Nycthemeral rhythm in

adrenal responsiveness to ACTH

MARY F. DALLMAN, WILLIAM C. ENGELAND, JAMES C. ROSE,

CHARLES W. WILKINSON, JEANETTE SHINSAKO, AND FREDERICK SIEDENBURG
Department of Physiology and the Metabolic Research Unit, University of California,
San Francisco, California 94143

DALLMAN, MARY F., WILLIAM C. ENGELAND, JAMES C. of changing circulating levels of endogenously secreted
ROSE, CHARLES W. WILKINSON, JEANETTE SHINSAKO, AND ACTH (14, 15, 18).
FREDERICK SIEDENBURG. Nycthemeral rhythm in adrenal re- Against the interpretation that circadian increases
sponsiveness to ACTH. Am. 3. Physiol. 235(5): RZlO-R218, in circulating levels of ACTH increase adrenocortical
1978 or Am. J. Physiol.: Regulatory Integrative Comp. Phys- responsiveness to ACTH, are reports that plasma cor-
iol. 4(3): RZlO-R218, 1978. -Adrenal adenosine 3’,5’-cyclic
monophosphate (CAMP) and corticosterone responses to ex- ticosterone levels retain a circadian rhythm after hy-
ogenous ACTH were found to be about 2.5 times greater in pophysectomy (and the presumed elimination of cycli-
the evening (at lights off) than in the morning (at lights on) cal changes in circulating ACTH levels) in both the
in rats. The rhythm in adrenal responsiveness to ACTH was kill&h (22) and the rat (11). These results strongly
fouad to persist in rats treated with dexamethasone 15 and 3 suggest that the rhythm in adrenocortical responsive-
h before exogenous ACTH (in the presumed absence of a ness to ACTH is determined by a mechanism that does
rhythm in endogenous ACTH). Treatment with p-chlorophen- not depend upon a daily rhythm in circulating ACTH
ylalanine did not affect the daily rise in circulating ACTH levels.
levels but did abolish the rhythm in adrenal responsiveness In this study we have attempted to determine
to ACTH. The magnitude of the rhythm in adrenal respon-
siveness to ACTH is greater than the magnitude of the whether there is a nycthemeral change in the respon-
rhythm in ACTH. Because the rhythms are dissociable, we siveness of the rat adrenal gland to exogenously admin-
conclude that in vivo measurements of adrenal corticosteroid istered ACTH, and whether such a change includes
levels do not necessarily reflect ACTH levels. alterations in the adenosine 3’ ,&cyclic monophosphate
(CAMP) response. In addition, we have attempted to
ACTH rhythm; adrenal rhythm; PCPA; dexamethasone; ad- dissociate the rhythms in ACTH and adrenal respon-
renal cyclic AMP; corticosterone siveness to ACTH using either multiple treatments
with dexamethasone to inhibit the ACTH rhythm, or
treatment with p-chlorophenylalanine, a drug that has
MARKED CIRCADIAN FLUCTUATIONS in PlaSIIIa COrtiCOSte- been reported to alter the nycthemeral rhythm in
roid levels have been thoroughly documented in man, plasma and adrenal corticosterone (20, 24, 25), to in-
other primates, and rodents (1, 9, 12, 17). There is hibit the adrenal rhythm.
general agreement that under normal conditions corti-
MATERIALS AND METHODS
costeroids rise to a peak at the start of the daily activity
cycle. Rats, which are active nocturnally, exhibit peak Male or female Sprague-Dawley rats weighing lOO-
plasma corticosterone levels at, or just prior to, dark 150 g at the time of the experiment were received from
(9) Simonsen suppliers lo-14 days before use and were
in a recent study of the plasma ACTH and corticos- placed two or three to a cage in hanging wire baskets.
terone responses of the rat to stressful stimuli applied The animals were maintained in rooms with a con-
at the peak (lights off) or trough (lights on) of the trolled light cycle (12 h light:12 h dark) and had
corticosterone rhythm, we obtained evidence for a dif- continuous access to food (Berkeley Diet rat and mouse
ference in adrenal responsiveness to endogenously se- chow) and water.
creted ACTH between the two times of the day (5). Adrenal responsiveness to ACTH. When used, pen-
Evidence for changing adrenocortical responsiveness to tobarbital sodium (Nembutal, Abbott) was given ip in
exogenously administered ACTH as a function of the doses of 4.5 mg/lOO g body wt to females and 5.0 mg/lOO
time of day when ACTH was given has been found in g body wt to males; dexamethasone (Decadron, Merck
man (7, 14, 15, 18). In both man and the rat, it appears Sharpe & Dohme) was given 25 pg/lOO g body wt SC.
that the adrenal is maximally responsive to ACTH Synthetic a!1-24-ACTH (Cortrosyn, Organon) was di-
when circulating ACTH levels are at their circadian luted in 0.005 N HCl:0.9% saline and kept on ice for
maximum. In studies of man the change in adrenal each experiment. When the individual experiments
responsiveness to ACTH was ascribed to the influence were separated by 12 h, ACTH was made up twice;
R210 0363.6119/78/0000-OOOO$Ol. 25 Copyright 0 1978 the American Physiological Society
NYCTHEMERAL RHYTHM AND ACTH R211

