- Email: [email protected]
Acute ethanol effects on blood pH, PCO2, and PO2 in LS and SS mice
Physiology & Behavior, Vol. 28, pp. 879-883. Pergamon Press and Brain Research Publ., 1982. Printed in the U.S.A.
Acute Ethanol Effects on Blood pH, PCO2, and PO2 in LS and SS Mice' D A V I D M. G I L L I A M A N D A L L A N
C. C O L L I N S z
Alcohol Research Center, Department of Psychology, School of Pharmacy and Institute for Behavioral Genetics, University of Colorado Boulder, CO 80309 R e c e i v e d 15 J u l y 1981 GILLIAM, D. M. AND A. C. COLLINS. Acute ethanol effects on blood pH, PC02, and P02 in LS and SS mice.
PHYSIOL. BEHAV. 28(5) 879--883, 1982.--The effects of ethanol on blood pH, PCO2, and POx were measured in LS and SS mice in an attempt to ascertain whether these lines of mice, which differ in CNS sensitivity to the behavioral effects of ethanol, also differ in sensitivity to physiological effects of this drug. Long-sleep (LS) female mice were injected intraperitoneally with 1.8, 2.5, 3.3, or 3.8 g/kg ethanol; short-sleep (SS) female mice were administered 2.5, 3.3, 4.1, or 4.7 g/kg. Blood pH, PCO2, and POz were assessed at 15, 30, 60, 120, or 180 min after injection of the 2.5 and 4.1 g/kg doses or at 60 min after injection of the 1.8, 2.5, 3.3, 3.8, 4.1, and 4.7 g/kg doses. Opposite effects on blood pH and PCO2 over time were obtained in LS and SS mice at the 2.5 g/kg dose. Acidosis characterized the LS line, whereas alkalosis characterized the SS. The results obtained with SS mice at the 4.1 g/kg dose were similar to those obtained with LS mice at the 2.5 g/kg dose. The dose-response curve for the SS mice generated at 60 min post-injection lies to the right of that for the LS mice. The effects of high ethanol doses on SS mice resemble the effects of low doses on LS animals. Thus, the two lines of mice differ in response to the effects of ethanol on these parameters related to respiration. The difference in sensitivity to the respiratory depressant effects of ethanol may contribute to the differences in behavioral sensitivity between the two lines. Ethanol
Blood pH, PCO2, and POz
Selected mice
T H E central nervous system (CNS) depressant actions of hypnotic agents are well known. Among CNS areas known to be susceptible to the depressant actions of hypnotic agents are respiratory centers o f the medulla. Several hypnotic agents have been shown to produce an increase in ventilation at low doses and a decrease in ventilation at high doses. Low doses o f ethanol, for example, produce an increase in rate of respiration [5, 14, 17], while a decrease in rate is seen at higher doses [8, 14, 17, 18, 22]. This change in rate may be a consequence of the effects of ethanol on centrally mediated responses to COz [8]. Biphasic responses are characteristic of ethanol, which is also known to produce biphasic behavioral responses [12,19]. Differences in response to ethanol administration have been well studied in the long-sleep (LS) and short-sleep (SS) lines that have been selectively bred for differential sensitivity to a hypnotic dose o f ethanol [13]. Research using the LS and SS mice has recently been reviewed [3]. Greater sensitivity to ethanol in LS mice has been demonstrated by a lower EDs0 for loss of the righting reflex [6], longer sleep time [6,21] (Gilliam and Collins, to be published), greater decrease in body temperature [9], and decrease in open-field activity [19] (Ritz, George and Collins, to be published) at lower doses of ethanol as compared to SS mice.
In view of these differences in behavioral sensitivity to ethanol, it seemed reasonable to suspect that ethanol may cause decreased ventilation in LS mice at lower doses of ethanol than in SS mice. Decreased ventilation should result in increased blood PCO2 and decreased blood pH, whereas increased ventilation should result in decreased PCO2 and increased pH. Blood PO2 is probably influenced by both respiration rate and tissue utilization. The current study was initiated in order to determine whether LS and SS mice do, indeed, differ in the effects of ethanol on respiration. Respiration rate is quite difficult to monitor in the mouse; fortunately, however, recent advances in the measurement of blood pH, PCO2, and PO2 have made it quite easy to measure these related parameters. The data presented in this report demonstrate that LS and SS mice differ in the effects of ethanol on respiration. This effect may contribute to the differences between the lines in their sensitivity to the behavioral effects of ethanol. METHOD
Subjects Female LS and SS mice from the 21st generation of the selective breeding program were used in this study. These
1This research was supported in part by a Research Scientist Development Award to A. C. Collins (AA-00029) and by NIAAA research grant AA-03527. It was also partially supported by BRSG grant RR-07013-14 awarded by the Biomedical Research Support Grant Program, Division of Research Resources, National Institutes of Health. ~Send reprint requests to Dr. Allan C. Collins, Institute for Behavioral Genetics, Box 447, University of Colorado, Boulder, CO 80309.
C o p y r i g h t © 1982 B r a i n R e s e a r c h P u b l i c a t i o n s Inc.--0031-9384/82/080879-05503.00/0
880
GILLIAM AND COLLINS
animals were tested at 70_+10 days of age between 1300 and 1600 hr. The normal 12-hr light cycle (lights on from 700 to 1900 hr) was maintained throughout the experiment. Ambient temperature was 22___1.5°C. Littermates were distributed randomly among all time points and doses in the following experiments.
•
~
-
-
'~, T
7.30C
~
•......
, . . . . .
,
<~
'
TLaA'---~-----(~---'-~
,=
........
SS 2 , S g / k # ss 4.1 gzkq
7.25(
7,2tX
Time Course
LS mice were injected intraperitoneally (IP) with 2.5 g/kg ethanol; SS mice received an IP ethanol dose of either 2.5 or 4.1 g/kg. Injection volumes were 0.01 ml/g body weight (BW) for the 2.5 g/kg dose and 0.013 ml/g BW for the 4.1 g/kg dose. Different groups of ethanol-treated mice were then tested for blood pH, PCO2, and PO2 at 15, 30, 60, 120, or 180 rain post-injection. Uninjected and saline-injected controls were also tested to establish control levels of pH, PCO2, and PO2. Blood samples were obtained from the retro-orbital sinus using 120-/zl heparinized capillary tubes. Blood pH, PCO2, and PO2 were determined by inserting the blood sample into a BMS3 Mk2 blood micro system (Radiometer Copenhagen) equipped with a PHM73 pH/blood gas monitor (Radiometer Copenhagen).
