| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Netherlands Institute for Brain Research (A.K., S.E.L.F., J.W., R.M.B.), and Department of Endocrinology and Metabolism, F5–171 (E.F., O.B.), and Department of Clinical Chemistry, Laboratory of Endocrinology (E.E.), Academic Medical Center, 1105 AZ Amsterdam; and Department of Endocrinology C4-R, Leiden University Medical Center (J.A.R.), 2300 RC Leiden, Amsterdam
Address all correspondence and requests for reprints to: Dr. A. Kalsbeek, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam, The Netherlands. E-mail: a.kalsbeek{at}nih.knaw.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Blood levels of leptin follow a diurnal rhythm in healthy lean, obese, and type II diabetic subjects (9, 10, 11, 12). The genesis of this diurnal pattern of leptin levels has not been elucidated. In fact, it is not even clear whether, like many other hormonal rhythms, the diurnal plasma leptin rhythm is under direct control of the biological clock or, conversely, whether it is a mere consequence of circadian rhythms in hormone release and/or behavior. Meal shifts entrain the diurnal rhythm of leptin secretion, suggesting that daytime feeding contributes to the nocturnal increase in plasma leptin levels (13, 14, 15, 16). This feeding effect would be in accordance with the acute effect of fasting and insulin on leptin secretion (1, 17). On the other hand, diurnal leptin release was maintained during prolonged hyperinsulinemia in normal human subjects (18), indicating that factors other than feeding behavior must also be involved. The inverse relationship between diurnal rhythms of leptin, on the one hand, and ACTH and cortisol, on the other, was apparent from the discovery of leptin (9, 19) and led to the suggestion that the circadian changes in cortisol might account for the circadian rhythm in leptin release (20, 21). However, experiments employing hydrocortisone infusions and observations in patients with Cushing’s syndrome showed persistent leptin rhythms despite constant (high or low) levels of cortisol (22, 23). Another argument against a predominant role of either endogenous hormone or feeding rhythms in the control of diurnal leptin rhythmicity is that in both rodents and humans the highest leptin levels are found at night despite the fact that these species have contrasting patterns of feeding and adrenal activity.
In mammals (including humans) the biological clock is located at the base of the anterior hypothalamus in the suprachiasmatic nuclei (SCN) (24, 25). The SCN plays a key role in the induction, synchronization, and entrainment to the environmental light/dark cycle of circadian rhythms in behavioral, metabolic, and endocrine functions (24, 26). Disappearance, however, of the diurnal leptin rhythm after ablation of the SCN does not prove a direct clock control of leptin release, because in these SCN-ablated animals other rhythmic signals (e.g. feeding and glucocorticoids) have also become arrhythmic. Therefore, to investigate separately the respective roles of the SCN, adrenal glucocorticoids, and feeding rhythms in the control of leptin rhythmicity, we determined daily plasma leptin rhythms in four experimental groups of rats, i.e. unoperated intact animals, SCN-lesioned animals, adrenalectomized animals that had received a chronic corticosterone-releasing pellet, and animals on a regular feeding schedule.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SCN lesions
For SCN lesions 36 animals, 180–200 g, anesthetized with
Hypnorm (Duphar, The Netherlands; 0.6 ml/kg, im), were mounted with
their heads in a David Kopf Stereotact (Tujunga, CA) with the toothbar
set at +5.0 mm, and sustained a bilateral lesion of the SCN
(coordinates: 1.4 mm rostral to bregma; 1.1 lateral to the midline; 8.3
mm below the brain surface) using bilateral lesion electrodes, 0.2 mm
in diameter, with temperature set at 85 C for 1 min (lesion generator,
Radionics, Burlington, MA). This temperature was found
empirically to result in lesions large enough to eliminate the SCN
bilaterally, but small enough to leave surrounding hypothalamic brain
structures, such as the paraventricular (PVN) and supraoptic (SON)
nuclei intact. A drawback of this restricted lesion size is the limited
yield of completely lesioned animals (i.e. 
