| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Institut National de la Recherche Agronomique, Physiologie de la Reproduction des Mammiferes Domestiques (D.C.S., B.M.), Nouzilly 37380, France; and The Babraham Institute (D.C.S.), Babraham, Cambridge, United Kingdom CB2 4AT
Address all correspondence and requests for reprints to: Dr. Donal C. Skinner, Department of Clinical Veterinary Science, University of Bristol, Langford, United Kingdom BS40 5DU.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is currently thought that melatonin reaches its neural targets via the peripheral circulation (10, 11). According to this approach, melatonin is secreted into the Galen vein, which drains into the sagittal sinus before entering the jugular vein. Venous melatonin is finally transported back to the brain via the carotid arteries. Considering that the pineal gland is a deep brain structure, this route of action would appear rather circuitous. An alternative hypothesis is that melatonin reaches its targets through the cerebroventricular system. Data to support this second hypothesis are, however, sparse, and less than 0.1% of the total amount of melatonin produced by the pineal gland is thought to enter the cerebrospinal fluid (CSF) (12).
Reported CSF melatonin concentrations vary enormously in the few studies published. Moreover, the majority of reports are based on cisternal (11, 12, 13, 14) or lumbar (15, 16) CSF samples, which consistently have concentrations similar to or lower than those found in blood. As this site of collection is somewhat distant from the cerebral ventricles, the concentrations present may not reflect levels found in the ventricles or, more importantly, the levels seen by neural tissue in target sites. Indeed, reports suggest that lateral ventricle CSF melatonin levels are higher than those in both plasma and cisternal/lumbar CSF (17, 18, 19). However, there are no studies investigating CSF melatonin concentrations in the third ventricle; moreover, it is thought that melatonin is not secreted directly into this ventricle (17). Our interest in this particular compartment of CSF has been raised by our discovery that the site of action responsible for transducing the effects of melatonin on reproduction are located in the premammillary hypothalamic area, a periventricular site extending laterally from the third ventricle by only 2 mm (20).
Using a technique we developed recently to continuously withdraw CSF from the third ventricle for long periods (21), the objective of our first study was to characterize third ventricular CSF melatonin concentrations over 24 h and compare these concentrations with levels found in jugular plasma. Further studies sought to establish whether melatonin enters the cerebral ventricular system directly from the pineal.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ventricular surgery
At least 2 weeks before experimentation, a third ventricular
guide cannula was permanently implanted using a method described in
detail previously (21). Briefly, the guide cannula was constructed from
a stainless steel luer-lock needle, which had the tube end blunted and
cut to a length of 40 mm. For placing the guide cannula, the head of
the anesthetized ewe (halothane, Pitman-Moore France, Meaux, France)
was positioned in a stereotaxic frame (Précision
Cinématographique, Paris, France), and 1 ml radioopaque liquid
(Omnipaque, Nycomed Ingenon SA, Paris, France) was injected into
the lateral ventricle. Frontal and lateral x-rays were taken to give
specific landmarks of the ventricular system, especially the thalamic
mass intermedia and the infundibular recess of the third ventricle for
the antero-posterior and vertical orientations and the middle of the
third ventricle for the laterality of the cannula. Using the lateral
x-ray, the cannula was aimed approximately 1 mm in front of and below
the vertical and horizontal tangents, respectively, of the thalamic
mass. When the tip of the cannula was in the third ventricle, CSF
flowed freely back up the tube. The cannula was fixed to the skull with
acrylic dental cement and, to maintain sterility and prevent CSF
leakage, a plastic obturator was inserted into the cannula. A threaded
cylindrical Teflon cap, anchored to the skull with four stainless steel
screws and acrylic cement, protected the whole device. In Exp 3, a
guide cannula was also placed in the lateral ventricle using the same
approach.
CSF and jugular blood collection
Approximately 6 h before sample collection commenced, a
catheter was inserted into the jugular vein and secured to the skin.
The catheter was connected to polyethylene tubing, which was connected
to a peristaltic pump. Animals were then injected with 25,000 IU
heparin, and blood was continuously withdrawn into collection
containers at a rate of 3–4 ml/10 min. Samples were centrifuged, and
plasma was stored at -20 C until assay.
