Вы находитесь на странице: 1из 5

Psychopharmacologia (Berl.

) 46, 301-305 (1976) - 9 by Springer-Verlag 1976

Circadian Rhythm of Corticosterone in Mice"

The Effect of Chronic Consumption of Alcohol
Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine,
Stanford, California 94305, U.S.A.
Received October 21, 1975
Abstract. The effect of chronic consumption of alcohol on
the circadian variations of the plasma corticosterone was
investigated in DBA/2J male mice. After 15 weeks of alcohol
consumption (3.8~ow/v for the first week and 7.5~ for
subsequent weeks) the alcohol groups exhibited a flattened
circadian corticosterone curve, the level being intermediate
between the peak and trough values of the water control
groups. The diurnal patterns of food and liquid consumption
were still present at the 10th week of alcohol treatment in the
alcohol groups, although the absolute amount of food and

liquid consumed at each of the 6-h intervals was somewhat

different between the alcohol and water groups. The blood
alcohol showed a peak at early morning with the mean of
100 mg/100 ml, but the levels of alcohol during the remaining
periods were remarkably stable, the means ranging from
30 to 46 mg/100 ml. Chronic consumption of alcohol, even
relatively low concentrations, appears to affect the neural
sites in the CNS controlling the circadian rhythm of ACTH

Key words: Alcohol - Corticosterone -Circadian rhythm - Mice.

The effect of alcohol on the pituitary-adrenal system

has been studied in laboratory mice and rats (Santisteban and Swinyard, 1956; Czaja and Kalant, 1961;
Ellis, 1966; K a k i h a n a et al., 1968). Earlier investigators (Santisteban and Swinyard, 1956; Czaja and
Kalant, 1961) measured the changes of ascorbic acid
content of the adrenals and more recent ones employed
the direct measurement of the circulating levels
of corticosterone (Ellis, 1966; K a k i h a n a et al., 1968).
The results of studies by Ellis (1966) strongly indicate
that alcohol activates the pituitary-adrenal system,
via the release of corticotrophin releasing factor (CRF)
from the hypothalamus. We also found that dexamethasone injection 1 h prior to alcohol challenge,
completely blocked its stimulatory effect on the
adrenal cortex in mice (Kakihana et al., unpublished
One of the well documented characteristics of the
pituitary-adrenal system is the circadian variation of
the circulating corticosterone. The levels of corticosterone typically rise toward late afternoon to early
evening just prior to the onset of darkness in the
laboratory r a t s and mice ( Z i m m e r m a n and Critchlow, 1967; Dixit and Buckley, 1967; Butte et al., 1975;
Halberg et al., 1959). One of the few studies on h u m a n
subjects on the effects of alcohol on the pituitaryadrenal function suggests that alcoholics show a

circadian rhythm with two peaks, rather than a

typical single daily peak, of the plasma 17-hydroxycorticosteroids (Margraf et al., 1967). The present
study was initiated to investigate the effect of chronic
consumption of alcohol on the circadian rhythm of
plasma corticosterone in laboratory mice.

Animals and Alcohol Treatment. The experiment was conducted on 114 male DBA/2J mice, purchased from the Jackson
Laboratory, Bar Harbor. The mice were housed in groups
of five on a 12-h light-dark cycle (7:00-19:00 hours) until
2 days prior to blood sampling at which time they were housed
Cages containing 5 mice each were randomly assigned
to alcohol and control groups. The alcohol group received
3.8~ ethanol (w/v) in water as the sole liquid source and
controls received water. After I week the alcohol concentration
was increased to 7.5 % and maintained at this concentration
for the next 14 weeks until the animals were sacrificed for
hormone determination. Berkeley Diet Rat and Mouse Food
was continuously available throughout the experiment.
Food and Liquid Intake. After 10 weeks of alcohol treatment,
food and liquid intake was measured for each cage at 12:00,
18 : 00, 00 : 00 and 6 : 00 hours for two consecutive 24-h
periods. Plastic syringes (30 ml, without plungers) fitted with
metal drinking tips were used to measure liquid consumption.
Mean food and liquid intake based on 2 days of observation
was used for statistical analysis. Body weights were recorded


Psychopharmacologia (Bed.), Vol. 46, Fasc. 3 (1976)

at the beginning of alcohol treatment, on the first day of food

and liquid consumption measurement, and 1 day before the
animals were sacrificed for hormone assay.