30%). To
restrict the number of (false positive) animals to be operated upon,
initially the effectiveness of the lesions was checked by measuring
daily water intake. After a rest period of 2 weeks to recover from the
trauma of anesthesia and brain trauma, the daily water intake of the
animals was measured during the middle 8 h of the light period.
Only animals showing a daytime water intake of more than 30% of the
total daily intake (in intact control animals this value is typically
<10%) were assumed to have complete lesions of the SCN until the
final immunocytochemical analysis and were allowed to enter the
experiment. After the experiments, SCN lesions were checked
histologically by immunocytochemical staining of hypothalamic sections
for the presence of vasopressin (VP) and/or vasoactive intestinal
polypeptide-containing cell bodies or fibers in the SCN area. If
animals had cell bodies that stained positively for either VP or
vasoactive intestinal polypeptide at the border of the lesion, or for
immunoreactive fibers in SCN target areas such as the PVN, they were
excluded from the data analysis (27, 28, 29).
Adrenalectomy (ADX)
Before ADX (via the flank approach) animals were anesthetized
with a combination of Hypnorm and Dormicum (Roche,
Mÿdrecht, The Netherlands; 0.4 ml/kg, sc). Together with
bilateral adrenalectomy the animals received a jugular vein catheter
(see below), and 100-mg slow release corticosterone pellets
(Innovative Research of America, Sarasota Beach, FL) were
implanted sc. The corticosterone pellets provided stable plasma
concentrations of corticosterone for up to 60 days. Adrenalectomized
animals (i.e. ADX+Cort animals) had free access to fresh
water and 0.9% saline to compensate for mineral loss.
Regular feeding
For the rats entrained to the six meals per day feeding schedule
(i.e. regular-fed animals) food pellets were available in
metal food hoppers. A rat could gnaw off pieces of food through
vertical stainless steel bars. Access to food could be prevented by a
sliding door situated in front of the food hopper. Door opening and
door closing were activated by an electrical motor and controlled by a
clock. The rats were entrained to a feeding schedule of six 10-min
meals spaced equally over the day-night cycle. Food became available at
zeitgeber time (ZT) 2, ZT6, ZT10, ZT14, ZT18, and ZT22 (ZT12 being
defined as the onset of the dark period).
Door-opening time was determined empirically in previous experiments, i.e. with longer door-opening times animals would start to skip one or more of the daytime meals. Animals did not save food for consumption during the 4-h intermeal period. Adaptation was considered completed when animals had learned to consume approximately 3 g at every meal. Despite the equally distributed feeding activity, general (locomotor) activity still showed a clear day-night rhythm, with the major part of activity occurring during the dark period. During the light period animals would wake up, eat, and resume sleeping. No anticipatory activity was observed. Diurnal glucose, insulin, and corticosterone profiles as a result of this feeding pattern have been reported previously (27, 30).
Blood sampling
For undisturbed blood sampling intact, regular-fed,
SCN-lesioned, and ADX+Cort animals were provided with a permanent
silicon heart cannula (id, 0.5 mm; od, 1.0 mm) at the entrance of the
right atrium (vena cava) via an external jugular venotomy as described
previously (27). The cannula was externalized and fixed on
top of the skull with three screws and dental cement. Animals were
operated on when their body weight was more than 325 g. After the
operation a 10-day recovery period was included to allow complete
reinstatement of circadian rhythms in activity, body temperature, and
plasma corticosterone. During this period animals became accustomed to
the experimental conditions in the Plexiglass cages designed to allow
blood sampling under unrestrained conditions. All experiments were
performed in the animal’s own home cage. Tubings were threaded through
a stainless steel support spring that was attached to the dental cement
on the skull. The entire assembly was suspended from the animal by a
counterbalanced beam and did not influence the animal’s posture or
motion. Daily changes in the plasma levels of leptin and corticosterone
were assessed in three experimental sessions of 8 or 10 h, each 1
week apart, by taking a 0.6-ml blood sample every 2 h. Within one
session sampling was initiated at different ZTs for the different
animals.
