Green (1975) have confirmed that the pigmented epithelial cells hyperpolarize in response to decreasing extracellular K+ as it is reabsorbed by the rods. Conductance of K+ into a neuron is a slow process (pre- sumably active) which accounts for the slow response of the c-wave. Attenuation of the c-wave under hyperbaric 02 could be the result of toxicity to the rods or to the pigmented epithelial cell itself. Since the attenuation of the frog c-wave in this study was greater than the attenuation of the a-wave a direct effect of 02 on the epithelial cells is suggested. Besides shielding the receptor cells from excess light, a major function of the pigmented epithelium is phagocytosis of spent pigment containing discs which are sloughed off from the receptors (Young, 1970). It is sug- gested that a toxic effect of 02 on the pigmented epithe- lium as indicated by attenuation of the c-wave might also adversely affect these important functions. 83 Mechanisms of Retinal Oxygen Toxicity It has been established that oxygen is toxic at the level. of cellular metabolism (Haugaard, 1968). Since the retina is neural tissue and often considered to be simply an extension of the brain itself it is possible to specu- late on the mechanisms of retinal oxygen toxicity based on what is known about effect of 02 on brain tissue and on cellular metabolism in general. Kaplan and Stein (1957) have demonstrated that there is a loss of K+ and a decrease in active uptake of glutamate by guinea pig brain cortex slices eXposed to hyperbaric oxygen. It has further been shown that K+ and Na+ gra- dients continually decrease in the direction of the Donnan equilibrium during exposure of guinea pig cortext to hyper- baric oxygen for 120 min (Joanny, Corriol and Brue, 1970) indicating a possible toxic effect of O on the metabolic 2 pump reSponsible for maintaining threse gradients. Main- tenance of transmembrane potentials in the retina also depends on a metabolic pump which can be inhibited by ouabain (Sillman, Ito and Tomita, 1969) indicating a depen- dence on Na-K dependent ATPase. Treatment of the retina with ouabain abolishes the ERG (Honda, 1972). Since the maintenance of transmembrane potentials in the retina depends on activity of Na-K ATPase and since a reduction in transmembrane potential is seen during oxygen toxicity to neural tissue it is possible that hyperbaric oxygen causes inhibition of the ATPase enzyme in the retina. It is, of 84 course, possible that a reduction in transmembrane poten- tial could result simply from changes in the permeability of the retinal cells due to ultrastructural changes in the cell membrane caused by hyperbaric oxygen. The retina is an obligate glucose user. The enzymes of glycolysis are highly susceptible to oxygen inhibition (Haugaard, 1968; Baeyens gt_a1., 1979) which would stop production of the ATP needed to provide energy for mainte— nance of membrane potentials. A reduction of ATP in neural tissue exposed to hyperbaric oxygen has been demon- strated (Joanny §t_§1., 1970). In summary, the attenuation of the ERG during hyper- baric oxygenation could be the result of a reduction in transmembrane potentials due to inhibition of the enzymes of the Na-K pump, inhibition of the enzymes responsible for supplying energy to the pump in the form of ATP, or structural changes in the cell membrane which alter ion permeability. The Resistance of the Teleost Retina to Oxygen Toxicity It has previously been shown that 02 at 1H7O torr has no toxic effect on teleost (trout) retinal metabolism and LDH activity (Baeyens §t_§i,, 1973; 197“). Fonner (1973) has shown that the teleost retina depends on the high PO 2 generated by the choroidal rete for normal function. In this study no toxic effect was seen on the gold- fish ERG during exposure of the retina to hyperbaric oxy- gen (Fig. 12) and the a-wave and c—wave of the trout ERG 85 were maintained at their initial amplitudes throughout the experiment indicating