when several groups were to be injected within 12 h, and corticosterone assays, so that there were equal
the same solution, kept cold, was used. Several proto- numbers of left and right adrenals in each assay. The
cols were used to test adrenal responsiveness to ACTH. adrenal to be used for CAMP assay was removed within
In the first experiments, female rats were anesthetized 60 s of decapitation and snap-frozen on dry ice until
with pentobarbital20 min before intravenous injections homogenized in 1.0 ml 5% trichloroacetic acid. After
of acid saline or 7.5 or 15 ng ACTH into a lateral tail four extractions with ether the supernatant was lyoph-
vein. Plasma corticosterone levels were measured 15 ilized and processed as directed for measurement with
min after the injections at lights on and off. In the the Amersham CAMP competitive protein binding kit
second experiment, female rats were treated with dex- (8). Adrenal s collected for corticosterone determination
amethasone 2 h before pentobarbital. After 20 min, were cleaned of fat and connective tissue, weighed to
acid saline, 4 or 8 ng ACTH was injected intravenously the nearest 0.01 mg and homogenized 1:500 (wt/vol) in
and the rats were killed 15 min later for determination 20% EtOH:O.9% saline (vol/vol). Plasma and adrenal
of adrenal and plasma corticosterone concentrations. corticosterone concentrations were measured by com-
Other, untreated rats were decapitated at the same petitive protein binding (13) using human transcortin,
times of day (6 times during a 24-h period) to determine and plasma ACTH was determined by radioimmunoas-
resting plasma ACTH and corticosterone levels. In the say (4, 19). All ACTH and corticosterone samples from
third experiment, female rats were again treated with a single experiment were distributed in the assays so
dexamethasone and pentobarbital and were injected that groups were run across assays and that represent-
with acid saline, 2 or 4 ng ACTH. Plasma and adrenal ative samples from each group were run within each
corticosterone levels were determined 5 and 10 min assay. The coefficient of interassay variation is 14% for
after the injections at lights on or at lights off. In the the ACTH assay and 10% for the corticosterone assay.
fourth experiment, male rats were treated with dexa- Data were analysed using one-, two-, or three-way
methasone and pentobarbital and injected with 2 or 4 analysis of variance (ANOVA), for either time of day,
ng ACTH intravenously. Groups of rats were killed at time of day vs. dose of ACTH or vs. treatment, time of
zero time (before) and 3, 7, or 15 min after ACTH at day vs. dose of ACTH vs. time after ACTH, or time of
lights on and lights off. Plasma corticosterone, and day vs. dose of ACTH vs. treatment.
adrenal corticosterone and CAMP concentrations were
determined. In the fifth experiment, male rats were
RESULTS
pretreated either once at -3 h or twice at - 15 and -3
h with dexamethasone. All rats were anesthetized with Effect of time-of-day on adrenal response to ACTH in
pentobarbital and injected intravenously at zero time pentobarbital-anesthetized female rats without or with
with acid saline, 1, 2, or 4 ng ACTH. Groups of rats dexamethasone pretreatment. In the first experiment,
were killed at 3 and 15 min after injections at lights on female rats (n = G/group) were given pentobarbital 20
and lights off. In the sixth, seventh, and eighth experi- min before injecting acid saline, 7.5.or 15 ng ACTH iv
ments, male rats were treated at -72 and -48 h with at the time of lights on, or 12 h later at lights off.
either p-chlorophenylalanine methyl ester (PCPA) 32 Pentobarbital anesthesia was used because it has been
mg/lOO g body wt or 0.9% saline (0.5 ml/rat). Some rats reported to decrease plasma corticosterone levels in the
were treated at -3 h with dexamethasone. At zero evening to levels observed in the morning (27). Com-
time, either at lights on or at lights off, rats not treated parison of the corticosterone levels in saline-treated
with dexamethasone were decapitated for measure- rats in the morning and the evening using an unpaired
ment of circulating ACTH and corticosterone levels t test revealed that there was no difference in these
and adrenal corticosterone concentration under resting levels (t = 1.73, P > 0.2). There were significant effects
conditions. The dexamethasone-blocked rats were in- of treatment with ACTH (F = 45.07, P < O.OOl), and of
jected with saline, 50 or 100 ng ACTHI g body wt ip. time of day (F = 90.25, P < 0.001) on plasma corticos-
Plasma and adrenal corticosterone concentrations were terone levels obtained 15 min after these treatments
measured 15 min after ACTH at lights on and lights (Fig. 1). However, there were no differences between
off. corticosterone responses to 7.5 or 15 ng ACTH, either
A final, nonexperiment was performed in which rats in the morning or in the evening, suggesting that both
were decapitated without prior treatment at lights on doses of ACTH had produced a maximal adrenal re-
or lights off and plasma ACTH and corticosterone and sponse 15 min later.
adrenal corticosterone were determined. These resting In the second experiment, groups of six rats were
levels were collected from not more than 8 rats/time (16 decapitated without prior treatment at 4-h intervals
rats/day) over a period of 6 mo. during the 24-h day and plasma ACTH and corticoster-
All rats were killed by decapitation and trunk blood one, and adrenal corticosterone levels were measured.
was collected into iced, heparinized plastic tubes. Blood Other groups of 14 rats were treated with dexametha-
was centrifuged at 4”C, and plasma was separated and sone 2 h before death, and 20 min after pentobarbital
frozen. Left adrenal glands were removed and placed these animals were injected intravenously with saline
on filter paper saturated with 0.9% NaCl in dishes kept (n = 2), 4 ng (n = 6), or 8 ng (n = 6) ACTH. The rats
on ice. In the experiment in which adrenal CAMP were killed 15 min after the injections at the same time
content was determined, left and right adrenals from as the untreated groups and plasma and adrenal corti-
alternate rats were assigned to the cyclic nucleotide costerone levels were measured. Figure 2 shows that
R212 DALLMAN ET AL.