7°150 i i
CONT~
1~
30
60 120 MINIJTE S P O S T -IN,,ECTION
180
FIG. 1. Mean blood pH values for LS mice given 2.5 g/kg (A) and for SS mice given 2.5 g/kg (O) or 4.1 g/kg (12]) ethanol IP. Different groups of mice were assessed at 15, 30, 60, 120, or 180 min postinjection. N=6 to 9 per group. Solid triangle and solid hexagon represent control values for LS (N= 15) and SS iN= 15) mice, respectively. Brackets indicate -+1 SE.
Dose Response
|
LS mice were injected IP with 1.8, 2.5, 3.3, or 3.8 g/kg ethanol; the ethanol dose received by SS mice was either 2.5, 3.3, 4.1, or 4.7 g/kg. Injection volumes were 0.01 mug BW for the 1.8, 2.5, and 3.3 g/kg doses; 0.016 ml/g BW for the 3.8 g/kg dose; 0.013 mug BW for the 4.1 g/kg dose; and 0.02 ml/g BW for the 4.7 g/kg dose. Increased volumes for the higher doses were used to insure survival of the subjects for the duration of the experiments. Different groups of ethanoltreated mice were tested for blood pH, PCO2, and PO2 at 60 min after injection. Blood samples were obtained and analyzed in the manner described above, and uninjected and saline-injected controls were tested to establish control levels of pH, PCO~, and POv Data on animals injected with the 2.5 or 4.1 g/kg doses and tested at 60 rain were combined within line in the time-course and dose-response analyses, as were data on uninjected and saline-injected controls.
L
I
A
,
[] ss4.~g,kg
D-
3Q
Analyses
The significance of variations in these parameters was assessed by multivariate analysis of variance with the Statistical Package for the Social Sciences [ 15]. The Wilks test was used to determine statistical significance. One-way analyses of variance with post hoc tests (Scheff~ [20]: p =0.05) were performed on scores generated by the first standardized discriminant function coefficients from the interaction term (for data resulting from the time-course experiment) or from the main effect (for data resulting from the dose-response experiment). Trend analysis was performed following the methods described by Kerlinger and Pedhazur [10]. RESULTS Saline injection did not have an effect on blood pH, PCO~, or POz at any time. Data collected from saline-injected or uninjected mice were combined and represent control levels. Significant differences between lines, F(3,26)=3.3, p<0.05, were found for control levels of pH, PCO~, and PO~ taken collectively. Control levels of blood pH were higher in LS than in SS mice, while control levels of PCO2 and POz did not differ significantly between lines.
MINUTES
POST o ~NJEC TION
FIG. 2. Mean blood PCO2 values for LS mice given 2.5 g/kg (A) and for SS mice given 2.5 g/kg (O) or 4.1 g/kg ([]) ethanol ]P. Different groups of mice were assessed at 15, 30, 60, 120, or 180 rain postinjection. N=6 to 9 per group. Solid triangle and solid hexagon represent control values for LS (N= 15) and SS (N= 15) mice, respectively. Brackets indicate _ 1 SE.
Time Course
Figures I, 2, and 3 show mean values of pH, PCO2, and POz, respectively, for control animals and for experimental animals at all time points. When composite blood pH, PCO2, and PO2 measures for LS and SS mice injected with 2.5 g/kg ethanol were compared at 15, 30, 60, 120, or 180 min postinjection, results indicated a significant interaction between line and time, F(15,218)=4.5, p<0.001. Results of the multivariate analysis revealed that SS composite values did not differ from control at any time point. However, LS compo-
ACUTE E T H A N O L EFFECTS ON BLOOD
T 50
/k LS 2,5 g/kg 0 SS 2,5 9tkg
\
,\
881
[] ss 4.1
/x
/k LS 7,35Q
~- ~ 4
Q~ SS
\
g~kg 7,30C
~ ~ C)
7,2O0
7,15Q
3o
i
1,8
i
i
2,5 3,,3 ETHANOL DOSE (g/k~
i
i
3,8 4,1
i
4,7
FIG. 4. Mean blood pH values at 60 min post-injection for LS mice (A) given 1.8, 2.5, 3.3, or 3.8 g/kg or for SS mice ( O ) given 2.5, 3.3, 4.1, or 4.7 g/kg ethanol IP. N=6 to 9 per group. Solid symbols indicate control values. Brackets indicate ___1 SE.
~ co.;~ ,'~ ~'o
i
CONTROL
go
Go
40
MINUTES POST-INJECTION
FIG. 3. Mean blood PO~ values for LS mice given 2.5 g/kg (A) and for SS mice given 2.5 g/kg (©) or 4.1 g/kg ([]) ethanol IP. Different groups of mice were assessed at 15, 30, 60, 120, or 180 rain postinjection. N=6 to 9 per group. Solid triangle and solid hexagon represent control values for LS (N=15) and SS (N=15) mice, respectively. Brackets indicate __.1 SE.
/1 LS
5o
0 SS
45
site values were found to differ from control at 15, 30, and 60 rain post-injection. For LS mice, pH decreased from control levels at 15 rain and remained depressed through 60 min (Fig. 1), PCO~ was markedly elevated at 60 rain (Fig. 2); and PO 2 was greatly elevated at 15, 30, and 60 rain, with a return to near normal levels at 120 and 180 rain post-injection (Fig. 3). An unexpected finding was that the composite blood measures in the SS mice injected with the 4.1 g/kg dose and LS mice injected with 2.5 g/kS showed similar changes over time, F(15,213)=0.1, p >0.05. Trend analysis revealed significant LS-SS differences in the equations that describe the effects of the 2.5 g/ks dose on blood pH, PCO2, and PO2 over time. However, when LS mice injected with 2.5 g/ks were compared with SS mice injected with 4.1 g/kg, similar equations were found to describe the time course of the PO~ measures. Trends for pH and PCO2 differed between lines. The composite blood measures for SS mice injected with 2.5 or 4.1 g/kg ethanol were compared at 15, 30, 60, 120, and I80 min post-injection, The results of this analysis indicated a significant interaction between time and dose, F(15,174)=3.5, p<0.001. Multivariate analysis revealed that none of the values for SS mice injected with the 2.5 g/ks dose differed from control, but that the measures for SS mice given the 4.1 g/kg dose differed from the control value at 15, 30, 60, and I20 rain. For SS mice injected with 4.1 g/kg, pH decreased from the control value at 15 min post-injection, remained depressed through 120 min, and returned to the control value at 180 rain (Fig. 1); PCO2 was elevated at 60 and 120 min post-injection (Fig. 2); and PO~ increased from the control value at 15 rain and remained elevated through 180 rain (Fig. 3).
u 4o 35
30
ETHANOL DOSE (g/kg)
FIG. 5. Mean blood PC02 values at 60 rain post-injection for LS
mice (A) given 1.8, 2.5, 3.3, or 3.8 g/kg or for SS mice ( O ) given 2.5, 3.3, 4.1, or 4.7 g/kg ethanol IP. N--6 to 9 per group. Solid symbols indicate control values. Brackets indicate _+1 SE.