Hormone assays
Blood samples were collected in heparinized tubes, placed on
ice, and centrifuged, and plasma was stored at -20 C until assay.
Plasma corticosterone was measured directly, without extraction, using
a RIA from ICN Biomedicals, Inc. (Costa Mesa, CA), with
iodinated corticosterone. From the samples 10 µl plasma were taken
and diluted in 4 ml assay buffer. The interassay coefficient of
variation for corticosterone was less than 4%, and the detection limit
was 1 ng/ml. Plasma leptin concentrations were determined by RIA
(Linco Research, Inc., St. Charles, MO) in 100 µl
plasma. The detection limit was 0.5 ng/ml, and inter- and intraassay
coefficients were 6% or less. The samples of the intact and
SCN-lesioned animals were analyzed in one assay, as were the samples of
the ADX+Cort and regular-fed animals.
RNA isolation and quantification
Experimental animals were killed at either ZT2 or at ZT14.
Material from the regular-fed groups was lost. In a follow-up
experiment, six groups of control rats (n = 8–10) were killed at
4-h intervals through a 24-h period starting at ZT2. From all animals
epididymal fat depots were removed, snap-frozen in liquid
N2, and stored at -80 C. Total RNA was purified
using TriPu; from this complementary DNA was synthesized using AMV
reverse transcriptase and random primers (Roche). Leptin
messenger RNA (mRNA) was quantified using real-time PCR in a
LightCycler (SYBRgreen format, Roche). The primers used
were 5'-gtgcctatccacaaagtccag-3' (forward) and
5'-tcagcattcagggctaaggtc-3' (reverse), giving a fragment of 441 bp. The
homologous fragment was used as a standard. All mRNA levels were
corrected for ß-actin expression measured with the LightCycler
as well, using the primer sequences
5'-gggtcagaaggattcctatg-3' (forward) and
5'-ggtctcaaacatgatctggg-3' (reverse).
Statistics
The mean ± SEM were calculated for plasma
leptin, plasma corticosterone, and adipocyte mRNA
concentrations for each sampling point. ANOVA with repeated measures
was used to test for significant effects of ZT, group, or ZT
vs. group interaction. ANOVA was followed by Student’s
t test (paired) to establish which time points differed
significantly from trough values. The diurnal rhythm of each leptin
profile was quantitatively described by cosinor analysis, as described
previously (10, 11). In short, a cosine curve with a fixed
period of 24 h (because of the 12-h light, 12-h dark schedule) was
fitted through all data, and the following chronobiological parameters
were obtained: acrophase (time of peak value), amplitude (half of the
total predictable change in a rhythm), and mesor (the average value of
a cosine curve fitted to the data; the mesor and 24-h mean leptin
concentration are equivalent). Because of a considerable individual
variation in the 24-h means of SCN-lesioned and regular-fed animals,
leptin levels were first normalized by calculating the absolute change
at different time points in relation to the 24-h mean. Due to its
nonuniform distribution leptin mRNA values were compared using the
nonparametric Kruskal-Wallis test. P < 0.05 was
considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). During the
remainder of the dark period plasma leptin concentrations slowly
decreased, but remained elevated compared with mean daytime levels
(i.e. 1.3 ± 0.2 vs. 1.9 ± 0.2 ng/ml;
light vs. dark, respectively). The diurnal trough occurred
during the first half of the light period. ANOVA showed a very
significant effect of time [F(11,66) = 9.9; P <
0.001]. Plasma corticosterone values in intact control animals varied
between 5.1 ± 0.1 (at dawn) and 109.8 ± 40.0 ng/ml (at
dusk).