the lowest plasma ACTH and corticosterone levels in
resting rats and those stimulated with ACTH were
observed at lights on (0600 h), and that peak resting
levels and responsiveness to ACTH were observed 12 h
later at lights off (1800 h). One-way ANOVA revealed
significant time-of-day effects for all parameters mea-
sured except for plasma< ACTH (resting ACTH: F =
2.09, P > 0.1; resting adrenal B: F = 7.14,P c 0.001;
resting plasma B: F = 9.52, P < 0.001; after 4 ng ACTH:
adrenal B: F = 3.00, P < 0.05 (data not shown); plasma
B: p = 14.09, P c 0.001). The results after 8 ng ACTH
were similar to those after 4 ng (data not shown); it
appeared that 4 ng ACTH was a dose that was adequate
to saturate the adrenal corticosterone response 15 min
later. These results show clearly that adrenal respon-
siveness to ACTH is lowest at lights on and highest at
lights off. Additionally, because the effect was observed
at the adrenal gland as well as in the plasma, the effect
is not simply a consequence of altered distribution,
binding, or metabolism (DBM) of corticosterone.
NS
Corticosterone 30
(yg/ IO0 ml plasma
ot I5 min) 2.
0
IV Injection Acid sol ine
FIG. 1. Plasma corticosterone responses 15 min after acid saline,
7.5 or 15 ng ACTH administered iv to pentobarbital-anesthetized
female rats in morning or evening, n = G/group. Verticd Lines
In the third experiment, rats were treated with
dexamethasone 3 h before pentobarbital anesthesia and
injection of saline, 2 or 4 ng ACTH iv @/group) at lights
on or at lights off. The rats were killed 5 or 10 min after
the injections and plasma and adrenal corticosterone
levels were measured. The results are shown in Fig. 3.
An additional group of rats was killed at zero time to
determine initial levels. The injection of saline did not
change adrenal or plasma corticosterone levels at 5 or
10 min compared to the zero time controls (F = 1.31,P
> 0.7). In contrast, there were main effects of both
ACTH and time of day on both adrenal and plasma
corticosterone levels (adrenal: F = 21.71,P c 0.001;F
= 23.15, P < 0.001, respectively; and plasma: F = 36.90,
P < 0.001;F = 22.36, P c 0.001, respectively). There
was, however only a suggestion of an increased effect
of the 4-ng dose above that observed after 2-ng dose.
Nycthemeral difference in the adrenal CAMP re-
sponse to ACTH in male rats. To determine whether
the daily change in adren .a1 responsiveness to ACTH
occu rred before or after the step of activation of adrenal
adenylate cyclase, male rats were treated in the morn-
cl A.M.
ing or the evening with dexamethasone 2.5 h before
.:.:.:.:.:.:*:..
:::>::::::::::: pentobarbital aneethesia and injection with 2 or 4 ng
izgzzg
.... P. M .
0.*.*.*.* ACTH @/group). Adrenal CAMP and corticosterone
p<O.Ol p<O.OOl levels and plasma corticosterone levels were measured
3, 7, and 15 min later (Fig. 4). The results of the three-
way ANOVA (dose of ACTH vs. time after ACTH vs.
time of day) show that there is a highly significant
effect of all three variables on adrenal CAMP and
plasma corticosterone levels and of dose of ACTH and
time of day on adrenal corticosterone levels (Table 1).
The time course of all three responses changes signifi-
cantly between morning and evening with a more
prolonged response accompanying the greater ampli-
tude of the response in the evening (time-time of day
interactions, Table i).
Nycthemeral difference in adrenal responses to
ACTH ACTH ACTH in male rats after single or multiple treatments
7.5 ng I5 ng with dexamethasone. Rats were treated with dexameth-
asone 3 h before pentobarbital anesthesia and were
injected with saline, 1, 2, or 4 ng ACTH iv (n = 81
group). Adrenal and plasma corticosterone levels were

t
represent *SE in this and succeeding figures. determined in the saline-treated group at 3 min and in

/t \ w!
60 -PO
40 20 Plosmo
Plosmo ACTH corticostorone
(fWml) 20 IO (JJgNxhnl)
/
L I I I I 1
0600 IO00 1400 1800 2200 0200
Pbsmo 40
cofticostorono
(JJg/lood) 20
IS min. oftw
4 ng ACfH LV. 0
limo of doy Mours)
FIG. 2. Top: plasma ACTH and corticosterone concentrations in pentobarbital-anesthetized female rats, n = G/group. Open area
resting female rats at 6 times a day (lights on 0600 h, lights off 1800 bounded by dashed Line at base of each bar represents mean plasma
h n = G/group). Bottom: plasma corticosterone levels 15 min after 4 corticosterone levels of 2 similarly treated rats injected with acid
ng ACTH iv at 6 times during day in dexamethasone-treated, saline.
NY ‘CTHEMERAL RHYTHM AND ACTH R213
ADRENAL AND PLASMA CORTICOSTERONE RES PONSES
TO ACTH ARE GREATER IN THE EVENING
-
4
Adrenal
corticosterone 2
0 ACTH at0600h FIG. 3. Adrenal and plasma cor-
( yg/lOO mg)
ticosterone levels obtained 5 or 10
n t m ACTH gt 1800h
min after injection of saline (dotted
Base of bat = Saline controls portion of bars), 2 or 4 ng ACTH at
lights on (0600 h) or lights off (1800
20 h). Female rats @/group) were pre-
Plasma treated with dexamethasone and
corticosterone IO pentobarbital before iv injections.