Dose Response Figures 4, 5, and 6 show mean values of pH, PCO~, and PO~, respectively, for control animals and for experimental animals at each dose at 60 rain post-injection. For both the LS and SS lines, composite blood measures varied as a function of dose, F(12,85)= 18.6, p <0.001; F(12,96)= 10.2, p<0.001, respectively. Values for LS mice given a dose of 2.5, 3.3, or 3.8 g/kg and for SS mice that received 3.3, 4.1, or 4.7 g/kg differed significantly from control. Both LS and SS mice show a dose-dependent decrease in pH and a dosedependent increase in PO~. There is an elevation of PCO~ for LS mice at the 2.5 g/kg dose and for SS mice at the 4.1 g/kg dose. Trend analysis revealed that the equations that describe the dose-response curves for pH, PCO~, or PO~ do not
~82
G I L L I A M AND C O I , I J N S
z~ LS 50
0
SS
F
5 35
30
CONTROL
1'.8
2'.5 3.3 ETHANOL DOSE (g/kg}
3.8 4.1
45
FIG. 6. Mean blood PO2 values at 60 min post-injection for LS mice (A) given 1.8, 2.5, 3.3, or 3.8 g/kg or for SS mice ( O ) given 2.5, 3.3, 4.1, or 4.7 g/kg ethanol IP. N=6 to 9 per group. Solid symbols indicate control values. Brackets indicate _+1 SE.
differ significantly between the LS and SS lines. The doseresponse curve for the SS mice thus represents a rightward shift in the dose-response curve for the LS mice. DISCUSSION
This study demonstrates that the effects of ethanol administration on blood pH, PCO2, and PO2 occur at lower doses for LS mice than for SS mice. LS mice given a dose of 2.5 g/kg show a decrease in blood pH, a general increase in PCO2, and a greatly increased PO2 over time. The same dose in SS animals produces very different effects: an increase in blood pH, a substantial decrease in PCO2, and only a slight increase in PO2 over time. The response of SS mice to a much higher dose of ethanol (4.1 g/kg) is similar to the response seen in LS mice at the 2.5 g/kg dose. Blood pH is decreased, and there is an increase in both PCO2 and PO2. The differences in sensitivity of LS and SS mice to ethanol's effects on blood pH, PCO2, and PO2 are particularly well demonstrated in their response to increasing doses of ethanol at 60 min post-injection. Blood pH in LS mice shows an increase at the 1.8 g/kg dose and a progressive decrease at higher doses. SS mice exhibit a similar pattern, but their responses lie to the right of LS mice on the doseresponse curve. For example, SS mice require higher doses of ethanol than LS mice in order to exhibit decreases in pH. The differential sensitivities of LS and SS mice with respect to changes in blood pH at increasing doses of ethanol are mirrored by changes in PCO2 and PO2. Interestingly, a change in the injection volume seems to influence the effect of ethanol on pH, PCOz, and PO2. LS mice given 3.3 g/kg ethanol in an injection volume of 0.01 ml/g BW showed a greater decrease in pH than did LS mice given 3.8 g/kg ethanol in an injection volume of 0.016 ml/g BW. A similar phenomenon was also seen for SS mice. SS mice given 4.1 g/kg ethanol in an injection volume of 0.013 ml/g BW showed a greater decrease in blood pH than SS mice given 4.7 g/kg ethanol in an injection volume of 0.02 ml/g BW. The cause of this reversal in effect of ethanol on blood pH is not clear. However, a concentration-dependent effect of ethanol on
blood pH, PCO2, and PO2, as well as on respiration rate, in the LS and SS mice has been verified in our laboratory (Gilliam and Collins, to be published). As noted previously, ethanol is known to produce respiratory depression at high doses [8, t4, 17, 18, 22], while low doses may increase respiration rate [5, 14, 17]. Hyperventilation leads to respiratory alkalosis (increased blood pH) and decreased PCOz. These changes characterize the responses of LS mice to the 1.8 g/kg dose and SS mice to the 2.5 g/kg dose at 60 rain post-injection. The results seen with the LS mice at 2.5 g/kg and the SS mice at 4.1 g/kg may be a consequence of ethanol's depressant action on central respiratory centers. Therefore, the medullary centers responsible for respiration may be differentially sensitive to ethanol's effects in the LS and SS mice. This notion receives support from the observation that LS cerebellar Purkinje cells are one to two orders of magnitude more sensitive to the depressant effects of locally administered ethanol than are those from the SS mice [23]. Ethanol has been shown to decrease the centrally mediated ventilatory response to CO ~ in a dose-dependent manner [8]. Respiratory depression and/or decreased responsiveness to CO2 result in respiratory acidosis, which would cause a decrease in blood pH. Metabolic acidosis resulting from the metabolism of ethanol to acetic acid or a rise in blood lactate during ethanol oxidation may contribute to the decrease in blood pH following the administration of ethanol in doses equal to or greater than 2.5 g/kg for LS mice or 3.3 g/kg for SS mice. A somewhat surprising finding was the increase in PO~ seen following ethanol administration. We had expected to find reduction in PO.,, especially at the higher ethanol doses, as a result of the respiratory depression. The observed increase in PO~ may be explained by the fact that decreases in blood pH are known to lower the affinity of the hemoglobin molecule for oxygen [2], the Bohr effect. Thus, the presence of more oxygen in the dissolved form may contribute to the initial increase in POz seen in LS and SS mice at higher ethanol doses. Decreased affinity of hemoglobin for oxygen and hypoventilation may result in limiting the amount of oxygen distributed to peripheral tissues. The central nervous system is particularly sensitive to a reduction in arterial oxygen saturation, which often results in unconsciousness [1]. The differential sensitivities of LS and SS mice to respiratory changes as a result of ethanol administration may contribute to the line difference in ethanol-induced narcosis. A more parsimonious explanation may be a differential effect of ethanol on the cardiovascular systems of LS and SS mice. Ethanol is known to produce a depression of the mechanical performance of cardiac muscle [4, 11, 16]. Decreasing cardiac output below normal causes a progressive decrease in total body oxygen consumption [7]. A decrease in oxygen consumption as a result of depressed cardiac output may be detected as an increase in PO2. LS mice show an increase in PO2 which may result from decreased oxygen consumption at much lower doses than in SS mice. Ethanol may thus have a depressant action on cardiac output which could result in decreased oxygen consumption for LS mice at 2.5 g/kg and for SS mice at 4.1 g/kg. Wallgren [24] has also shown that ethanol reduces cerebral oxygen uptake. Thus, a differential effect of ethanol on cardiac output may result in differences in oxygen consumption that contribute to the line difference in sensitivity to a hypnotic dose of ethanol. The results obtained in this study may be the consequence of differential sensitivities to ethanol of respiratory and cardiovascular regulatory mechanisms in LS and SS
ACUTE ETHANOL EFFECTS ON BLOOD
883
mice. Ascertaining w h e t h e r these differences are centrally and/or peripherally mediated will require further research. This study, h o w e v e r , further substantiates the differential
sensitivities of LS and SS mice to the acute administration o f ethanol and provides support for the notion that responses to ethanol are partially genetic in origin.