a resistance of the teleost retina to oxygen toxicity. This is especially apparent when compared to the data obtained in the frog and rat experi- ments. The decline of the trout b-wave (Fig. 15) under hyperbaric oxygen is probably not due to oxygen toxicity since it is not statistically different from the decline of the control b-wave. It might be argued that the O at 2 760 torr is also toxic to the retina, however, we do not feel that this is the case since in other studies of oxygen toxicity the severity of the toxic effect is seen to increase with increasing P The decline of the b-wave O O 2 may be due to deterioration of the preparation caused by problem of oxygen diffusion through the vitreous which was discussed above. Oxygen Supply to the Teleost Retina from the Choroidal Rete The teleost retina is avascular. This characteris- tic is probably an adaptation to the conditions of reduced light under water. The lack of retinal blood vessels allows more light to reach the receptor cells and allows greater visual acuity. In animals with a vascularized retina oxygen reaches the deep layers of the retina through the retinal arteries. In the teleost all oxygen must diffuse from the choroid. The counter-current multiplier of the choroidal rete generates 86 oxygen tensions in excess of “00 torr in order to supply adequate oxygen to the whole retina (Fairbanks g§_a1., 1969). Calculations by Dollery, Bulpitt and Kohner (1969) have shown that theoretically 97% of the human retina between the choroid and deep retinal arteries, a distance 2 when the P02 in the choroid is ”00 torr. At a PO2 of 1U3O torr in the choroid the whole retina (260 u thick) can be supplied with of about 100 u, can be supplied with 0 oxygen. Because Dollery made certain assumptions about ocular blood flow and oxygen consumption by the retina which are not applicable to the fish it is not possible to extrapolate his data directly to the trout. It seems clear, however, that trout should have no problem in supplying its whole retina (200 u thick, unpublished data of Hoffert) with oxygen at a choroidal P of N00 torr, O 2 especially when one considers that the metabolic rate of the teleost retina at 15°C is only about one-third that of the mammalian retina at 37°C (calculated from data of Baeyens et al., 1973). Dollery et al. (1969) have also shown that there must be a PO gradient in the retina due 2 2 consumption; PO decreasing with increasing distance 2 from the choroid. This means that whole teleost retina is to 0 not exposed to high PO , i.e. the normal level of Moo torr, 2 and raises the question of whether all retinal cells are resistant to oxygen toxicity. (Hoffert et al. (1973) have shown that teleost lens and cornea as well as retina are 87 resistant to oxygen toxicity). The presence of this gra- dient also means that the P probably did not reach 3800 02 torr at the pigmented epithelium in the hyperbaric oxygen experiments, however, at this high pressure it is reason- able to assume that PO2's significantly higher than normal . existed throughout the retina. Mechanisms of Resistance to Oxygen Toxicity There is much evidence that teleost retinal tissue is resistant to oxygen toxicity, although at this time the source of this resistance is unknown. Three mechanisms of protection against oxygen toxicity are proposed here which could serve as a basis for further investigation on this subject. A mechanism of protection against oxygen toxicity may be the enzyme superoxide dismutase (SOD). This enzyme has been extensively studied by I. Fridovich (1974). Dur- ing electron transport 02 may be partially reduced to superoxide radical (O -) by univalent reduction of 0 or 2 2 H202. Univalent pathways are actually favored over mechan- isms using electron pairs. Thus in the presence of ele- vated PO2 high levels of toxic 02' may be produced. The function of SOD is to eliminate these poisonous radicals by catalyzing the reaction 202- + 2H ———> H2O2 + 02. The H202, which is also toxic, is decomposed by the reaction H202 ——> H survive in 100% 02 (at 1 atm) if acclimatized in 85% O2 20 + 1/202 catalyzed by catalase. Rats can 88 for seven days. Survival is correlated with an increase in SOD in the lungs during the seven day period (Fridovich, 197“) which indicates that SOD can be induced in the pre- sence of high P02 and may then be produced in relatively large amounts at the teleost retina thereby protecting against oxygen toxicity. The teleost eye may contain high levels of gluta- thione. It has been shown that an important function of this compound is to maintain the SH groups of enzymes in their reduced state (Haugaard, 1968). The SH group of glutathione is easily oxidized and may be oxidized more readily than the SH groups of enzymes, or it may reduce SH groups of enzymes which have already been oxidized. Bae- yens (1975) has shown that addition of reduced glutathione to mouse brain homogenate prevents inhibition of LDH under hyperbaric oxygen. These proposed high levels of reduced glutathione may originate in the teleost retina or be produced in the pseudobranch, which itself is resis- tant to oxygen toxicity (Hoffert §§_§l,, 1973), and carried to the retina by the afferent pseudobranch artery. Finally, the enzymes of the metabolic pathways of the teleost retina may in themselves be resistant to oxi- dation. Possibly these isozymes are specially adapted to high PO2 through structural modifications such as the pre- sence of non-essential SH groups which are more easily oxidized than the SH groups of the active site of the enzyme. SUMMARY Through the use of isolated,runbperfused eyecups it has been shown that hyperbaric 02 has a direct effect on the retinal cells irrespective of changes mediated by other systems, e.g. the cardiovascular system. Exposure to hyperbaric oxygen (3800 torr) causes ini- tial enhancement of the goldfish, frog, and rat ERG's as compared to the responses from the control prepar- ations (7ND torr). Hyperbaric oxygen is toxic to the frog retina causing attenuation of all components of the ERG. The b-wave and c-wave decline more than the a-wave indicating a direct effect of oxygen on the neurons lying post- synatpic to the receptors and on the pigmented epithe- lium respectively. The toxicity probably does not involve the Muller cells as evidenced by longer per- sistence of the slow PIII. Hyperbaric oxygen is highly toxic to the rat retina. The attenuation of the a—wave and slow PIII components of the ERG indicates a direct toxic effect of oxygen on the photoreceptors. The decline of the b-wave under hyperbaric oxygen parallels that of the a-wave reflect- ing the decline in receptor activity. 89 90 The teleost ERG is not attenuated by exposure to hyperbaric oxygen. This indicates that the retina is well adapted to the high oxygen tensions generated by the choroidal rete and that no toxic effect is seen from exposure to oxygen tensions well above those nor- mally encountered. 1. RECOMMENDATIONS FOR FURTHER STUDIES A study similar to the one reported here using micro- electrodes, intracellular recording techniques, and perfused retinas in order to study the effect of hyperbaric oxygen on individual cells. The attenuation of the ERG during oxygen toxicity may be due to a reduction in transmembrane potentials. If this is the case the ERG could be restored by perfusing the retina with a high Na+ solution which initially would restore the transmembrane potential (Sillman e§_a1., 1969). A study of the effect of hyperbaric oxygen on Na-K dependent ATPase. Compare the amount of glutathione and superoxide dis— mutase in the teleost eye to the levels found in other eyes and tissues. 91 LITERATURE CITED LITERATURE CITED Anderson, 3., and H. A. Saltzman. 1965. Hyperbaria, Hybaroxia, and the Retinal and Cerebral Vessels. Headache, 5:73-77. Armington, J. C. 197A. The Electroretinogram. Academic Press, New York. h78pp. Ashton, N., B. Ward and G. Serpel. 