(jJg/lOOml)
0
lime of ter injection
5 IO 5 IO
(min)
Dose of ACTH 2 w 4 ng

A----A AM (OSOOh)
the ACTH-treated groups at 3 and 15 min. The adrenal o-0 PM ( 1800 h)
corticosterone response-to ACTH was similar to earlier
experiments in that ACTH given in the evening was ( pMoles//adrenol)
significantly more effective than the same dose given
in the morning (Fig. 5) (time of day: F = 339.47, P c
0.001). Plasma corticosterone levels 3 min after ACTH
were not different from saline-treated controls either in
the morning or in the evening. In agreement with the Adrenal corticosterone 8
adrenal measurements, the two lower doses of ACTH OJgAOO mg)
did not result in increases of plasma corticosterone
above control levels at 15 min in the morning; the 4-ng
dose of ACTH resulted in plasma corticosterone levels
of 6.8 t 0.4 kg/100 ml (mean t SE) at 15 min,
significantly above the control level of 51.5 pg/lOO ml. Plasma cortkosterone 2O
I
Fif%een minutes after 1,2, and 4 ng ACTH given in the
evening plasma corticosterone levels were 4.3 t 0.8, 7.0
t 1.3, and 24.3 t 1.2 pg/lOO ml, respectively. I I
To test whether changing levels of ACTH during the 0 3 7 I5
preceding 12 h were necessary for the altered adrenal 4 Time after injection (min.)
2 ng ACTH
responsiveness to ACTH observed between morning 0 300 r
and evening, additional groups were injected twice
with dexamethasone 15 and 3 h before pentobarbital CAMP
200
treatment and injection. with 2 ng ACTH. There were (p MblesIadrenaI) I o.
significant main effects of time of day and of dexameth-
asone treatments when the adrenal and plasma corti- 0
costerone responses to 2 ng ACTH were compared in
rats treated with dexamethasone either once or twice
(Fig. 5) (adrenal: F = 112.24, P < 0.001, and F = 28.06, Adrenal car ticosterone 8
P c 0.001, respectively; and plasma: F = 31.92, P c (yg/lOOmg)
4
0.001 and F = 7.01, P < 0.01, respectively). Adrenal
corticosterone at 3 min and plasma corticosterone at 15
min are shown in Fig. 6. The adrenal corticosterone 30
response to 2 ng ACTH at 3 min was significantly
./I
greater in the evening than in the morning following Plasma corticosterona 2O A- --c--
15-h pretreatment with dexamethasone; there was no / /
(flg/lOOml) / 0
difference between groups treated at -3 or at - 15 and l
/ 0
0
_
-3 h with dexamethasone. However, the 15.min re- IO/, --
-A’
I I
0
sponse in adrenal (Fig. 5) and plasma (Fig. 6) corticos- 0 3 7 IS
terone was affected by longer duration dexamethasone 4 Time after injection (min .)
4 ng ACTH
treatment. FIG. 4. Adrenal CAMP and adrenal and plasma corticosterone
Effect of treatment with PCPA on plasma ACTH and responses to 2 ng (A) and 4 ng (B) ACTH given at 0600 h (lights on)
plasma and adrenal corticosterone rhythms. The re- and 1800 h (lights off). Male rats @/group) were pretreated with
sults of two similar experiments were pooled and are dexamethasone and pentobarbital before iv injection of ACTH.
R214 DALLMAN ET AL.

TABLE 1. ANOVA for’results shown in Fig. 4

Adrenal
Plasma Corticosterone
CAMP Corticosterone

Dose of ACTH F (1,79) = 41.67; P < 0.001 F (1,81) = 21.83; P < 0.001 F (1,81) = 20.24; P < 0.001
Time after ACTH F (2,79) = 21.02; P < 0.001 F (2,81) = 4.49; P < 0.001 F (2,81) = 118.17; P < 0.001
Time of day F (1,79) = 71.21; P < 0.001 F (1,81) = 30.39; P < 0.001 F (1,81) = 42.46; P c 0.001
Dose-time F (2,79) = 7.25; P c 0.005 F (2,81) = 3.39; P < 0.05 F (2,81) = 7.27; P < 0.005
Time-time of day F (2,79) = 3.25; P < 0.05 F (2,81) = 20.08; P c 0.001 F (2,81) = 5.26; P < 0.005