REFERENCES 1. Anderson, D. P., W. J. Allen, H. Bancroft, O. G. Edholm and G. W, Manning. Circulatory changes during fainting and coma caused by oxygen lack. J. Physiol. '104: 426-434, 1946. 2. Brobeck, J. R., editor. Best and Tailors' Physiological Basis of Medical Practice. Baltimore: Williams and Wilkins, 1979, pp. 4-43. 3. Collins, A. C. A review of research using the short-sleep and long-sleep mice. In: The Development of Animal Models as Pharmacogenetic Tools, edited by G. E. McClearn, R. A. Deitrich, and V. G. Erwin. (DHEW Publication No. (DM) 82-000). Washington, DC: U. S. Government Printing Office, in press. 4. Degerli, I. V. and W. R. Webb. Alcohol, cardiac function and coronary flow. Surg. Forum 14: 252-257, 1963. 5. Dundee, J. W., M. Isaac, S. K. Pandit and S. A. McDowell. Clinical studies of induction agents. XXIV. Further investigations with ethanol, Br. J. Anaesth. 42: 300-310, 1970. 6. Erwin, V. G., W. D. W. Heston, G. E. McClearn and R. A, Deitrich. Effect of hypnotics on mice genetically selected for sensitivity to alcohol. Pharmac. Biochem. Behav. 4: 679-683, 1976. 7. Guyton, A. C. Textbook of Medical Physiology. Philadelphia: W. B. Saunders Co., 1971, p. 488. 8. Johnstone, R. E. and C. E. Reier. Acute respiratory effects of ethanol in man. Clin. Pharmac. 14: 501-508, 1973. 9. Kakihana, R. Endocrine and autonomic studies in mice selectively bred for different sensitivity to ethanol. In: Alcohol Intoxication and Withdrawal, vol. 3 (Vol. 85a of Advances in Experimental Medicine and Biology), edited by M. Gross. New York: Plenum Press, 1977, pp. 83-95. 10. Kerlinger, F. N. and E. J. Pedhazur. Multiple Regression in Behavioral Research. New York: Holt, Rinehart and Winston, 1973, pp. 199-230. 11. Knott, D. H. and J. D. Beard. Changes in cardiovascular activity as a function of alcohol intake. In: The Biology of Alcoholism, vol. 2, Physiology and Behavior, edited by B. Kissin and H. Begleiter. New York: Plenum Press, 1972, pp. 345-365. 12. Matchett, J. A. and C. K. Erickson. Alteration of ethanolinduced changes in locomotor activity by adrenergic blockers in mice. Psychopharmacology 52: 201-206, 1977.
13. McClearn, G. E. and R. Kakihana. Selective breeding for ethanol sensitivity in mice. Behav. Genet. 3: 409-410, 1973. 14. Murphree, H. B. Electroencephalographic and other evidence for mixed depressant and stimulant actions of alcoholic beverages. Ann. N.Y. Acad. Sci. 215: 325-331, 1973. 15. Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner and D. H. Bent. SPSS, Statistical Package for the Social Sciences. New York: McGraw-Hill, 1975. 16. Regan, T. J., A. B. Weisse, C. B. Moschos, L. J. Lesnick, M. Nadami and W. K. Hellems. The myocardial effects of acute and chronic usage of ethanol in man. Trans. Ass. Am. Physns 78: 282-286, 1965. 17. Rh~iih/i, N. and P. M~ieni~i~. The influence of ethanol in the acid-base balance of the blood in man and rat. Scand. J. clin, Lab. Invest. 16: 267-272, 1964. 18. Rosenstein, R., L. E. McCarthy and H. L. Borison. Respiratory effects of ethanol and procaine injected into the cerebrospinal fluid of the brainstem in cats. J, Pharmac. exp. Ther. 162: 174-181, 1968. 19. Sanders, B. Sensitivity to low doses of ethanol and pentobarbital in mice selected for sensitivity to hypnotic doses of ethanol. J. comp. physiol. Psychol. 19: 394-398, 1976. 20. Scheffr, H. A. The Analysis of Variance. New York: Wiley, 1959. 21. Siemens, A. J. and A. W. K. Chan. Differential effects of pentobarbital and ethanol in mice. Life Sci. 19: 581-590, 1976. 22. Smith, A. A., C. Engelsher and M. Crofford. Respiratory or analgesic actions of ethanol and other narcotics: Modulation by biogenic amines. Ann. N.Y. Acad. Sci. 273: 256-262, 1976. 23. Sorensen, S., M. Palmer, J. Dunwiddie and B. Hoffer. Electrophysiological correlates of ethanol-induced sedation in differentially sensitive lines of mice. Science 210:1143-1145, 1980. 24. Wallgren, H. Effect of ethanol on intracellular respiration and cerebral function. In: The Biology of Alcoholism, vol. 1, Biochemistry, edited by B. Kissin and H. Begleiter. New York: Plenum Press, 1971, pp. 103-125.