1953. Role of Oxygen in the Genesis of Retrolental Fibroplasia. Brit. J. Ophthal., 37:513—519. Baeyens, D. A. 1972. Comparative Aspects of Oxygen Toxi- city. Ph.D. Thesis, Michigan State University, East Lansing, Michigan. . 1975. Oxygen Induced Inhibition of Mouse Brain Lactate Dehydrogenase. Aviat. Space Environ. Med., 46:772-77A. Baeyens, D. A., and J. R. Hoffert. 1972. Inhibition of Vertebrate Retinal Lactic Dehydrogenase During Expo- sure to Normoboric Oxygen. Aerospace Med. A3:263- 265. Baeyens, D. A., J. R. Hoffert and P. O. Fromm. 1973. A Comparative Study of Oxygen Toxicity in the Retina, Brain and Liver of the Teleost, Amphibian and Mammal. Comp. Biochem. Physiol., 45A: 925- 932. . 1974. A Comparative Study of the Influence of Oxygen on Lactate Dehydrogenase. Comp. Biochem. Physiol., 47B:l-7. Baylor, D. A. and M. G. F. Fourtes. 1970. Electrical Responses of Single Cones in the Retina of the Tur- tle. J. Physiol., 207:77-92. Bean, J. W., D. Zee and B. Thom. 1966. Pulmonary Changes with Convulsions Induced by Drugs and Oxygen at High Pressure. J. Appl. Physiol., 21:865-872. Beckman, D. L., and R. T. Houlihan. 1973. Hyperbaric Oxygen and Alveolar Surfactants. Aerospace Med., uu:u22—u2u. 92 93 Bergren, D. R., and D. L. Beckman. 1975. Hyperbaric Oxy- gen and Pumonary Surface Tension. Aviat. Space Environ. Med., M6:99fl-995. Bresnick, G. H. 1970. Oxygen Induced Visual Cell Degen- eration in the Rabbit. Invest. Ophthal., 9:372-387. Bridges, w. Z. 1966. Electroretinographic Changes During Hyperoxia. Arch. Ophthal., 75:812-817. Brown, K. T. 1968. The ERG: Its Components and Their Origins. Vis. Res., 8:633-677. D'Aoust, B. G. 1969. Hyperbaric Oxygen: Toxicity to Fish at Pressures Present in Their Swimbladders. Science, 163:576-578. Dollery, C. T., C. J. Bulpitt and E. M. Kohner. 1969. Oxygen Supply to the Retina frOm the Retinal and Choroidal Circulations at Normal and Increased Arter- ial Oxygen Tensions. Invest. Ophthal., 8:588-59h. Dollery, c. T., D. w. Hill, c. M. Mailer and P. s. Ramalho. 196“. High Oxygen Pressure and the Retinal Vessels. The Lancet, 2:291-292. Dowling, J. E. 1970. Organization of Vertebrate Retinas. Invest. Ophthal., 9:655-680. Eldred, G. E. 1975. An Atlas of the Choroidal Ultrastruc— ture with the Histochemical Localization of Choroidal and Pseudobranchial Carbonic Anhydrase Activity in the Rainbow Trout (Salmo gairdneri). M.S. Thesis, Michigan State University, East Lansing, Michigan. Eperon, G., M. Johnson and N. J. David. 1975. The Effect of Arterial PO on Relative Retinal Blood Flow in Monkeys. Invegt. Ophthal., iuz3u2-352. Fain, G. L., and J. E. Dowling. 1973. Intracellular Recordings from Single Rods and Cones in the Mud- puppy Retina. Science, 180:1178-1181. Fairbanks, M. B. 1970. The Importance of the Enzyme Carbonic Anhydrase in Ocular Oxygen Concentration in the Rainbow Trout (Salmo gairdneri). Ph.D. The- sis, Michigan State University, East Lansing, Mich. Fairbanks, M. B., J. R. Hoffert and P. O. Fromm. 1969. The Dependence of the Oxygen Concentrating Mechanism of the Teleost Eye (Salmo gairdneri) on the Enzyme Carbonic Anhydrase. J. Gen. Physiol., 5A:203-211. 9N . 197“. Short Circuiting of the Ocular Oxygen Concentrating Mechanism in the Teleost Salmo gairdneri Using Carbonic Anhydase Inhibitors. J. Gen. Physiol., 6uz263-273. Fonner, D. B. 1973. Electrophysiological Evidence for the Necessity of the Counter Current Oxygen Multi- plier Mechanism in Maintaining Teleost Function. Ph.D. Thesis, Michigan State University, East Lansing, Michigan. Fonner, D. B., J. R. Hoffert and P. O. Fromm. 1973. The Importance of the Counter Current Oxygen Multiplier in Maintaining Retinal Function in the Teleost. Comp. Biochem. Physiol., N6Az559-567. Fridovich, I. 197“. Superoxide Dismutases. Adv. Enzymol., Alz35-97- Goodman, L. 8., and A. Gilman. 1975. The Pharmacological Basis of Therapeutics. 