|
Complete 24-h plasma leptin profiles were obtained from 10 SCN-lesioned
animals. Upon histological verification (Fig. 2) one more animal of the SCN-lesioned
group had to be discarded due to an incomplete lesion. Mean plasma
leptin concentrations in the 9 remaining SCN-lesioned animals ranged
between 2.6 ± 0.2 and 3.7 ± 0.4 ng/ml (Fig. 1A
), with
individual samples varying between 0.5 and 5.4 ng/ml. ANOVA did not
detect a significant effect of time [F(11,88) = 1.4;
P = 0.185]. The lack of a significant diurnal
fluctuation is also reflected in the absence of a day/night difference,
i.e. mean day and night values were 3.3 ± 0.3 and
2.9 ± 0.2 ng/ml, respectively. When comparing SCN-lesioned and
control animals, ANOVA indicated clear effects of group and time
x group [F(1, 14) = 22.89; P < 0.001 and
F(11,154) = 3.54; P < 0.001], confirming the
overall higher mean levels in SCN-lesioned animals and the changed
diurnal profile, respectively. Plasma corticosterone values in
SCN-lesioned animals varied between 13.7 ± 2.1 and 187.0 ±
50.7 ng/ml; the overall 24-h mean value (80.3 ± 10.8 ng/ml) was
almost twice as high as that in control animals (39.5 ± 3.4
ng/ml; P < 0.005).
|
Complete 24-h profiles of plasma leptin were obtained in six animals in
the regular-fed group. Plasma leptin levels varied between 4.0 ±
0.4 and 6.0 ± 0.6 ng/ml (Fig. 1B), with individual data varying
between 2.2–10.3 ng/ml. ANOVA detected a significant effect of time
[F(14,70) = 3.87; P < 0.001]. When comparing
the 24-h plasma leptin profiles of the regular-fed animals with those
of intact and SCN-lesioned animals, ANOVA indicated significant
differences for both comparisons [intact: F(11,121) = 2.83;
P = 0.002 and F(1,11) = 44.64; P
< 0.001; SCN lesion: F(11,143) = 2.32; P = 0.012
and F(1,13) = 17.70; P = 0.001; effects of group
vs. time and group, respectively]. Plasma corticosterone
values in the regular-fed animals varied between 5.0 ± 0.1 and
153.5 ± 46.2 ng/ml; 24-h mean values (39.5 ± 2.1 ng/ml)
were comparable to those in control animals.
Effect of (the lack of) diurnal changes in plasma corticosterone on
plasma leptin levels
In the adrenalectomized animals complete 24-h plasma
corticosterone and leptin profiles were obtained in eight of the nine
animals. Plasma corticosterone values in the ADX+Cort animals varied
between 13.8 ± 3.2 and 23.2 ± 6.0 ng/ml, with no apparent
diurnal rhythmicity. Thus, 24-h plasma corticosterone levels in
ADX+Cort animals were higher than the trough levels in control animals,
but were well below the daily peak values, as also indicated by their
lower 24-h means (17.8 ± 3.8 ng/ml). Mean plasma leptin levels
varied between 1.1 ± 0.5 and 2.3 ± 0.5 ng/ml (Fig. 1B),
with individual values varying between 0.5 and 6.1 ng/ml. In the
ADX+Cort group ANOVA detected a significant effect of time
[F(13,91) = 10.00; P < 0.001] for plasma leptin
levels. The overall plasma leptin levels in the ADX+Cort animals
differed from those in SCN-lesioned and regular-fed animals, but not
from those in the control animals [F(1,13) = 0.11;
P = 0.75]. However, ANOVA did indicate a significant
interaction of group vs. time [F(11,143) = 4.19;
P < 0.001], probably due to the earlier nocturnal
decline in plasma leptin levels in the ADX+Cort group.
Diurnal rhythmicity
All groups, except that with SCN lesions, showed trough leptin
levels during the first part of the light period and an acrophase
during the light/dark transition (Figs. 1 and 3). Cosinor analysis of the data from the
SCN-intact group confirmed a clear daily rhythm (explaining 73% of the
variation; P < 0.01), with an acrophase at ZT16.5 and
an amplitude of 42%. Cosinor analysis of the data from SCN-lesioned
animals did not produce a significant correlation (P >
0.05); introducing a daily rhythmicity could at best explain 28% of
the variation. On the other hand, cosinor analysis of the data from the
regular-fed and ADX+Cort animals produced significant fits
(P < 0.01), and the daily rhythmicity could explain
48% and 60% of the variation, respectively. In both groups the
acrophase was slightly phase-advanced compared with that in intact
animals (ZT131/2 and ZT121/2, respectively). Whereas
ADX+Cort animals also showed an approximately 40% amplitude, the
(relative) amplitude was clearly reduced in the regular-fed animals,
i.e. about 10%, due to the higher 24-h means of this group.