Circadian change in adrenal sensitivity to ACTH similar excursions occurred in plasma ACTH levels in
both saline- and PCPA-treated rats, the normal rhythm
in adrenal responsiveness to ACTH is abolished in rats
treated with PCPA.
Plasma ACTH and adrenal corticosterone levels ob-
tained from resting rats in the morning or evening.
5
Adrenal l 1800 h
Plasma ACTH and adrenal corticosterone levels were
car ticosterone 0 0600 h determined at lights on or at lights off in 152 rats under
4
concentration I t- t/ I
A I5 hr Dex 1800h resting conditions. The results are presented as a
fyg/lOO mg) 3
t I
A I5 hr Dex 0600 h histogram with mean adrenal corticosterone levels
plotted against lo-20 pg/ml increments in ACTH con-
2 / centration (Fig. 9). For the same range of plasma
+ 1' 44- ACTH levels (between 0 and 49 pg/ml) adrenal corticos-
I 0 Q---- 0' 4 '4 terone levels in the morning were significantly lower
0 I 1 I I I than in the evening (P < 0.001). Individual compari-
0 I 2 4 1 2 4 sons between mean morning and evening adrenal cor-
ng ACTH (IV) at’3mln ng ACTH (IV) at 15min ticosterone levels were made for the ranges of ACTH O-
FIG. 5. Adrenal corticosterone 3 min (Zefi) and 15 min (right) 19, 20-39, and 40-49 pg/ml; Student’s t values are
after ACTH given iv to male rats at lights on (0600 h) or lights off indicated at the top of Fig. 9.
(1800 h). Cirdes represent mean values obtained from rats treated
once with dexamethasone 3 h before injection. Triangles represent DISCUSSION
mean values obtained from rats injected twice with dexamethasone
15 and 3 h before ACTH, n = 8/group. The results of these studies provide unequivocal
evidence that adrenal responsiveness to ACTH changes
shown in Figs. 7 and 8. Resting ACTH levels at lights between morning and evening in the rat as it is
on and lights off were not affected by treatment with reported to in man (7, 14, 15, 18). The change in
PCPA although there was a significant effect of time of responsiveness observed between lights on and lights
day (F = 2.24, P > 0.1, and F = 6.68, P = 0.011, off was initially inferred by us to occur from a study of
respectively). In contrast, adrenal and plasma corticos- the nycthemeral responses of the unanesthetized rat to
terone levels were significantly affected both by PCPA ACTH-releasing stimuli. In that study we found that
treatment and by time of day (adrenal: F = 4.93, P = the plasma ACTH response to histamine injected intra-
0.028, and F = 49.91, P < 0.001respectively; plasma: F peritoneally was greater in the morning whereas the
= 27.96, P c 0.001 and F = 60.96, P c 0.001respec- plasma corticosterone response in the same rats was
tively). In agreement with the results from control rats probably greater in the evening. Moreover, although
shown in Fig. 7, 15 min after 50 or 100 ng ACTH the basal circulating ACTH levels were only found to
usual time-of-day effect was observed in saline-treated double between morning and evening, plasma corticos-
rats, but no time-of-day effect was noted in the adrenal terone levels increased ninefold (5). Because plasma
response to ACTH in PCPA-treated animals (Fig. 8). corticosterone levels represent the net result of adrenal
ACTH-dose: F = 32.04, P < 0.001; PCPA: F = 5.09, P < corticosterone secretion rates and corticosterone distri-
0.05; time of day: F = 4.22, P < 0.05). bution-binding and metabolism rates, it was possible
In a third experiment, treatment with PCPA ap- in that study (in which only plasma corticosterone
peared to be ineffective. Rats were treated with dexa- levels were measured) that a marked nycthemeral
methasone, and saline or 10, 50, or 100 ng ACTH were decrease in corticosterone clearance rate and/or distri-
injected ip 3 h later at lights on and lights ofl, animals bution volume might have explained the apparent
were killed 15 min after the injection. ANOVA revealed increase in adrenal responsiveness to ACTH. Yasuda
no significant effect of PCPA treatment on adrenal et al. (27) have also reported a greater ACTH response
corticosterone (F = 3.38, P = 0.065), however, the to ether in the morning than in the evening. However,
usual effects of dose of ACTH (F = 21.17,P < 0.001) these authors concluded from changes in plasma corti-
and of time of day (F = 4.30,P = 0.038) were observed. costerone levels (from different initial levels) that the
In Table 2 the results of the first two experiments with adrenal response to ACTH was greater in the morning
PCPA are expressed as a ratio of the mean levels than in the evening.
obtained in the evening divided by those in the morn- In the present study, both adrenal and plasma corti-
ing. Examination of Table 2 reveals that although costerone responses to ACTH administered at different
 NYCTHEMERAL RHYTHM AND ACTH R215

3
Adrenal 8 content t
(pg/100 mg)
3min *
after ACTH (IV)

FIG. 6. Adrenal and plasma corticosterone levels

after 2 ng ACTH in rats treated with dexamethasone 3
8
- cl Dex 25ygAOOgb3 h) h (open bars) of 15 and 3 h (stippled bar) before ACTH
injection, n = 8/group.