A SILASTIC brand catheter (Dow Corning Corp., Midland, MI) was inserted through the ventricular guide cannula so that the distal tip ended at the tip of the guide cannula. To collect a sample of CSF, the proximal end of the catheter was connected to approximately 1 m of PVC tubing, which, in turn, was connected to peristaltic tubing. The peristaltic tubing was threaded through a rate-adjustable peristaltic pump (Minipuls 2, Gilson, Villiers-le-bel, France), and CSF fractions were extracted for the desired period. For each ewe, the time taken for CSF to reach the fraction tubes was recorded, which enabled the true time the sample was extracted from the third ventricle to be calculated. Unless stated otherwise, CSF was collected at a rate of 30 µl/min, and fractions were stored at -20 C until assayed.
Exp 1: profile of melatonin in third ventricular CSF
To characterize the nightly profile of melatonin in CSF, six
Clun-Forest ewes were prepared for CSF collection as described. To
obtain a prenight average, hourly sampling started 5 h before
lights off and continued for 24 h. Half an hour before lights off,
samples were collected every 3 min for the next 90 min. Similarly, 30
min before lights on, samples were collected every 3 min for the
following 90 min. One hour of 3-min samples was also collected in the
middle of the night to establish whether CSF melatonin release was
pulsatile.
To determine the effect of CSF withdrawal on melatonin release, the flow rate was decreased (10 µl/min) for the second night in four of the six ewes, during which hourly samples were collected.
To see whether the light-mediated suppression of melatonin release, well established in plasma, also occurred in third ventricular CSF, the lights were turned on at midnight during the second night in the remaining two ewes. In these animals, CSF was collected in 3-min fractions for 30 min before and after lights on.
Exp 2: do CSF and plasma melatonin levels increase
coincidentally?
In the previous experiment, it was apparent that the 3-min
sampling regimen did not permit sufficient resolution of the time
courses to determine whether melatonin increased simultaneously in the
blood and the CSF or whether there was a time delay in a particular
compartment. To investigate this further, six ewes were prepared for
CSF and jugular blood collection as described. Half an hour before
lights off, two 10-min samples were collected. Samples were collected
every 30 sec for 50 min, starting 10 min before lights off. At the end
of this period, two additional 10-min samples were collected to
establish whether melatonin secretion was still increasing.
Exp 3: is melatonin released directly into the third
ventricle?
The delay evident in the increase in CSF melatonin may have been
a result of recirculation of melatonin through the choroid plexus and
into the lateral ventricle. To test this hypothesis, five ewes were
prepared for the simultaneous collection of CSF from the lateral and
third ventricles. At the time of the experiment, it was possible to
collect CSF simultaneously from both compartments in only three ewes.
In these ewes, 30-sec samples were collected starting 10 min before
lights off and continuing for an additional 40 min. At the end of the
night, 30-sec CSF samples were collected from both compartments
starting 10 min before lights on and continuing for an additional 40
min.
Melatonin assay
Jugular plasma melatonin concentrations were determined in
100-µl aliquots, using a well validated RIA (22). Melatonin
concentrations in 2- to 30-µl aliquots of CSF samples were measured
in the same assay system, except that the standard curve was made in
assay buffer instead of plasma. Parallelism between the standard curve
and sample dilution was checked for volumes in the range 0–50 µl,
and accuracy was checked by assaying melatonin after loading daytime
samples with exogenous melatonin. In Exp 2, to eliminate the assay
sensitivity difference caused by the higher melatonin concentrations in
CSF, the daytime level in plasma and CSF was first determined. The
volume of CSF that would yield the same amount as that found in 100
µl plasma was then calculated, and samples of CSF corresponding to
this volume were assayed simultaneously with 100-µl plasma samples.
Assay sensitivity averaged 0.4 pg/tube, and intra- and interassay
sensitivities were 4% and 1.1%, respectively, for plasma and 6% and
13%, respectively, for CSF.