Expressed as absolute deviations from the 24-h means, however, all
three groups showed very similar daily variations in their plasma
leptin levels, with the difference between trough and peak values
amounting to approximately 1.3 ng/ml. Cosinor analysis of the plasma
corticosterone data of the intact animals also produced a significant
fit that explained 68% of the variation. No significant fits were
detected in the corticosterone data of the three remaining groups,
although the normal circadian corticosterone pattern could be
recognized in the data of the regular-fed group (Fig. 3).
|
; r =
0.37; P = 0.04), with 24-h mean plasma leptin levels
increasing by 2 ng/ml with every 100 g in body weight. This
positive relation was not found when the different treatment groups
were analyzed separately. Therefore, SCN-lesioned and, even more so,
regular-fed animals show higher 24-h mean plasma leptin levels compared
with control and ADX+Cort animals despite comparable body weights.
|
). Taking together
the data from the intact, ADX+Cort, and SCN-lesioned animals showed a
very significant positive relation between plasma corticosterone and
plasma leptin levels (P < 0.0001), indicating that the
differences in 24-h mean plasma corticosterone levels could explain
73% of the variation in 24-h mean plasma leptin levels. Although the
grouping of intact, ADX+Cort, and regular-fed animals resulted in a
positive relation (P = 0.03), in this case the changes
in corticosterone levels could explain only 47% of the variation in
plasma leptin levels. Analyzing the different treatment groups
separately or taking together the data from all four treatment groups
did not produce a significant linear regression.
Adipocyte leptin mRNA
Leptin expression in the epididymal fat depots of none of the
three experimental groups used for the blood-sampling experiments
(i.e. intact, ADX+CORT, and SCN-lesioned) showed a
significant light/dark difference. The mean values of the three groups
also did not differ (2.3 ± 0.4, 4.3 ± 0.9, and 3.5 ±
0.5 relative arbitrary units for intact, ADX+CORT, and
SCN-lesioned groups, respectively; P = 0.165). As
the absence of significant differences was probably due to the limited
number of time points sampled (i.e. two), we performed an
additional experiment. With more frequent sampling a diurnal difference
in leptin mRNA levels in ad libitum-fed control animals just
escaped significance (P = 0.068; Fig. 5). Excluding the two outliers at ZT22,
however, produced a very significant diurnal difference
(P = 0.003). Trough values were found during the end of
the dark period and the beginning of the light period, as also observed
previously (31). Leptin mRNA values in SCN-lesioned
animals showed less variation and were restricted to the median ranges
of control animals.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The diurnal rhythm of plasma leptin and leptin mRNA levels described in the present study corresponds with the rhythms described recently in three other reports (31, 32, 33), with a daytime trough and peak levels in the early dark period. The recent descriptions of plasma leptin rhythms in the rat clearly show the similarity between leptin rhythms in rodents and humans. The similar phasing of the leptin rhythm in humans and rodents despite their opposite, i.e. nocturnal vs. daytime, feeding patterns indicates that it is unlikely that the rhythmic features of 24-h plasma leptin levels are the sole consequence of their feeding behavior. Nevertheless, a number of experiments have indicated that changes in the daily pattern of food intake can affect plasma leptin levels (1, 13, 14, 16, 17). For instance, it was shown that a 6.5-h phase delay in the timing of the meals also induces a phase delay in the nocturnal leptin peak (14). In addition, Ahima et al. (16) showed that mice on a feeding schedule consisting of one daytime meal restricted to 4 h, exhibited an extra leptin surge during the daytime. However, during continuous enteral feeding in humans the plasma leptin rhythm remained intact (13). Together with our present results, these data indicate that in addition to food intake, other factors are involved in the diurnal regulation of leptin secretion.
