Less than one hundred years ago Paul Bert discovered that oxygen at raised pressures was injurious to a wide range of organisms and, in a number of experiments conducted over several years, he investigated this and other problems which form the basis for what has subsequently become high-pressure physiology. It seems however that the large amount of research which has been done in an attempt to understand the nature of oxygen poisoning has touted to confuse the original simple view of oxygen as a universal poison. A very large number of reactions which depend upon enzymes are adversely affected by raised pressures of oxygen, sufficient indeed to make it likely that the 'cause' of oxygen poisoning will not be found, in the sense that no single critical reaction will be identified as giving rise to any one of the various toxic manifestations. Instead the emphasis must be upon studying the way in which cellular metabolism is modified and finally disrupted by raised pressures of oxygen; and in particular by trying to identity the rate-limiting step within the complex of reactions.</p> The present work has been upon one aspect of oxygen toxicity, namely the development of convulsions in mice which have been exposed to raised pressures of oxygen. These experiments have been based upon two hypotheses; the first of these was that the dose-response relationship was related to the dynamics of the underlying biochemical processes and particularly, that the rate at which toxic signs developed would be determined by whichever reaction was rate-limiting. Attempts to correlate biochemical and physiological information have not so far been possible since there hare been no estimates available of the rate at which toxicity develops. The first task was therefore to define as accurately as possible the dose-response relationship between the ambient oxygen pressure and the time taken for convulsions to appear, and then, by dividing each exposure into two parts, to attempt to measure the rate at which toxicity develops. The second hypothesis was that the time to convulsions was not uniquely determined by the partial pressure of oxygen inspired and hence that the convulsion time would be modified by the presence of other respirable gases. Since it was thought that the nature of such modifications would give further information about the process which leads to convulsions, the interactions between oxygen and other gases, name, Carbon Dioxide, Nitrous Oxide, Nitrogen and Helium therefore form the second part of this work. The first notable feature of oxygen convulsions in mice was that at least two types of convulsion were distinguishable. The first type, referred to as minor convulsions, tended to occur early, was of short duration and was minor in intensity, while the second tended to occur later, was more prolonged and more violent and was called a major convulsion. A few animals showed chronic rolling movements after relatively long exposures. The second point of importance was that, for any particular oxygen pressure, convulsions in mice were found to be distributed in a skew manner. This could however be converted to a normal distribution by a logarithmic transformation, which also had the desirable feature that the standard deviation of the transformed results was independent of the mean, unlike the untransformed data. Consideration of the dose-response curve led to an approximate relationship in which the product of the ambient oxygen pressure and the mean logarithmic convulsion time for that pressure was found to be constant. A similar relationship was also found to describe the occurrence of minor convulsions and the relationship between minor and major convulsions expressed as the ratio between the slopes of the respective dose-response lines. When an initial exposure, short of convulsions, was immediately followed by a second exposure leading to convulsions, such divided exposures were found to be additive; that is, the parts were capable of being described by the same mathematical relationship as the whole. These divided exposures also led to the definition of a provoked threshold which for mice was at the pressure of about 3 atm. of oxygen. Exposure to pressures above this threshold shortened subsequent exposures, while exposures below the threshold had not detectable effect upon subsequent exposures. Other experiments using divided exposures, in which the time of the first exposure was varied, were interpreted as showing that the process leading to convulsions was, as a first approximation, exponential in form and that its rate was proportional to the ambient oxygen pressure. Estimates for the half-time of this process at 3 atm. and 4 atm. were 35 mins. and 22 mins. respectively. The first interaction to be investigated was that between oxygen and carbon dioxide and the results obtained were found to be in agreement with the findings of Marshall and Lambertsen (1961), since carbon dioxide at low levels hastened the onset of convulsions, although at high levels no significant effect could be demonstrated. This would seem to be consistent with the view that high carbon dioxide levels protect against convulsions and would point to a cross-over point at about 7 atm. of oxygen for carbon dioxide levels of 0.03-0.07 atm. Similar results were obtained for nitrogen, these tow gases, carbon dioxide and nitrogen both changed the slope and the intercept of the dose-response line as is shown diagrammatically in Figure 17. While the mechanism of action is not known it is speculated that both may lead to changes in intracellular pH. at lower concentrations, while at higher concentrations the anaesthetic properties of these gases may preponderate. The evidence obtained from the interaction with nitrous oxide seemed to show that more than one type of interaction was possible, since this gas altered the slope of the dose-response line. 0.54 atm. of N<sub>2</sub>O gave a marked protection against convulsions, but did not appreciably alter the intercept. Helium was found to produce a marked acceleration in the onset of oxygen convulsions, bat no evidence was found for any convulsant action of helium itself, since supra-liminal oxygen pressures were necessary in order to show the effects of helium. The magnitude of the effect, judged by the change in slope of the dose-response time, was proportional to the helium partial pressure; hence the possibility has been considered that this effect was due not to any physiological action of helium, but rather to the effects of pressure 'per se'. The significance of the threshold, i.e. the minimum effective pressure to produce convulsions and the minimum response time, has been discussed. Two possible mechanisms to account for the threshold were considered. The first of these depends upon the fact that the tissue oxygen consumption tends to keep the tissue oxygen tension within normal limits until the total quantity of oxygen supplied exceeds the quantity being consumed. The second possibility which was considered was that of a competing recovery process which removed some product of the convulsion-producing reaction at a rate different from that at which it was produced. While on balance the first explanation was thought more likely, recovery undoubtedly does occur, and both may therefore be involved. Regarding the minimum response time, this was thought to be a simple consequence of the rate of the process to convulsions reaching a value; although the reason for this is not obvious, it was suggested that it might be due to the progressive inactivation of enzymes by oxygen at pressure. Consideration of the possible sites of action of oxygen in producing convulsions may perhaps be premature, since it is uncertain whether the convulsions represent a generalised reaction of brain tissue or whether there are particularly sensitive areas on the basal ganglia from which the convulsions originate. It was thought however that there was sufficient evidence to show that oxygen acts by disturbing cellular oxidations to make the choice of the mitochondrion, as the site of action, not unlikely. Using the previous assumption and available data, an estimate was made of the toxic level of oxygen in the brain of man as being of the order of 40-85 min. Hg. at the mitochondria, or 70-100 min. Hg. in the venous blood. It was shown that the dose-response curve might be interpreted as resulting from a process leading to convulsions. It was further suggested that this hypothetical process to convulsions represented the rate-limiting step of a cyclic or chain reaction involved in tissue oxidations and that the kinetics of this process might help to identify the site or sites at which oxygen acts to produce convulsions. The existence of interactions between oxygen and other gases has been confirmed and it has been suggested that these interactions may represent interference with tissue oxidations either at the same sites as those involved in oxygen toxicity or at adjacent sites. The evidence suggested that there were two principal modes of interaction, the first represented by the anaesthetic gases and the second by the action of helium, although alterations in intra-cellular pH. may also be of some importance.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:644618 |
| Date | January 1969 |
| Creators | Barnard, E. E. P. |
| Publisher | University of Oxford |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://ora.ox.ac.uk/objects/uuid:28d3cf00-12be-440e-b595-5181fc42f693 |
Page generated in 0.0029 seconds