Blood Collection. Five weeks after measurement of food and

liquid consumption 9 or 10 mice from each group were sacririced at 6:30, 10:30, 14:30, 18:30, 22:30 and 2:30hours
and blood samples were collected in the following manner,
Animals were etherized 4 0 - 5 0 s and blood was drawn from
the heart in a 1 ml syringe with a 25 g" needle rinsed in
heparinized saline. Less than 3 min elapsed between the time
the animal was picked up and the completion of blood collection. Ten microliters of whole blood was then transferred to
culture tubes containing 0.09 ml of cold 3.4% perchloric
acid and frozen to be assayed for blood alcohol. The rest of
the blood was centrifuged at 1500 g for 10 min and the
plasma frozen to be assayed for corticosterone.










Corticosterone and Alcohol Assay. Plasma corticosterone was

determined by the method of Butte and Noble (1969) with the
following modifications. For the column chromatographic
separation of the steroids, 1,1,2 trichloro-l,2,2 trifluoroethane
(Freon 113, Dupont) was used after redistillation instead of
carbon tetrachloride as used in the original procedure. Corticosterone was eluted from the column as follows. After discarding two fractions (10 ml of 100~ Freon 113 and 6 ml of
65 ~ Freon 113/35 ~ methylene chloride, v/v), a third fraction
of 12 ml 65 ~ Freon 113/35 ~ methylene chlorine contained
all of the corticosterone. This fraction was evaporated and
assayed. Blood alcohol levels were determined by the method
of Lundquist (1959).











Figure 1 shows the variation in f o o d and liquid cons u m p t i o n over 24-h periods in the alcohol and control
groups. The overall m e a n f o o d intake o f the alcohol
groups was 3.3 g 0.1/mouse/day in contrast to the
m e a n o f 4.4 g + 0 . 2 / m o u s e / d a y in the water groups,
the difference being statistically significant by
t-test (t = 5.63, df = 23, P < 0.01). The m e a n liquid
c o n s u m p t i o n was also significantly lower in alcohol
groups t h a n in the controls (3.5 and 5.9 m l / m o u s e / d a y ,
respectively; t -- 8.65, df = 23, P < 0.01).
The m e a n b o d y weights o f water and alcohol groups
are presented in Table 1. The small but significant
difference between the two g r o u p s prior to the treatment m u s t have resulted by chance, since cages containing 5 mice each were assigned r a n d o m l y to each
treatment group. A l t h o u g h the m e a n f o o d and liquid
intake o f alcohol groups were 25 ~ and 41 O//oless than
the respective means o f the water groups, the total
calories c o n s u m e d (food and alcohol) per g r a m o f
b o d y weight during the 10th week o f alcohol treatm e n t were r e m a r k a b l y similar (0.387 calories/g m o u s e /
day in alcohol g r o u p s c o m p a r e d to 0.418 in water
groups). After 15 weeks o f alcohol c o n s u m p t i o n , the
alcohol g r o u p s showed a slightly, but significantly,
lower m e a n b o d y weight than the water groups
(t = 2.61, d f = 113, P < 0.01).





Fig. 1. Mean food and liquid consumption of DBA/2J male

mice during alcohol treatment. Food and liquid intake was
measured for each cage containing 5 mice for two consecutive
24-h periods after 10 weeks of alcohol treatment.
* P < 0.005; ** P < 0.001; vertical bars = S.E.M.

Table 1. Mean body weight (g) of water and alcohol groups

on treatment

Water group

Alcohol group

M _+ S.E.M.

M S.E.M.


24.9 _+_0.3
28.3 0.3
27.5 0_+0.3


26.0* 0.3
27.9 0.3
26.6* _+ 0.3

* Significantly different from water group at P < 0.01.

The blood alcohol concentrations over a 24-h

period are shown in Table 2. The analysis o f variance
applied to these data indicates a statistically significant
difference a m o n g these groups (F = 5.31, df = 5/49,
P < 0.01). F u r t h e r m o r e , Scheffee's post hoc comparison (Hays, 1963) revealed that the peak b l o o d alcohol
concentration at 6 : 3 0 hours is significantly higher
than any other time point examined. This perhaps

R. Kakihana and J. A. Moore: Alcohol and Adrenocortical Activity

Fig. 2. Circadian variations of the plasma corticosterone in the chronically alcohol consuming
DBA/2J male mice. Blood samples were collected after 15 weeks of alcohol consumption,
3.8 ~ (w/v) during the first week and 7.5 ~ for
the next 14 weeks. Numbers in parentheses
represent number of mice assayed for each
time point. Vertical bars indicate S.E.M.













I- 15.00







1o-ot \ /






' ~









Table 2. Blood alcohol concentration of alcohol groups over

24-h period
Time of day

Blood alcohol (mg/100 ml)

M S.E.M.