1 Is!
r
6
Dex 25ygAOOij t-15 & -3 h)
Plasma 8 content
(flg/lOO ml) 4
15 min
after ACTH (IV) 2
Lower limit
of assay
0
0600 h 1800 h

Basal ACTH and corticosterone levels more, mean adrenal corticosterone concentration was
in saline 0 and PCPA lZl treated rats approximately 2.6 times higher in the evening than in
the morning in untreated resting rats whose plasma
-
ACTH levels ranged between 0 and 49 pglml (Fig. 9).
50 Thus, the nycthemeral rhythm in adrenal responsive-
ness to ACTH appears to be independent of these
PI asma ACTti experimental conditions.
(P4/ ml) 25 Not only were the adrenal as well as plasma corticos-
terone responses to ACTH increased at lights off com-
pared to lights on, but the magnitude and time course
12 of the adrenal CAMP response to ACTH was increased
and prolonged at lights off (Fig. 4, compare response to
4 ng ACTH at lights on with that to 2 ng ACTH at
Adrenal 6 8t - lights off). Because the nycthemeral change in adrenal
tyg/lOO mg) responsiveness to ACTH is evident at the level of the
cAMP response, the data suggest that the altered
responsiveness may involve either the number or afin-
ity of ACTH receptors on the adrenal cortical cell
membrane or the transduction step between receptor
occupancy and activation of adenylate cyclase. Because
the levels of CAMP are elevated for the entire 15-min
observation period at lights off, whereas they have
returned to base line by 15 min at lights on, it is
possible that the nycthemeral change in adrenal re-
sponsiveness to ACTH also involves changes in phos-
phodiesterase activity or in the kinetics of the reaction
between ACTH and its receptor.
Lights on Beyond the 2.5.fold amplification of the ACTH
FIG. 7. Plasma ACTH and adrenal and plasma corticosterone rhythm by the nycthemeral increase in adrenal respon-
levels at lights on and lights off in male rats treated with saline or siveness to ACTH, our results suggest that further
PCPA 72 and 48 h earlier. Number of animals in each group is
indicated at base of bars. amplification of the plasma corticosterone rhythm may
be effected by DBM elements of the adrenocortical
system. Low adrenal corticosterone levels at lights on
times of the day were measured. In every experiment, are usually reflected by low to undetectable plasma
the rat adrenal was found to be at least 2.5 times more corticosterone concentration, and high adrenal corticos-
responsive to ACTH at lights off than at lights on. This terone levels at lights off are usually reflected by much
was true when exogenously administered ACTH was higher plasma corticosterone levels. Because. plasma
given to rats treated a) only with pentobarbital (Fig. corticosterone levels ranged between ~1.5 and approx-
l), b) only with dexamethasone (Fig. 8), and c) with imately 30 pg/lOO ml, it is likely that corticosterone
dexamethasone and pentobarbital (Figs. Z-6). Further- was bound primarily to transcortin and was held within
R216 DALLMAN ET AL.

16 Saline treated PCPA treated

Adrenal corticosterone
(jJg/lOOmg at 15min)
12

8 t
-Ix FIG. 8. Adrenal corticosterone levels 15 min after
ACTH injected ip in morning or evening. Results in
saline-treated rats are shown on left, those from

L
0 At lights on PCPA-treated rats are shown on right.
m At lights off 4
14 t4
100 0 50 100
ng ACTH IP