15th ed., Macmillan Co., New York. l70h pp. Granit, R. 1933. The Components of the Retinal Action Potential in Mammals and their Relation to the Discharge in the Optic Nerve. J. Physiol., 77:207- 239. Grunbaum, B. w. and H. A. Leon. 1975. Changes in Hapto- globin and Other Plasma Proteins of Rats during Expo- sure to Oxygen at 760 Torr. Aviat. Space Environ. Med., ”6:136—lh0. Hardenbergh, E., R. G. Buckles, J. A. Miles, Jr., P. W. Schilling and J. B. Eckenhoff. 1973. Cardiovascu- lar Changes in Anesthetized Dogs at 3 and 5 Atmos- pheres Absolute Pressure. Aerospace Med., AA 1231— 1235. Haugaard, N. l9u6. Oxygen Poisoning. XI. The Relation Between Inactivation of Enzymes by Oxygen and Essential Sulfhydryl Groups. J. Biol. Chem., 16“: 265-270. . 1968. Cellular Mechanisms of Oxygen Toxicity. Physiol. Rev., "8:311-373. Hoffert, J. R., D. A. Baeyens and P. O. Fromm. 1973. The Resistance of Teleost Ocular Tissues to Oxygen Toxi- city. Invest. Ophtha1., 12:858-861 . 1975. In Vivo Oxygen Toxicity in Invertebrates and Vertebrates. Comp. Biochem. Physiol., 52Az351- 353. 95 Honda, Y. 1972. Quantitative Analysis of Some Effects of Ouabain Upon the Electrical Activity of Mammalian Retinas. Invest. Ophthal., 11:699-705. Horn, R. S., and N. Haugaard. 1966. Inhibition of Carbo- hydrate Metabolism by Oxygen and N-Ethylmaleimide in Rat Heart Homogenates. J. Biol. Chem., 241: 3078-3082. Joanny, P., J. Corriol and F. Brue. 1970. Hyperbaric Oxygen: Effects on Metabolism and Ionic Movement in Cerebral Cortex Slices. Science, 167:1508-1510. Kaplan, S. A., and S. N. Stein. 1957. Effects of Oxygen at High Pressure on the Transport of POtassium, Sodium and Glutamate in Guinea Pig Brain Cortex. Am. J. Physiol., 190:157-162. Krishnamartu, s., and M. L. Wadhawan. l97u. Central Ner- vous System Toxicity in Hyperbaric Oxygen: Case Reports. Aerospace Med., “5:782-78H. Lambertsen, C. J., R. H. Krough, D. Y. Cooper, G. L. Emmel, H. H. Loescheke and C. F. Schmidt. 1953. Oxygen Toxicity. Effects in Man of Oxygen Inhalation at l and 3.5 Atmospheres on Blood Gas Transport, Cerebral Circulation and Cerebral Metabolism. J. Appl. Physiol., 5:“7l-A85. Lucas, D. R., and O. A. Trowell. 1958. In vitro Culture of the Eye and the Retina of the Mouse and Rat. J. Embryol. and Exptl Morph., 6:178-182. MeiJne, N. G. 1970. Hyperbaric Oxygen and Its Clinical Value. Charles C. Thomas, Springfield, Ill., 261 pp. Miller, R. F. 1973. Role of K+ in Generation of the b- Wave of the Electroretinogram. J. Neurophysiol., 33:28-38. Miller, R. F., and J. E. Dowling. 1970. Intracellular Responses of the Mfiller (Glial) Cells of the Mud- puppy Retina: Their Relation to the b-Wave of the Electroretinogram. J. Neurophysiol., 33:323-341. Nichols, C. w., and C. J. Lambertsen. 1969. Effects of High Oxygen Pressures on the Eye. New England J. Med., 281:25-30. Nichols, C. N., M. Yanoff, D. A. Hall, C. J. Lambertsen. 1972. Histologic Alterations Produced in the Eye by Oxygen at High Pressures. Arch. Ophthal., 87:M17-A2l. 96 Noell, W. K. 1955. Visual Cell Effect of High Oxygen Pressure. Fed. Proc., 14:107-108. . 1962. Effect of High and Low Oxygen Tension on the Visual System. in Schaeffer, K. E. Ed. Environ- mental Effects on Consciousness. Macmillan, New York. pp. 3-18. Oakley, B., and D. G. Green. 1975. The Ionic Basis of the c-Wave of the Electroretinogram. Abstr. ARVO Spring Meeting. p. 5. Patz, A. 1965. The Effect of Oxygen on Immature Retinal Vessels. Invest. Ophthal., 4:988-999. Patz, A., L. E. Hoeck and E. De La Cruz. 1952. Studies on the Effect of High nygen Administration in Retro- lental Fibroplasia: Nursery Observations. Am. J. Ophthal., 35:12A8-1253. Penn, R. D., and W. A. Hagins. 1969. Signal Transmission along Retinal Rods and the Origin of the Electrore- tinographic a-Wave. Nature, 223:201-205. Radomski, M. W., and W. J. Watson. 