Acute Ethanol Effects on Blood pH, PCO2, and PO2 in LS and SS Mice' D A V I D M. G I L L I A M A N D A L L A N
C. C O L L I N S z
Alcohol Research Center, Department of Psychology, School of Pharmacy and Institute for Behavioral Genetics, University of Colorado Boulder, CO 80309 R e c e i v e d 15 J u l y 1981 GILLIAM, D. M. AND A. C. COLLINS. Acute ethanol effects on blood pH, PC02, and P02 in LS and SS mice.
PHYSIOL. BEHAV. 28(5) 879--883, 1982.--The effects of ethanol on blood pH, PCO2, and POx were measured in LS and SS mice in an attempt to ascertain whether these lines of mice, which differ in CNS sensitivity to the behavioral effects of ethanol, also differ in sensitivity to physiological effects of this drug. Long-sleep (LS) female mice were injected intraperitoneally with 1.8, 2.5, 3.3, or 3.8 g/kg ethanol; short-sleep (SS) female mice were administered 2.5, 3.3, 4.1, or 4.7 g/kg. Blood pH, PCO2, and POz were assessed at 15, 30, 60, 120, or 180 min after injection of the 2.5 and 4.1 g/kg doses or at 60 min after injection of the 1.8, 2.5, 3.3, 3.8, 4.1, and 4.7 g/kg doses. Opposite effects on blood pH and PCO2 over time were obtained in LS and SS mice at the 2.5 g/kg dose. Acidosis characterized the LS line, whereas alkalosis characterized the SS. The results obtained with SS mice at the 4.1 g/kg dose were similar to those obtained with LS mice at the 2.5 g/kg dose. The dose-response curve for the SS mice generated at 60 min post-injection lies to the right of that for the LS mice. The effects of high ethanol doses on SS mice resemble the effects of low doses on LS animals. Thus, the two lines of mice differ in response to the effects of ethanol on these parameters related to respiration. The difference in sensitivity to the respiratory depressant effects of ethanol may contribute to the differences in behavioral sensitivity between the two lines. Ethanol
Blood pH, PCO2, and POz
Selected mice
T H E central nervous system (CNS) depressant actions of hypnotic agents are well known. Among CNS areas known to be susceptible to the depressant actions of hypnotic agents are respiratory centers o f the medulla. Several hypnotic agents have been shown to produce an increase in ventilation at low doses and a decrease in ventilation at high doses. Low doses o f ethanol, for example, produce an increase in rate of respiration [5, 14, 17], while a decrease in rate is seen at higher doses [8, 14, 17, 18, 22]. This change in rate may be a consequence of the effects of ethanol on centrally mediated responses to COz [8]. Biphasic responses are characteristic of ethanol, which is also known to produce biphasic behavioral responses [12,19]. Differences in response to ethanol administration have been well studied in the long-sleep (LS) and short-sleep (SS) lines that have been selectively bred for differential sensitivity to a hypnotic dose o f ethanol [13]. Research using the LS and SS mice has recently been reviewed [3]. Greater sensitivity to ethanol in LS mice has been demonstrated by a lower EDs0 for loss of the righting reflex [6], longer sleep time [6,21] (Gilliam and Collins, to be published), greater decrease in body temperature [9], and decrease in open-field activity [19] (Ritz, George and Collins, to be published) at lower doses of ethanol as compared to SS mice.
In view of these differences in behavioral sensitivity to ethanol, it seemed reasonable to suspect that ethanol may cause decreased ventilation in LS mice at lower doses of ethanol than in SS mice. Decreased ventilation should result in increased blood PCO2 and decreased blood pH, whereas increased ventilation should result in decreased PCO2 and increased pH. Blood PO2 is probably influenced by both respiration rate and tissue utilization. The current study was initiated in order to determine whether LS and SS mice do, indeed, differ in the effects of ethanol on respiration. Respiration rate is quite difficult to monitor in the mouse; fortunately, however, recent advances in the measurement of blood pH, PCO2, and PO2 have made it quite easy to measure these related parameters. The data presented in this report demonstrate that LS and SS mice differ in the effects of ethanol on respiration. This effect may contribute to the differences between the lines in their sensitivity to the behavioral effects of ethanol. METHOD
Subjects Female LS and SS mice from the 21st generation of the selective breeding program were used in this study. These
1This research was supported in part by a Research Scientist Development Award to A. C. Collins (AA-00029) and by NIAAA research grant AA-03527. It was also partially supported by BRSG grant RR-07013-14 awarded by the Biomedical Research Support Grant Program, Division of Research Resources, National Institutes of Health. ~Send reprint requests to Dr. Allan C. Collins, Institute for Behavioral Genetics, Box 447, University of Colorado, Boulder, CO 80309.
C o p y r i g h t © 1982 B r a i n R e s e a r c h P u b l i c a t i o n s Inc.--0031-9384/82/080879-05503.00/0
880
GILLIAM AND COLLINS
animals were tested at 70_+10 days of age between 1300 and 1600 hr. The normal 12-hr light cycle (lights on from 700 to 1900 hr) was maintained throughout the experiment. Ambient temperature was 22___1.5°C. Littermates were distributed randomly among all time points and doses in the following experiments.
•
~
-
-
'~, T
7.30C
~
•......
, . . . . .
,
<~
'
TLaA'---~-----(~---'-~
,=
........
SS 2 , S g / k # ss 4.1 gzkq
7.25(
7,2tX
Time Course
LS mice were injected intraperitoneally (IP) with 2.5 g/kg ethanol; SS mice received an IP ethanol dose of either 2.5 or 4.1 g/kg. Injection volumes were 0.01 ml/g body weight (BW) for the 2.5 g/kg dose and 0.013 ml/g BW for the 4.1 g/kg dose. Different groups of ethanol-treated mice were then tested for blood pH, PCO2, and PO2 at 15, 30, 60, 120, or 180 rain post-injection. Uninjected and saline-injected controls were also tested to establish control levels of pH, PCO2, and PO2. Blood samples were obtained from the retro-orbital sinus using 120-/zl heparinized capillary tubes. Blood pH, PCO2, and PO2 were determined by inserting the blood sample into a BMS3 Mk2 blood micro system (Radiometer Copenhagen) equipped with a PHM73 pH/blood gas monitor (Radiometer Copenhagen).