Soon after the discovery of the pronounced diurnal rhythmicity of plasma leptin levels (9, 12), a number of reports indicated a stimulatory effect of glucocorticoids on leptin release (20, 21, 34, 35). In combination with the inverse relationship between plasma leptin levels and adrenal glucocorticoid activity (9, 19), the above-mentioned results led to the proposition that the diurnal rhythm in plasma leptin levels might be due to a 12-h lag in the stimulatory effect of circulating glucocorticoids on the fat cell (22). In the meantime, however, a number of observations in humans have indicated that the circadian rhythm in plasma leptin levels cannot be accounted for by the diurnal variations in circulating plasma cortisol levels (9, 22, 23). In accordance with that observation, in our experiments in rats the absence of a diurnal fluctuation in circulating corticosterone levels in the ADX+Cort animals did not prevent a clear diurnal variation in plasma leptin levels comparable to that in intact control animals. Therefore, the present experiments provide further evidence for control of the daily leptin rhythm independently of the hypothalamo-pituitary-adrenal axis. Furthermore, the persistence of diurnal leptin rhythms in patients with perinatal stalk transsection syndrome also makes it unlikely that other pituitary hormones are involved in the diurnal control of plasma leptin levels (36).
Our finding of increased mean 24-h plasma leptin levels but unchanged daily rhythmicity of plasma leptin in the regular-fed animals is very much in line with the recent observations by Boden et al. (18) during 72-h euglycemic-hyperinsulinemic clamp studies. Although these prolonged euglycemic-hyperinsulinemic clamps caused a dose-dependent increase in serum leptin levels, the circadian rhythmicity remained intact. Therefore, data from our regular-fed animals and the results of Boden et al. (18) clearly show that a rhythm in insulin release cannot be the primary cause of the diurnal leptin rhythm. As the six meals per day feeding schedule does not result in increased body weight or increased plasma insulin levels (27, 30, 37), it is not clear what causes the increased plasma leptin levels in our regular-fed group. In fact, as these animals are slightly food-deprived, a lowering of plasma leptin levels would be expected. As it has been suggested that leptin acts as a sensor of nutrient flux in adipose tissue (38, 39), we are currently testing the hypothesis that the aberrant feeding schedule causes an increased flux of nutrients through the adipocyte, resulting in increased leptin release. Indeed, it has been reported that rats on a food restriction schedule develop high rates of lipolysis in adipose tissue and liver (40).
The question then remains which SCN-generated mechanism is responsible
for the diurnal rhythm in plasma leptin levels if it is not the rhythm
in feeding behavior or in the activity of the
hypothalamo-pituitary-adrenal axis. Bearing in mind the reports that
glucose metabolism regulates insulin-mediated leptin expression and
secretion by adipocytes (38), the meal-independent
circadian rhythm of leptin could also result from daily fluctuations in
adipocyte glucose utilization via changes in insulin sensitivity
(18). Both glucocorticoids and GH are known to reduce
insulin-mediated glucose uptake, but in view of the evidence presented
previously (36) it is unlikely that the rhythmic release
of either of these hormones is responsible for a (insulin
sensitivity-mediated) leptin rhythm. A number of experiments have
indicated an important role for the autonomic nervous system in the
regulation of leptin secretion. Reduction of noradrenaline release from
sympathetic nerve endings by administration of
-methyl-p-tyrosine or 6-hydroxydopamine induces
hyperleptinemia (41, 42). Conversely, ß-adrenergic
receptor agonists decrease plasma leptin levels and leptin mRNA
expression (43, 44). This regulation of leptin secretion
by sympathetic activity resembles the regulation of melatonin release
by the pineal gland. The pronounced day/night rhythm of plasma
melatonin levels is due to the nocturnal release of noradrenaline from
the sympathetic nerve endings in the pineal gland (45).