100" J9
40 6
30 4
41 11
46 10
40 7

Blood alcohol samples were taken when the animals were

sacrificed after 15 weeks of alcohol consumption, 3.8 ~ (w/v)
for the first week and 7.5 ~ for remaining 14 weeks.
* P < 0.05.

reflects the observation that almost 7 5 ~ of alcohol

drinking takes place between 00:00 and /2:00 hours
(Fig. l).
The circadian variations of the plasma corticosterone in the alcohol and water groups are shown in
Figure 2. The typical circadian variations are evident in
the water groups (F = 3.38, df= 5/46, P < 0.05).
The mean plasma corticosterone of the water groups
at 18 : 30 hours was significantly higher than the 10:30
and 22:30-hours means (t = 3.7, dr= 16, P < 0.005
and t = 2.32, df = 16, P < 0.025, respectively). The
14:30 and 18:30-hours levels were not statistically
different. In contrast to the water groups, corticosterone levels of alcohol groups show no group differences due to the time of day sampled (F = 0.34,
df = 5/48, n.s.).

The main finding of the present study is that the
typical circadian variations of plasma corticosterone
disappear entirely in mice drinking alcohol chronically
(15 weeks). Figure 2 clearly shows that during the
normal trough period the corticosterone levels of
the alcohol groups are higher than those of the water
groups, but the relationship reverses itself during the
normal peak period, presenting a flattened circadian
profile. Whether such an endocrine profile indicates
a general hypoactivity or hyperactivity of the adrenocortical system is hard to assess at this point. The only
available data concerning alcohol and the circadian
variations of the adrenocorticosteroids are those
reported by M a r g r a f et al. (1967) on h u m a n alcoholics.
These data suggest that the diurnal variations of the
alcoholics are more accentuated and show two peaks
rather than a typical one peak. The possible reasons
for the differences in these findings are easily conceived, however. The circadian data reported by
M a r g r a f et al. (/967) are based on three subjects with
" k n o w n histories of chronic alcoholism". But it is
not clear from the report whether these alcoholic
patients were drinking alcohol or not at the time of the
study. Since no blood alcohol data are reported, it
seems safe to assume that the alcoholic subjects were
already withdrawn from alcohol. Therefore, the
unusual two-peaked circadian profile of these alcoholic
subjects m a y represent the recovery phase of the
normal endocrine function. F o r these reasons the
direct comparison of our results with those of M a r g r a f
et al. (1967) are not valid.

It is interesting to note that exactly the same change
in the circadian rhythm of corticosterone, as we have
shown in the present study, was produced in rats by
the injection of p-chlorophenylalanine (pCPA), a
drug that blocks serotonin (5-HT) synthesis (Scapagnini et al., 1971). These investigators further demonstrated that the pCPA produced a significant reduction
in the 5-HT content of the amygdala, hippocampus
and hypothalamus. Based on these data, they made a
suggestion that serotonergic neurons may be mediating
the diurnal fluctuation in the pituitary-adrenal function. The current literature on the effect of alcohol on
brain serotonin and its metabolism is still varied and
conflicting (for discussion, see Feldstein, 1971; Blum
et al., 1973; Hunt and Majchrowicz, 1974). While
some studies seem to indicate a decrease in serotonin
levels in the brain of rabbits, mice and rats following
a single injection of alcohol (Gursey and Olson, 1960;
Feldstein, 1973; Kuriyama et al., 1971), others show
increased 5-HT levels after acute treatment with ethanol but no change during chronic treatment in rat
brain (Paiai6 et al., 1971) or no change in mouse brain
during alcohol withdrawal (Tabakoff and Boggan,
1974). A recent paper by Hunt and Majchrowicz (1974)
provided evidence suggesting alcohol exerts little or
no effect on the steady-state levels of serotonin nor on
the rate of serotonin turnover in rats rendered dependent on alcohol by intragastric intubation of alcohol.
The conflicting results from various laboratories may
imply the lack of systematic and functional relationship between ethanol and serotonin metabolism or it
may imply the complexity of the relationship influenced by specific experimental procedures, species
and, possibly, by genetic factors. In support of the
genetic variable, Ahtee and Eriksson (1972) reported
that chronic alcohol drinking resulted in an increase
of the 5-HT content in the brain of alcohol addicted
rats, but not in alcohol non-addicted rats. These are
distinct genetic strains produced by selective breeding
over 1 6 - 1 7 generations (Eriksson, 1971). At present
there are no data available as to the serotonin content
of various brain regions of the mice consuming
relatively low concentrations of alcohol chronically.
It is therefore possible that chronic consumption of
alcohol alters the activity of serotonergic neurons in
certain regions of the CNS, which mediate the circadian rhythmicity of corticosterone.
We have previously reported that the adrenocortical response to acute stress was modified as a
result of chronic consumption of alcohol in C57BL
mice (Kakihana et al., 1971). The rise of corticosterone and the return to the pre-stress level after an
acute injection of alcohol was faster in the alcohol
adapted mice than in controls, indicating that the
responsiveness of the adrenals in the alcohol-adapted