7°150 i i
CONT~
1~
30
60 120 MINIJTE S P O S T -IN,,ECTION
180
FIG. 1. Mean blood pH values for LS mice given 2.5 g/kg (A) and for SS mice given 2.5 g/kg (O) or 4.1 g/kg (12]) ethanol IP. Different groups of mice were assessed at 15, 30, 60, 120, or 180 min postinjection. N=6 to 9 per group. Solid triangle and solid hexagon represent control values for LS (N= 15) and SS iN= 15) mice, respectively. Brackets indicate -+1 SE.
Dose Response
|
LS mice were injected IP with 1.8, 2.5, 3.3, or 3.8 g/kg ethanol; the ethanol dose received by SS mice was either 2.5, 3.3, 4.1, or 4.7 g/kg. Injection volumes were 0.01 mug BW for the 1.8, 2.5, and 3.3 g/kg doses; 0.016 ml/g BW for the 3.8 g/kg dose; 0.013 mug BW for the 4.1 g/kg dose; and 0.02 ml/g BW for the 4.7 g/kg dose. Increased volumes for the higher doses were used to insure survival of the subjects for the duration of the experiments. Different groups of ethanoltreated mice were tested for blood pH, PCO2, and PO2 at 60 min after injection. Blood samples were obtained and analyzed in the manner described above, and uninjected and saline-injected controls were tested to establish control levels of pH, PCO~, and POv Data on animals injected with the 2.5 or 4.1 g/kg doses and tested at 60 rain were combined within line in the time-course and dose-response analyses, as were data on uninjected and saline-injected controls.
L
I
A
,
[] ss4.~g,kg
D-
3Q
Analyses
The significance of variations in these parameters was assessed by multivariate analysis of variance with the Statistical Package for the Social Sciences [ 15]. The Wilks test was used to determine statistical significance. One-way analyses of variance with post hoc tests (Scheff~ [20]: p =0.05) were performed on scores generated by the first standardized discriminant function coefficients from the interaction term (for data resulting from the time-course experiment) or from the main effect (for data resulting from the dose-response experiment). Trend analysis was performed following the methods described by Kerlinger and Pedhazur [10]. RESULTS Saline injection did not have an effect on blood pH, PCO~, or POz at any time. Data collected from saline-injected or uninjected mice were combined and represent control levels. Significant differences between lines, F(3,26)=3.3, p<0.05, were found for control levels of pH, PCO~, and PO~ taken collectively. Control levels of blood pH were higher in LS than in SS mice, while control levels of PCO2 and POz did not differ significantly between lines.
MINUTES
POST o ~NJEC TION
FIG. 2. Mean blood PCO2 values for LS mice given 2.5 g/kg (A) and for SS mice given 2.5 g/kg (O) or 4.1 g/kg ([]) ethanol ]P. Different groups of mice were assessed at 15, 30, 60, 120, or 180 rain postinjection. N=6 to 9 per group. Solid triangle and solid hexagon represent control values for LS (N= 15) and SS (N= 15) mice, respectively. Brackets indicate _ 1 SE.
Time Course
Figures I, 2, and 3 show mean values of pH, PCO2, and POz, respectively, for control animals and for experimental animals at all time points. When composite blood pH, PCO2, and PO2 measures for LS and SS mice injected with 2.5 g/kg ethanol were compared at 15, 30, 60, 120, or 180 min postinjection, results indicated a significant interaction between line and time, F(15,218)=4.5, p<0.001. Results of the multivariate analysis revealed that SS composite values did not differ from control at any time point. However, LS compo-
ACUTE E T H A N O L EFFECTS ON BLOOD
T 50
/k LS 2,5 g/kg 0 SS 2,5 9tkg
\
,\
881
[] ss 4.1
/x
/k LS 7,35Q
~- ~ 4
Q~ SS
\
g~kg 7,30C
~ ~ C)
7,2O0
7,15Q
3o
i
1,8
i
i
2,5 3,,3 ETHANOL DOSE (g/k~
i
i
3,8 4,1
i
4,7
FIG. 4. Mean blood pH values at 60 min post-injection for LS mice (A) given 1.8, 2.5, 3.3, or 3.8 g/kg or for SS mice ( O ) given 2.5, 3.3, 4.1, or 4.7 g/kg ethanol IP. N=6 to 9 per group. Solid symbols indicate control values. Brackets indicate ___1 SE.
~ co.;~ ,'~ ~'o
i
CONTROL
go
Go
40
MINUTES POST-INJECTION
FIG. 3. Mean blood PO~ values for LS mice given 2.5 g/kg (A) and for SS mice given 2.5 g/kg (©) or 4.1 g/kg ([]) ethanol IP. Different groups of mice were assessed at 15, 30, 60, 120, or 180 rain postinjection. N=6 to 9 per group. Solid triangle and solid hexagon represent control values for LS (N=15) and SS (N=15) mice, respectively. Brackets indicate __.1 SE.
/1 LS
5o
0 SS
45
site values were found to differ from control at 15, 30, and 60 rain post-injection. For LS mice, pH decreased from control levels at 15 rain and remained depressed through 60 min (Fig. 1), PCO~ was markedly elevated at 60 rain (Fig. 2); and PO 2 was greatly elevated at 15, 30, and 60 rain, with a return to near normal levels at 120 and 180 rain post-injection (Fig. 3). An unexpected finding was that the composite blood measures in the SS mice injected with the 4.1 g/kg dose and LS mice injected with 2.5 g/kS showed similar changes over time, F(15,213)=0.1, p >0.05. Trend analysis revealed significant LS-SS differences in the equations that describe the effects of the 2.5 g/ks dose on blood pH, PCO2, and PO2 over time. However, when LS mice injected with 2.5 g/ks were compared with SS mice injected with 4.1 g/kg, similar equations were found to describe the time course of the PO~ measures. Trends for pH and PCO2 differed between lines. The composite blood measures for SS mice injected with 2.5 or 4.1 g/kg ethanol were compared at 15, 30, 60, 120, and I80 min post-injection, The results of this analysis indicated a significant interaction between time and dose, F(15,174)=3.5, p<0.001. Multivariate analysis revealed that none of the values for SS mice injected with the 2.5 g/ks dose differed from control, but that the measures for SS mice given the 4.1 g/kg dose differed from the control value at 15, 30, 60, and I20 rain. For SS mice injected with 4.1 g/kg, pH decreased from the control value at 15 min post-injection, remained depressed through 120 min, and returned to the control value at 180 rain (Fig. 1); PCO2 was elevated at 60 and 120 min post-injection (Fig. 2); and PO~ increased from the control value at 15 rain and remained elevated through 180 rain (Fig. 3).
u 4o 35
30
ETHANOL DOSE (g/kg)
FIG. 5. Mean blood PC02 values at 60 rain post-injection for LS
mice (A) given 1.8, 2.5, 3.3, or 3.8 g/kg or for SS mice ( O ) given 2.5, 3.3, 4.1, or 4.7 g/kg ethanol IP. N--6 to 9 per group. Solid symbols indicate control values. Brackets indicate _+1 SE.