TABLE 2. Results of first two experiments
with PCPA
Treatment PCPA Saline
Plasma ACTH 2.13 1.61
Adrenal corticosterone 1.69 4.55
Plasma corticosterone 1.75 10.00
Increase in adrenal corticosterone after
50 ng ACTH 1.12 4.55
100 na ACTH 1.06 1.92
Treatment with PCPA inhibits the normal rhythm in adrenal
responsiveness te ACTH without markedly affecting the rhythm in
ACTH. (The values are mean PM levels/mean AM levels, data from
Figs. 7 and 8.)
the plasma compartment (10). It is unlikely, therefore,
that the distribution volume changed markedly be-
tween morning and evening, since transcortin levels
have been reported to remain constant throughout the
day (10). Assuming that the system is linear over the
ranges examined and that adrenal corticosterone con-
centration is a direct reflection of secretion rate, a 50%
reduction in the rate of corticosterone metabolism be-
tween morning and evening would account for our
observations that overall, ACTH doubles between
morning and evening, adrenal responsiveness to ACTH
increases by a factor of 2.5, and plasma corticosterone
concentration increases 9- to lo-fold (e.g., see Table 2;
Ref. 5).
From these experiments, it seems to us unlikely for
several reasons that the increased levels of ACTH that
are found just before or at lights out are responsible for
the increased adrenal responsiveness to ACTH in the
evening. First, the relative increase in responsiveness
persisted when rats were treated twice at -15 and -3
h with dexamethasone, although the adrenals of these
rats were less responsive to ACTH than those of rats
treated only once with dexamethasone (Figs. 5 and 6).
Treatment with dexamethasone 15 h before injection
with ACTH probably effectively inhibited ACTH secre-
tion for at least 12 h, thus the normal increase in
circulating ACTH levels that occurs during the lights-
on hours was prevented in the rats injected with ACTH
at lights off. In contrast, the rats injected with ACTH
at lights on had experienced most of the previous day’s
increase in ACTH levels before dexamethasone treat-
ment (see Fig. 2). We suspect that prolongation of the
treatment with dexamethasone for @iods longer than
15 h would result in such profound adrenocortical
atrophy and unresponsiveness (6) that testing differen-
tial adrenal responsiveness under these conditions
would not be useful.
Second, in two of three experiments, treatment of
rats with PCPA appeared to abolish the rhythm in
adrenal responsiveness to ACTH without affecting the
rhythm in plasma ACTH (Table 2). Plasma ACTH was
twice as high in the evening as in the morning in both
saline- and PCPA-treated rats. Adrenal corticosterone
concentration was also twice as high in the evening as
in the morning in PCPA-treated rats, and the response
to ACTH was the same at both times of day. In
contrast, in saline-treated rats, adrenal corticosterone
concentration was 4.5 times greater in the evening as
was the adrenal response to 50 ng ACTH.
We have no certain explanation for the failure of
PCPA to affect the adrenocortical system in the third
experiment. In that experiment, failure of PCPA treat-
ment was strongly suspected before the corticosterone
results were obtained. In two experiments, rats treated
with PCPA were noted to be unusually jumpy during
the daylight hours and to have gained less weight
during the 48 h after initiation of treatment than the
saline-injected controls. Furthermore, in the first two
experiments, adrenal weight of PCPA-treated rats was
significantly greater than that of saline-treated con-
trols at 72 h. Similar effects of PCPA have also been
noted by Van Delft et al. (24). In the third experiment,
none of these differences was noted and, from the
apparent lack of effect of PCPA treatment on these
variables as well as on adrenal function, it seems
possible to us that the lot of PCPA used was impotent.
The present study demonstrates a rhythm in adrenal
responsiveness to ACTH that is dissociable from the
rhythm in ACTH. We have presented preliminary
results elsewhere showing that rhythms in adrenal and
plasma corticosterone levels may persist in the absence
of a rhythm in ACTH levels either 24-48 h after
changing the light cycle by 180” (21), or by limiting
access to food to a period of 2 h/day either at lights on
or lights off (26). Although rhythmic secretion of corti-
costerone in vitro has been reported to occur from
adrenals in organ culture (2, 3, 23) but not from
monolayers of primary cultures of adrenal cells (16), it
seems necessary to postulate the existence of another
factor (or factors) that changes in response to environ-
mental cues and that drives the rhythm in adrenal
responsiveness to ACTH.
The magnitude of the rhythm in adrenal responsive-
ness to ACTH is large, and the results of the present
NYCTHEMERAL RHYTHM AND ACTH R217
,- 3.17 6.26 3.26
(29) (66) (211
IOr
FIG. 9. Adrenal corticosterone levels from resting male
rats killed at lights on or lights off grouped in accordance
to level of plasma ACTH found in same rat. Numbers of
Adrenal 6 rats are indicated at base of each bar. Mean ACTH levels
corticastuonr (beneath each pair of bars) did not differ between lights on
(open bars) and lights off (closed bars) for any range of
Qg/lOO mg) 4 ACTH. However, daily elevation of plasma ACTH at lights
out is shown by greater numbers of rats with high ACTH
2 levels at this time as well as by higher overall mean level
of ACTH. Numbers above 3 sets of bars on the left are t
0 values with numbers of rats compared shown parentheti-
Plasma ACTH (pg/ml) - O-19 20-39 40-49 50-59 60-79 cally. Only in first 3 groups were there sufficient numbers
4
;;;i iiF SiT; ii 44 of rats found at lights on and lights off to make meaningful
ACTH moan co* t-w comparisons.
I58 ?“, InY) rt
- -kTH At Lights Adrrnal B
28f2 Ia 2.86fO.25
Overall msans
44f2 fa 7.43 f0.42

experiments demonstrate that measurement of adrenal Career Development Award AM-00072. C. W. Wilkinson is sup-
corticosteroid production in vivo is not a reliable reflec- ported by National Institutes of Health postdoctoral fellowship AM-
05681.
tion of ACTH levels except under very strictly con- Present address of W. C. Engeland: Dept. of Biomedical Engineer-
trolled conditions such as those- in bioassays for ACTH. ing, John Hopkins University, School of Medicine, Baltimore, MD
21205.
This research was supported by Public Health Service Grants Present address of J. C. Rose: Dept. of Physiology, Bowman Gray
AM-06704 and GM-00927 and National Aeronautics and Space Avia- School of Medicine, Winston Salem, NC 27103.
tion University Interchange Agreement NCA2-OR665602. M. F.
Dallman is supported by National Institutes of Health Research Received 9 January 1978; accepted in final form 8 May 1978.