Psychopharmacologia (Berl.), Vol. 46, Fasc. 3 (1976)

mice was intact, if not enhanced. The rate of the
corticosterone metabolism was shown to be the same
in the alcohol-adapted and control groups (Kakihana
et al., 197l). The phenomenon of faster return of the
plasma corticosterone to pre-stress level was also
observed in the alcohol-adapted male and female
DBA/2 mice (Kakihana, unpublished study). Thus,
the responsiveness of the adrenals of the alcohol
drinking mice to acute stress appeared to be intact.
The notion that the neuroendocrine mechanisms controlling the circadian variations of the ACTH and
adrenocortical steroid are dissociable from those controlling the release of ACTH in response to acute
stress, has ample experimental support (Slusher, 1964;
Zimmerman and Critchlow, 1965, 1969; Halasz et al.,
1967; Hodges, 1970; Hodges and Mitchley, 1970).
It appears therefore that chronic alcohol consumption
is affecting those sites in the CNS involved in the
circadian control of the hypothalamo-pituitaryadrenal system, possibly via its effect on CRF activity
in the median eminence.
The diurnal variations of food and liquid intake
were still present after 10 weeks of alcohol consumption, although the absolute levels of intake were
different during certain periods of the day. In addition
to light which has been considered as a major factor
in entraining the adrenocortical rhythmicity, restriction of food and water alters the normal circadian
adrenocortical function in rats (Johnson and Levine,
1973; Krieger, 1974). For example, Krieger (1974)
demonstrated that restriction of food and water for
2 h in the morning shifted the normal circadian peak
of adrenocorticosteroid from late afternoon to early
morning without affecting the overall levels of the
hormone. Since there was no restriction on the time
and amount of food and liquid intake imposed in
the present study, the difference in total food and
liquid intake seen in alcohol and water groups per se
is assumed not to be a major factor for the observed
disappearance of the endocrine rhythmicity.

Acknowledgment. This work was supported by a Grant from

the National Institute of Alcoholism and Alcohol Abuse
(AA 00498). We thank Dr. John Butte of NASA for reading
the manuscript and Husein Halilovic and Paul Roesler for
their expert technical assistance. We also thank Dr. Dora
B. Goldstein of Stanford University School of Medicine for
the use of her laboratory facilitiesfor blood alcohol assay.

Ahtee, L., Eriksson, K. : 5-hydroxytryptamineand 5-hydroxyindolylacetic acid content in brain of rat strains selected
for their alcohol intake. Physiol.Behav. 8, 123 - 126 (1972)
Blum, K., Calhoun, W., Wallace,J. E. : Soporific action of
ethanol in mice: Possible role of biogenic amines. Pharmacol. Biochem.Behav. 1,271-276 (1973)