Dose Response Figures 4, 5, and 6 show mean values of pH, PCO~, and PO~, respectively, for control animals and for experimental animals at each dose at 60 rain post-injection. For both the LS and SS lines, composite blood measures varied as a function of dose, F(12,85)= 18.6, p <0.001; F(12,96)= 10.2, p<0.001, respectively. Values for LS mice given a dose of 2.5, 3.3, or 3.8 g/kg and for SS mice that received 3.3, 4.1, or 4.7 g/kg differed significantly from control. Both LS and SS mice show a dose-dependent decrease in pH and a dosedependent increase in PO~. There is an elevation of PCO~ for LS mice at the 2.5 g/kg dose and for SS mice at the 4.1 g/kg dose. Trend analysis revealed that the equations that describe the dose-response curves for pH, PCO~, or PO~ do not
~82
G I L L I A M AND C O I , I J N S
z~ LS 50
0
SS
F
5 35
30
CONTROL
1'.8
2'.5 3.3 ETHANOL DOSE (g/kg}
3.8 4.1
45
FIG. 6. Mean blood PO2 values at 60 min post-injection for LS mice (A) given 1.8, 2.5, 3.3, or 3.8 g/kg or for SS mice ( O ) given 2.5, 3.3, 4.1, or 4.7 g/kg ethanol IP. N=6 to 9 per group. Solid symbols indicate control values. Brackets indicate _+1 SE.
differ significantly between the LS and SS lines. The doseresponse curve for the SS mice thus represents a rightward shift in the dose-response curve for the LS mice. DISCUSSION
This study demonstrates that the effects of ethanol administration on blood pH, PCO2, and PO2 occur at lower doses for LS mice than for SS mice. LS mice given a dose of 2.5 g/kg show a decrease in blood pH, a general increase in PCO2, and a greatly increased PO2 over time. The same dose in SS animals produces very different effects: an increase in blood pH, a substantial decrease in PCO2, and only a slight increase in PO2 over time. The response of SS mice to a much higher dose of ethanol (4.1 g/kg) is similar to the response seen in LS mice at the 2.5 g/kg dose. Blood pH is decreased, and there is an increase in both PCO2 and PO2. The differences in sensitivity of LS and SS mice to ethanol's effects on blood pH, PCO2, and PO2 are particularly well demonstrated in their response to increasing doses of ethanol at 60 min post-injection. Blood pH in LS mice shows an increase at the 1.8 g/kg dose and a progressive decrease at higher doses. SS mice exhibit a similar pattern, but their responses lie to the right of LS mice on the doseresponse curve. For example, SS mice require higher doses of ethanol than LS mice in order to exhibit decreases in pH. The differential sensitivities of LS and SS mice with respect to changes in blood pH at increasing doses of ethanol are mirrored by changes in PCO2 and PO2. Interestingly, a change in the injection volume seems to influence the effect of ethanol on pH, PCOz, and PO2. LS mice given 3.3 g/kg ethanol in an injection volume of 0.01 ml/g BW showed a greater decrease in pH than did LS mice given 3.8 g/kg ethanol in an injection volume of 0.016 ml/g BW. A similar phenomenon was also seen for SS mice. SS mice given 4.1 g/kg ethanol in an injection volume of 0.013 ml/g BW showed a greater decrease in blood pH than SS mice given 4.7 g/kg ethanol in an injection volume of 0.02 ml/g BW. The cause of this reversal in effect of ethanol on blood pH is not clear. However, a concentration-dependent effect of ethanol on
blood pH, PCO2, and PO2, as well as on respiration rate, in the LS and SS mice has been verified in our laboratory (Gilliam and Collins, to be published). As noted previously, ethanol is known to produce respiratory depression at high doses [8, t4, 17, 18, 22], while low doses may increase respiration rate [5, 14, 17]. Hyperventilation leads to respiratory alkalosis (increased blood pH) and decreased PCOz. These changes characterize the responses of LS mice to the 1.8 g/kg dose and SS mice to the 2.5 g/kg dose at 60 rain post-injection. The results seen with the LS mice at 2.5 g/kg and the SS mice at 4.1 g/kg may be a consequence of ethanol's depressant action on central respiratory centers. Therefore, the medullary centers responsible for respiration may be differentially sensitive to ethanol's effects in the LS and SS mice. This notion receives support from the observation that LS cerebellar Purkinje cells are one to two orders of magnitude more sensitive to the depressant effects of locally administered ethanol than are those from the SS mice [23]. Ethanol has been shown to decrease the centrally mediated ventilatory response to CO ~ in a dose-dependent manner [8]. Respiratory depression and/or decreased responsiveness to CO2 result in respiratory acidosis, which would cause a decrease in blood pH. Metabolic acidosis resulting from the metabolism of ethanol to acetic acid or a rise in blood lactate during ethanol oxidation may contribute to the decrease in blood pH following the administration of ethanol in doses equal to or greater than 2.5 g/kg for LS mice or 3.3 g/kg for SS mice. A somewhat surprising finding was the increase in PO~ seen following ethanol administration. We had expected to find reduction in PO.,, especially at the higher ethanol doses, as a result of the respiratory depression. The observed increase in PO~ may be explained by the fact that decreases in blood pH are known to lower the affinity of the hemoglobin molecule for oxygen [2], the Bohr effect. Thus, the presence of more oxygen in the dissolved form may contribute to the initial increase in POz seen in LS and SS mice at higher ethanol doses. Decreased affinity of hemoglobin for oxygen and hypoventilation may result in limiting the amount of oxygen distributed to peripheral tissues. The central nervous system is particularly sensitive to a reduction in arterial oxygen saturation, which often results in unconsciousness [1]. The differential sensitivities of LS and SS mice to respiratory changes as a result of ethanol administration may contribute to the line difference in ethanol-induced narcosis. A more parsimonious explanation may be a differential effect of ethanol on the cardiovascular systems of LS and SS mice. Ethanol is known to produce a depression of the mechanical performance of cardiac muscle [4, 11, 16]. Decreasing cardiac output below normal causes a progressive decrease in total body oxygen consumption [7]. A decrease in oxygen consumption as a result of depressed cardiac output may be detected as an increase in PO2. LS mice show an increase in PO2 which may result from decreased oxygen consumption at much lower doses than in SS mice. Ethanol may thus have a depressant action on cardiac output which could result in decreased oxygen consumption for LS mice at 2.5 g/kg and for SS mice at 4.1 g/kg. Wallgren [24] has also shown that ethanol reduces cerebral oxygen uptake. Thus, a differential effect of ethanol on cardiac output may result in differences in oxygen consumption that contribute to the line difference in sensitivity to a hypnotic dose of ethanol. The results obtained in this study may be the consequence of differential sensitivities to ethanol of respiratory and cardiovascular regulatory mechanisms in LS and SS
ACUTE ETHANOL EFFECTS ON BLOOD
883
mice. Ascertaining w h e t h e r these differences are centrally and/or peripherally mediated will require further research. This study, h o w e v e r , further substantiates the differential
sensitivities of LS and SS mice to the acute administration o f ethanol and provides support for the notion that responses to ethanol are partially genetic in origin.