REFERENCES
1. ALLEN, C. F., J. W. KENDALL, AND M. A. GREER. The effect of 13. MURPHY, B. E. P. Some studies on the protein binding of steroids
surgical isolation of the basal hypothalamus on the nycthemeral and their application to the routine micro and ultra micro
rhythm of plasma corticosterone concentration in rats with measurement of various steroids in body fluid by competitive
heterotopic pituitaries. Endocrinology 91: 873-876, 1972. protein-binding radioassay. J. Clin. Endocrinol. 27: 973-990,
2. ANDREWS, R. V., AND G. E. FOLK. Circadian metabolic patterns 1967.
in cultured hamster adrenal glands. Camp. &o&em. Physiol. 14. NUGENT, C. A., K. EIK-NES, L. T. SAMUEL~, AND F. H. TYLER.
11: 393-409,1964. Changes in plasma levels of 17-hydroxycorticosteroids during
3. ANDREWS, R. V., G. E. Fomc, AND R. HEDGE. Metabolic perio- the intravenous administration of adrenocorticotropin (ACTH).
dicity in adrenal glands cultured from arctic rodents (Abstract). J. Clin. Endocrinol. Metab. 19: 334-343, 1959.
Federation Proc. 24: 508,1964. 15. NUGENT, C. A., K. EIK-NES, H. S. KENT, L. T. SAMUELS, AND F.
4. DALLMAN, M. F., D. DE MANICOR, AND J. SHINSAKO. Diminish- H. TYLER. A possible explanation for Cushing’s Syndrome asso-
ing corticotrope capacity to release ACTH during sustained ciated with adrenal hvnernlasia. J. Clin. Endocrinol. Metab. 20:
stimulation: the twenty-four hours after bilateral adrenalectomy 1259-1268, 1960. ”- -
in the rat. Endocrinology 95: 65-73, 1974. 16. O’HARE, M. J., AND P. J. HORNSBY. Absence of a circadian
5. ENGELAND, W. C., J. SHINSAKO, C. M. WINGET, J. VERNIKOS- rhythm of corticosterone secretion in mono layer cultures of
DANE~IS, AND M. F. DALLMAN.
Circadian patterns of stress- adult rat adrenocortical cells. Eqerientia 31: 378-380, 1975.
induced ACTH secretion are modified by corticosterone re- 17. ORTH, D. N., D. P. ISLAND, AND G. W. LIDDLJZ. Experimental
sponses. Endocrinology 100: 138-147, 1977. alteration of the circadian rhythm in plasma cortisol (17-OHCS)
6. ENGELAND, W. C., J. SHINSAKO, AND M. F. DALLMAN. Corticoste- concentration in man. J. Clin. Endocrinol. Metab. 27: 549-555,
roids and ACTH are not required for compensatory adrenal 1967.
growth. Am. J. Physiol. 229: 1461-1464, 1975. 18. PERKOFF, G. T., K. EIK-NES, C. A. NUGENT, H. L. FRED, R. A.
7. FORBHAM, P. H., V. DI RAIMOND~, D. ISLAND, A. P. RINFRET, NIMER, L. RUSH, L. T. SAMUELB, AND F. H. TYLER. Studies of
AND R. H. Oaa. Dynamics of adrenal function in man. Ciba the diurnal variation of plasma 17-hydroxycorticosteroids in
Found. Colloa. Endocrinol. 8: 279-302.1955. man. J. Clin. Enokxrinol. Metab. 19: 432-443. 1959.
8. GILMAN, A. d. A protein binding assay for adenosine 3’:5’ cyclic 19. REES, L. H., D. M. COOK, J. W. KENDALL, C. F. ALLEN, R. M.
mononhosnhate. Proc. Natl. Acad. Sci. US 67: 305-312. 1970. KRAMER, J. G. ~WLI~, AND R. A. KNIGHT. A radioimmunoas-
9. GIJI~EMI& R., W. E. DEAR, AND R. A. LIEBELT. Nycthemeral say for rat nlasma ACTH. Endocrinologv 89: 254-261. 1971.
variations in plasma free corticosteroid levels of the rat. Proc. 20. S&PAGNIN~, U., G. P. MOBERG, G. R.-VAN LOON, J: DE GROOT,
Sot. Exvtl. Biol. Med. 101: 394-395, 1959. AND W. F. GANONG. Relation of brain 5-hydroxytryptamine
10. KELLJ&, N., L. R. SENDELBECK, U. I. RICHARDSON, C. MOORE, content to diurnal variation in plasma corticosterone in the rat.
AND F. E. YATES. Protein binding of corticosteroids in undiluted Neuroendocrinology 7: 90-96,197l.
rat plasma. Endocrinology 79: 884-906, 1966. 21. SIEDENBURG, F., J. SHINSAKO, AND M. F. DALLMAN. Separation
11. MEIER, A. H. Daily variation in concentration of plasma corti- of circadian rhythms in circulating ACTH and adrenal corticos-
costeroid in hypophysectomized rats. Endocrinology 98: 1475- terone (B) by a 180” phase shift in light cycle (Abstract).
1479, 1976. Endocrinology lOOA: 272, 1977.
12. MIGEON, C. J., A. B. FRENCH, L. T. SAMUELS, AND J. 2. BOWERS. 22. SRIVASTAVA, A. K., AND A. H. MEIER. Daily variation in concen-
Plasma 17-hydroxycorticosteroid levels and leucocyte values in tration of cortisol in plasma in intact and hypophysectomized
the rhesus monkey including normal variation and the effect of gulf killitish. Science 177: 185-186, 1972.
ACTH. Am. J. Physiol. 182: 462-468, 1955. 23. UNGAR, F., AND F. HALBERG. In vitro demonstration of circadian
R218 DALLMAN ET AL.

rhythm in the in vitro response of mouse adrenal to adrenocor- Res. 39: 301-309, 1973.
ticotropic hormone. Science 137: 1058-1060, 1962. 26. WILKINSON, C. W., J. SHINSAKO, AND M. F. DALLMAN. Insulin,
24. VAN DELFT, A. M. L., J. KAPLANSKI, AND P. G. SMELIK. Circa- corticosterone and ACTH rhythm in rats fed 2 hr/day (Abstract).
dian periodicity of pituitary-adrenal function after p-chlorophen- Physiologist 20: 102, 1977.
ylalanine administration in rats. J. Endocrinol. 59: 465-474, 27. YASUDA, N., K. TAKEBE, AND M. A. GREER. Evidence of nycte-
1973. rohemeral periodicity in stress-induced pituitary-adrenal acti-
25. VERNIKOS-DANELLIS, J., P. BERGER, AND J. D. BARCHAS. Brain vation. Neuroendocrinology 21: 214-224, 1977.
serotonin and pituitary adrenal function. Recent Progr. Brain