R. Kakihana and J. A. Moore: Alcohol and Adrenocortical Activity

Butte, J. C., Kakihana, R., Noble, E. P. : Circadian rhythm of
corticosterone levels in rat brain. J. Endocr. (in press)
Butte, J. C., Noble, E. P. : Simultaneous determination of
plasma or whole blood cortisol and corticosterone. Acta
endocr. (Kbh.) 61, 678 686 (1969)
Czaja, C., Kalant, H. : The effect of acute alcoholic intoxication on adrenal ascorbic acid and cholesterol in the rat.
Canad. J. Biochem. 39, 3 2 7 - 334 (1961)
Dixit, B. N., Buckley, J. P. : Circadian changes in brain
5-hydroxytryptamine and plasma corticosterone in the rat.
Life Sci. 6, 755-758 (1967)
Ellis, F. W. : Effect of ethanol on plasma corticosterone levels.
J. Pharmacol. exp. Ther. 153, 121-127 (1966)
Eriksson, K. : Rat strains specifically selected for their voluntary alcohol consumption. Ann. Med. exp. Fenn. 49,
6 7 - 72 (1971)
Feldstein, A. : Effect of ethanol on neurohumoral amine
metabolism. In: The biology of alcoholism, vol. 1, B.
Kissin and H. Begleiter, eds., pp. 127-159. New York:
Plenum Press 1971
Feldstein, A. : Ethanol-induced sleep in relation to serotonin
turnover to 5-hydroxyindoleacetaldehyde, 5-hydroxytryptophol, and 5-hydroxyindoleacetic acid. Ann. N. Y. Acad.
Sci. 215, 7 1 - 7 6 (1973)
Gursey, D., Olson, R. E. : Depression of serotonin and norepinephrine levels in brain stern of rabbit by ethanol.
Proc. Soc: exp. Biol. (N.Y.) 104, 280-281 (1960)
Halasz, B., Slusher, M . A . , Gorski, R. A. : Adrenocorticotrophic hormone secretion in rats after partial or total
interruption of neural afferent to the medial basal hypothalamus. Neuroendocrinology 2, 4 3 - 4 5 (1967)
Halberg, F., Peterson, R. E., Silber, R. H. : Plasma relations
of 24-hour periodicities in blood corticosterone, mitoses
in cortical adrenal parenchy and fetal body activity.
Endocrinology 64, 2 2 2 - 230 (1959)
Hays, W. L. : Statistics for psychologists. New York: Holt,
Rinehart and Winston 1963
Hodges, J. R. : The hypothalamus and pituitary ACTH release.
In: Progress in brain research, D. DeWied and J. A. W.
M. Weijnen, eds., pp. 12-20. Amsterdam: Elsevier 1970
Hodges, J. R., Mitchley, S. : The effect of betamethasone on
circadian and stress-induced pituitary-adrenocortical function in the rat. Brit. J. Pharmacol. 38, 719-724 (1970)
Hunt, W. A., Majchrowicz, E. : Turnover rates and steadystate levels brain serotonin in alcohol-dependent rats.
Brain Res. 72, 181 - 184 (1974)


Johnson, J. T., Levine, S. : Influence of water deprivation on

adrenocortical rhythms. Neuroendocrinology 11,268- 273
Kakihana, R., Butte, J. C., Hathaway, A., Noble, E. P. :
Adrenocortical response to ethanol in mice: Modification
by chronic ethanol consumption. Acta endocr. (Kbh.) 67,
653-664 (1971)
Kakihana, R., Noble, E. P., Butte, J. C. : Corticosterone
response to ethanol in inbred strains of mice. Nature
(Lond.) 218, 3 6 0 - 361 (1968)
Krieger, D. : Food and water restriction shifts corticosterone,
temperature, activity and brain amine periodicity. Endocrinology 95, 1195-1201 (1974)
Kuriyama, K., Rauscher, G. E., Sze, P. Y. : Effect of acute
and chronic administration of ethanol on the 5-hydroxytryptamine turnover and tryptophan hydroxylase activity
of the mouse brain. Brain Res. 26, 450-454 (1971)
Lundquist, F. : The determination of ethyl alcohol in blood
and tissues. Meth. Biochem. Anal. 7, 217-251 (1959)
Margraf, H . W . , Moyer, C. A., Ashford, L. E., Lavalle, L.
W. : Adrenocortical function in alcoholics. J. Surg. Res. 7,
5 5 - 62 (1967)
Palai6, Dj., Desaty, J., Albert, J. M., Panisset, J. C. : Effect
of ethanol on metabolism and subcellular distribution of
serotonin in rat brain. Brain Res. 25, 381-386 (1971)
Santisteban, G . A . , Swinyard, C. A. : The effect of ethyl
alcohol on adrenocortical activity in mice. Endocrinology
59, 391 - 397 (1956)
Scapagnini, U., Moberg, G. P., Van Loon, G. R., DeGroot,
J., Ganong, W. F. : Relation of brain 5-hydroxytryptamine
content to the diurnal variation in plasma corticosterone
in the rat. Neuroendocrinology 7, 9 0 - 9 6 (1971)
Slusher, M. A. : Effect of chronic hypothalamic lesions on
diurnal and stress corticosteroid levels. Amer. J. Physiol.
206, 1161 - 1164 (1964)
Tabakoff, B., Boggan, W. O. : Effects of ethanol on serotonin
metabolism in brain. J. Neurochem. 22, 759-764 (1974)
Zimmerman, E., Critchlow, V.: Dissociation in feedback
control of ACTH release. Compound "B" in corticoidtreated rats before and after stress. Fed. Proc. 24, 191
Zimmerman, E.I Critchlow, V. : Effect of diurnal variation
in plasma corticosterone levels on adrenocortical response
to stress. Proc. Soc. exp. Biol. (N.Y.) 125, 658-663 (1967)
Zimmerman, E., Critchlow, V. : Effects of intracerebral dexamethasone on pituitary-adrenal function in female rats.
Amer. J. Physiol. 217, 392-396 (1969)

Dr. Ryoko Kakihana, Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine
Stanford, CA 94305, U.S.A.