REFERENCES 1. Anderson, D. P., W. J. Allen, H. Bancroft, O. G. Edholm and G. W, Manning. Circulatory changes during fainting and coma caused by oxygen lack. J. Physiol. '104: 426-434, 1946. 2. Brobeck, J. R., editor. Best and Tailors' Physiological Basis of Medical Practice. Baltimore: Williams and Wilkins, 1979, pp. 4-43. 3. Collins, A. C. A review of research using the short-sleep and long-sleep mice. In: The Development of Animal Models as Pharmacogenetic Tools, edited by G. E. McClearn, R. A. Deitrich, and V. G. Erwin. (DHEW Publication No. (DM) 82-000). Washington, DC: U. S. Government Printing Office, in press. 4. Degerli, I. V. and W. R. Webb. Alcohol, cardiac function and coronary flow. Surg. Forum 14: 252-257, 1963. 5. Dundee, J. W., M. Isaac, S. K. Pandit and S. A. McDowell. Clinical studies of induction agents. XXIV. Further investigations with ethanol, Br. J. Anaesth. 42: 300-310, 1970. 6. Erwin, V. G., W. D. W. Heston, G. E. McClearn and R. A, Deitrich. Effect of hypnotics on mice genetically selected for sensitivity to alcohol. Pharmac. Biochem. Behav. 4: 679-683, 1976. 7. Guyton, A. C. Textbook of Medical Physiology. Philadelphia: W. B. Saunders Co., 1971, p. 488. 8. Johnstone, R. E. and C. E. Reier. Acute respiratory effects of ethanol in man. Clin. Pharmac. 14: 501-508, 1973. 9. Kakihana, R. Endocrine and autonomic studies in mice selectively bred for different sensitivity to ethanol. In: Alcohol Intoxication and Withdrawal, vol. 3 (Vol. 85a of Advances in Experimental Medicine and Biology), edited by M. Gross. New York: Plenum Press, 1977, pp. 83-95. 10. Kerlinger, F. N. and E. J. Pedhazur. Multiple Regression in Behavioral Research. New York: Holt, Rinehart and Winston, 1973, pp. 199-230. 11. Knott, D. H. and J. D. Beard. Changes in cardiovascular activity as a function of alcohol intake. In: The Biology of Alcoholism, vol. 2, Physiology and Behavior, edited by B. Kissin and H. Begleiter. New York: Plenum Press, 1972, pp. 345-365. 12. Matchett, J. A. and C. K. Erickson. Alteration of ethanolinduced changes in locomotor activity by adrenergic blockers in mice. Psychopharmacology 52: 201-206, 1977.
13. McClearn, G. E. and R. Kakihana. Selective breeding for ethanol sensitivity in mice. Behav. Genet. 3: 409-410, 1973. 14. Murphree, H. B. Electroencephalographic and other evidence for mixed depressant and stimulant actions of alcoholic beverages. Ann. N.Y. Acad. Sci. 215: 325-331, 1973. 15. Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner and D. H. Bent. SPSS, Statistical Package for the Social Sciences. New York: McGraw-Hill, 1975. 16. Regan, T. J., A. B. Weisse, C. B. Moschos, L. J. Lesnick, M. Nadami and W. K. Hellems. The myocardial effects of acute and chronic usage of ethanol in man. Trans. Ass. Am. Physns 78: 282-286, 1965. 17. Rh~iih/i, N. and P. M~ieni~i~. The influence of ethanol in the acid-base balance of the blood in man and rat. Scand. J. clin, Lab. Invest. 16: 267-272, 1964. 18. Rosenstein, R., L. E. McCarthy and H. L. Borison. Respiratory effects of ethanol and procaine injected into the cerebrospinal fluid of the brainstem in cats. J, Pharmac. exp. Ther. 162: 174-181, 1968. 19. Sanders, B. Sensitivity to low doses of ethanol and pentobarbital in mice selected for sensitivity to hypnotic doses of ethanol. J. comp. physiol. Psychol. 19: 394-398, 1976. 20. Scheffr, H. A. The Analysis of Variance. New York: Wiley, 1959. 21. Siemens, A. J. and A. W. K. Chan. Differential effects of pentobarbital and ethanol in mice. Life Sci. 19: 581-590, 1976. 22. Smith, A. A., C. Engelsher and M. Crofford. Respiratory or analgesic actions of ethanol and other narcotics: Modulation by biogenic amines. Ann. N.Y. Acad. Sci. 273: 256-262, 1976. 23. Sorensen, S., M. Palmer, J. Dunwiddie and B. Hoffer. Electrophysiological correlates of ethanol-induced sedation in differentially sensitive lines of mice. Science 210:1143-1145, 1980. 24. Wallgren, H. Effect of ethanol on intracellular respiration and cerebral function. In: The Biology of Alcoholism, vol. 1, Biochemistry, edited by B. Kissin and H. Begleiter. New York: Plenum Press, 1971, pp. 103-125.