Another appealing similarity between leptin and melatonin rhythms is
the nocturnality of its acrophase, i.e. peak levels of both
leptin and melatonin are found at night, independently of the diurnal
or nocturnal preference of a species. Both anatomical and
(electro)physiological techniques have demonstrated that the SCN
controls the sympathetic input to the pineal gland by contacting those
PVN neurons that possess descending projections to the preganglionic
neurons in the spinal cord (46, 47, 48, 49). We hypothesize that a
similar neural pathway underlies the circadian control of plasma leptin
levels. This idea is fully supported by the observation that retrograde
virus tracing from the white adipose tissue labels neurons in spinal
cord, PVN, and SCN (50), and the fact that insulin
sensitivity can be changed via the autonomic nervous system
(51).
Finally, SCN-lesioned animals showed increased 24-h mean plasma leptin levels compared with intact control animals. In view of the restricted size of our lesions we are confident that this increase is not due to collateral damage to other brain areas, such as the paraventricular, ventromedial, or arcuate nuclei (52, 53). Moreover, in these lesion models the increased plasma leptin levels are accompanied by obesity, which is certainly not the case in our SCN-lesioned animals. At present, however, we cannot completely exclude some additional effects of damage to the medial preoptic area on the observed changes. The preoptic area has been implicated in glucose metabolism (54, 55, 56), but we are not aware of any data on the effects of lesions in the preoptic area on plasma leptin levels. Therefore, it is possible that, similar to the rhythm in melatonin release, SCN control of leptin release is mainly inhibitory (46, 57), and thus removal of this inhibitory SCN control results in increased leptin levels. However, at least two other explanations cannot be excluded at present. First, the SCN-lesioned animals also show a regular feeding pattern somewhat comparable to that of the regular-fed animals. Therefore, if the regular feeding pattern itself induces increased plasma leptin levels, similar changes might be expected in SCN-lesioned animals, although the effect is more pronounced in the regular-fed group. A second alternate explanation is found in the positive relation between 24-h mean leptin and 24-h mean corticosterone levels. This correlation is in agreement with several previous observations (20, 21, 34, 35) and indicates that the higher plasma leptin levels in the SCN-lesioned animals might be due to the higher corticosterone levels in these animals. Future experiments using SCN transmitter (ant)agonist administration at the level of the PVN, SCN-lesioned animals with an enforced nocturnal feeding pattern, and SCN lesions combined with ADX+CORT, respectively, will allow further investigation of these possibilities. In addition, the data from the SCN-lesioned animals indicate an effect on leptin release and not so much leptin synthesis. Previous experiments have also shown differential effects on leptin release and synthesis (3, 32, 35). Although the present data suggest that leptin secretion can be stimulated from a preexisting intracellular pool, more information about the cellular and molecular biology of leptin is needed before a justified interpretation can be made.
In conclusion, the present study shows that the SCN directly orchestrates the diurnal changes in plasma leptin levels. This study therefore indicates the presence of an efferent pathway, probably mediated by the autonomous nervous system, between the central nervous system and the adipose tissue involved in the control of leptin secretion.