1973. Effect of Lithium on Actue Oxygen Toxicity and Associated Changes in Brain Gamma Amino Butyric Acid. Aero- Space Med., uhz387-392. Robinson, F. R., R. L. Sopher, C. E. Witchett and V. L. Carter, Jr. 1969. Pathology of Normobaric Oxygen Toxicity in Primates. Aerospace Med., H0:879-88u. Saltzman, H. A. 1967. Hyperbaric Oxygen. Medical Clinics of North America, 51:1301-131u Schmidt, R., and R. H. Steinberg. 1971. Rod Dependent Intracellular Responses to Light Recorded from the Pigment Epthelium of the Cat Retina. J. Physiol., 217:71-91. Stadie, W. C., B. C. Riggs and N. Haugaard. 1945. Oxygen Poisoning. IV. The Effect of High Oxygen Pressures upon the Metabolism of Liver, Kidney, Lung and Mus- cle Tissue. J. Biol. Chem., 160:191-208. Sillman, A. J., H. Ito and T. Tomita. 1969. Studies on the Mass Receptor Potential of the Isolated Frog Retina. II. On the Basis of the Ionic Mechanism. Vis. Res., 9:1AA3-1H51. Steinberg, R. H., and S. Miller. 1973. Aspects of Elec- trolyte Transport in Frog Pigment Epithelium. Exp. Eye Res., 16:36N-372. 97 Steinberg, R. H., R. Schmidt and K. T. Brown. 1970. Intracellular Response to Light from Cat Pigment Epithelium: Origin of the Electroretinogram c-Wave. Nature, 277:728-730. Tomita, T. 1976. ElectrOphysiological Studies of Retinal Cell Function. Invest. Ophthal., 15:171-187. von Schnakenberg, K. and H. Nolte. 1970. Histological Studies of Alterations in the Rat Brain under Oxygen at High Pressure. Aerospace Med., Ul:lOl3-10l7. Weidner, C. 1976. The c-Wave in the ERG of Albino Rat. Vis. Res., 16:753-763. Widdell, P. J., P. B. Bennett, P. Kivlin and W. Gray. 197“. Pulmonary Oxygen Toxicity in Man at 2 Atm with intermittent Air Breathing. Aerospace Med., #5: 07-410. Witkovsky, P., F. E. Dudek and H. Ripps. 1975. Slow PIII Component of the Carp Electroretinogram. J. Gen. Physiol., 65:119-13“. Wittenberg, J. B., and B. S. Wittenberg. 1962. Active Secretion of Oxygen into the Eye of the Fish. Nature, 19A:lO6-107. Wood, C. D., G. F. Perkins, A. G. Smith and J. M. Reaux. 1972. Response of the Cardiovascular System in Oxygen Toxicity. Aerospace Med., “3:162-167. Wood, J. D., W. J. Watson and G. W. Murray. 1969. Corre- lation Between Decreases in Brain Gamma-Amino-Butyric Acid and Susceptibility to Convulsions Induced by Hyperbaric Oxygen. J. Neurochem., 16:281-287. Yanoff, M., W. W. Miller and J. A. Waldhausen. 1970. Oxy- gen Poisoning of the Eyes. Comparison of Cyanotic and Acyanotic Dogs. Arch. Ophthal., 8hz627-629. Young, R. W. 1970. Visual Cells. Sci. Am., 223:81. APPENDICES Composition of Cold Blooded Ringers NaCl CaCl2 KC1 Dist. H20 APPENDIX I 6.40 g 0.20 g 0.10 g to 1 liter. 98 111.0 mM 1.8 mM 1.3 mM APPENDIX II CHAMBER COMPARISON The control and hyperbaric chambers (not to be con- fused with the recording chamber which held the eye cup itself) used in this study were quite different from each other in construction. The controlchamber was a clear Lucite box, whiletnuehyperbaric chamber was a steel cylin- der painted white on the inside with a glass window. These differences could lead to differences in lighting conditions so it was necessary to compare the ERG's recorded from these chambers under similar atmospheric conditions. This comparison was made by recording ERG's from trout eye cups with both chambers containing 02 at 7H0 torr. ERG's were recorded every five min for 70 min following 30 min of dark adaptation. The average wave amplitudes for seven experiments are shown in Table 5. Table 5. Comparison of trout ERG amplitudes recorded at 15°C from the hyperbaric and control chambers under 02 at 7A0 torr. a-wave (uV) b-wave (uV) c-wave (uV) Hyperbaric chamber 1A.6:2.7(7) “2.7:5.6(7) 23.916.3(6) Control chamber 18.3:3.N(7) 52.619.5(7) 27.5ill.6(6) Mean : SE(N) Differences are not significant at 5% level. 99 100 These results indicate that there is no significant difference in the ERG amplitudes recorded from the two chambers under similar conditions. The ERG amplitude from the hyperbaric chamber is slightly lower than the response from the control chamber which lends support to the obser- vation that the ERG is initially enhanced by hyperbaric oxygen. "I11111111111111“