|
Acknowledgments |
|---|
Received October 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-methyl-p-tyrosine. FEBS Lett 429:395–398[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  A. Kalsbeek, F. Kreier, E. Fliers, H. P. Sauerwein, J. A. Romijn, and R. M. Buijs Minireview: Circadian Control of Metabolism by the Suprachiasmatic Nuclei Endocrinology, December 1, 2007; 148(12): 5635 - 5639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Kok, F. Roelfsema, M. Frolich, J. van Pelt, A. E. Meinders, and H. Pijl Activation of Dopamine D2 Receptors Lowers Circadian Leptin Concentrations in Obese Women J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 3236 - 3240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Zvonic, A. A. Ptitsyn, S. A. Conrad, L. K. Scott, Z. E. Floyd, G. Kilroy, X. Wu, B. C. Goh, R. L. Mynatt, and J. M. Gimble Characterization of peripheral circadian clocks in adipose tissues. Diabetes, April 1, 2006; 55(4): 962 - 970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kalsbeek, R. M. Buijs, R. van Schaik, E. Kaptein, T. J. Visser, B. Z. Doulabi, and E. Fliers Daily Variations in Type II Iodothyronine Deiodinase Activity in the Rat Brain as Controlled by the Biological Clock Endocrinology, March 1, 2005; 146(3): 1418 - 1427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bodosi, J. Gardi, I. Hajdu, E. Szentirmai, F. Obal Jr., and J. M. Krueger Rhythms of ghrelin, leptin, and sleep in rats: effects of the normal diurnal cycle, restricted feeding, and sleep deprivation Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2004; 287(5): R1071 - R1079. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R G P Denis, C Bing, S Brocklehurst, J A Harrold, R G Vernon, and G Williams Diurnal changes in hypothalamic neuropeptide and SOCS-3 expression: effects of lactation and relationship with serum leptin and food intake J. Endocrinol., October 1, 2004; 183(1): 173 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Vicentic, G. Dominguez, R. G. Hunter, K. Philpot, M. Wilson, and M. J. Kuhar Cocaine- and Amphetamine-Regulated Transcript Peptide Levels in Blood Exhibit a Diurnal Rhythm: Regulation by Glucocorticoids Endocrinology, September 1, 2004; 145(9): 4119 - 4124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Z. Doulabi, M. P.-T. Schiphorst, A. Kalsbeek, E. Fliers, O. Bakker, and W. M. Wiersinga Diurnal Variation in Rat Liver Thyroid Hormone Receptor (TR)-{alpha} Messenger Ribonucleic Acid (mRNA) Is Dependent on the Biological Clock in the Suprachiasmatic Nucleus, whereas Diurnal Variation of TR{beta}1 mRNA Is Modified by Food Intake Endocrinology, March 1, 2004; 145(3): 1284 - 1289. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Watts, S. Tanimura, and G. Sanchez-Watts Corticotropin-Releasing Hormone and Arginine Vasopressin Gene Transcription in the Hypothalamic Paraventricular Nucleus of Unstressed Rats: Daily Rhythms and Their Interactions with Corticosterone Endocrinology, February 1, 2004; 145(2): 529 - 540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Utsumi, R. Iwakiri, B. Wu, T. Fujise, H. Sakata, R. Shimoda, S. Amemori, S. Tsunada, A. Ootani, and K. Fujimoto Intracerebroventricular Administration of Leptin-Induced Apoptosis in the Rat Small Intestinal Mucosa Experimental Biology and Medicine, November 1, 2003; 228(10): 1239 - 1244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Terazono, T. Mutoh, S. Yamaguchi, M. Kobayashi, M. Akiyama, R. Udo, S. Ohdo, H. Okamura, and S. Shibata Adrenergic regulation of clock gene expression in mouse liver PNAS, May 27, 2003; 100(11): 6795 - 6800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-L. Garidou, B. Vivien-Roels, P. Pevet, J. Miguez, and V. Simonneaux Mechanisms regulating the marked seasonal variation in melatonin synthesis in the European hamster pineal gland Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2003; 284(4): R1043 - R1052. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. W. Kok, F. Roelfsema, S. Overeem, G. J. Lammers, R. L. Strijers, M. Frolich, A. E. Meinders, and H. Pijl Dynamics of the Pituitary-Adrenal Ensemble in Hypocretin-Deficient Narcoleptic Humans: Blunted Basal Adrenocorticotropin Release and Evidence for Normal Time-Keeping by the Master Pacemaker J. Clin. Endocrinol. Metab., November 1, 2002; 87(11): 5085 - 5091. [Abstract] [Full Text] [PDF]
|
|
) is
compared with that in SCN-lesioned (n = 9; •; A) or ADX+Cort
(n = 8; 
) and regular-fed (n = 6; 
) animals (B).
Asterisks indicate plasma leptin values significantly
different (P < 0.01) from trough values between
ZT0 and ZT6. In B, the shaded area indicates the
mean ± SEM for the nonoperated control animals.