Analysis
The onset of the melatonin increase was taken as the time
melatonin concentrations exceeded either the mean plus 3SD
of samples collected during the hour preceding darkness or 3 times the
assay sensitivity, whichever was the greatest. The offset was taken as
the time melatonin concentrations fell below the mean -
1SD of samples collected during the hour preceding lights
on. The time taken for CSF and plasma melatonin levels to reach 10% of
the mean nocturnal level was also calculated in Exp 1. The nocturnal
maximum and the mean concentration during darkness were determined
using hourly values. Statistical comparisons between the two
compartments were made using Student’s t test for paired
data, with significance set at P < 0.05. Pearson’s
correlation was performed to establish the relationship between the
mean melatonin concentration in third ventricular CSF and plasma for
Exp 1 and 2 combined (n = 11).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. In
both CSF and plasma, the duration of the night is accurately reflected
by the increase in the duration of melatonin release. CSF melatonin
increased above the concentration (34 ± 4 pg/ml) detected in the
hour preceding lights off within 14 ± 3 min of darkness and
started to fall within 13 ± 2 min of lights on. Similarly, plasma
melatonin increased above a prenight baseline (5 ± 1 pg/ml)
within 10 ± 3 min of darkness and started to fall within 8
± 3 min after lights on. Melatonin levels fell below 10% of the
nocturnal mean after lights on in 39 ± 4 and 36 ± 4 min in
the CSF and plasma, respectively.
|
Reducing the flow rate to 10 µl/min during the second night did not
affect the measurable level of melatonin in CSF (Fig. 2). Neither the mean nightly CSF (night
1, 1327 ± 106 pg/ml; night 2, 1175 ± 196 pg/ml) nor plasma
(night 1, 139 ± 22 pg/ml; night 2, 110 ± 12 pg/ml) level
was significantly different over the 2 days. This suggests that the
high melatonin concentrations in third ventricular CSF found in this
study are not due to disturbances in the normal CSF flow pathway and
that the withdrawal of CSF does not affect the neural circuitry
involved in transmitting photic information to the pineal gland.
|
Exp 2: do CSF and plasma melatonin levels increase
coincidentally?
Melatonin increased earlier in the plasma than in CSF in all six
ewes (Fig. 3). After a delay of only
7.8 ± 1.7 min in the plasma, but 12.7 ± 1.4 min in the CSF,
melatonin levels increased significantly above their daytime baseline.
This amounted to a significant (P < 0.05) delay of
4.9 ± 1.1 min between the two compartments. As before, mean
nocturnal CSF melatonin concentrations were significantly (21 ±
8-fold; range, 7–62) higher than those found in plasma.
|
Exp 3: is melatonin released directly into the third
ventricle?
In the three ewes from which it was possible to collect CSF from
the lateral and third cerebral ventricles simultaneously, melatonin
concentrations were 7 ± 3 times higher in the third ventricle
(Fig. 4). In these ewes, melatonin
concentrations averaged 2247 ± 629 pg/ml in the third ventricle,
but were only 525 ± 212 pg/ml in the lateral ventricle. Melatonin
levels also started to increase in the third ventricle before
increasing in the lateral ventricle in all ewes (ewe 040, 8
vs. 9 min; ewe 120, 10 vs. 13 min; ewe 354, 26
vs. 50 min).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our finding that CSF melatonin concentrations are significantly higher than plasma levels concurs with earlier studies on goats (18) and calves (17), in which hourly to every 2 h instantaneous samples of lateral ventricle CSF were shown to exceed jugular plasma concentrations by between 5- and 18-fold. Similarly, in the sheep, hourly lateral ventricle CSF melatonin concentrations were reported to be up to 8-fold higher, whereas in the cisterna magna, CSF levels were comparable to those in plasma (19). However, Rollag and co-workers (12) found that cisterna magna CSF melatonin levels in sheep were significantly lower than those in the plasma, which led these researchers to conclude that less than 0.1% of the total melatonin released by the pineal gland enters the cerebral ventricles. The results from the present study showing high concentrations of melatonin in the CSF as well as a rapid clearance of melatonin from the CSF after the lights had been switched on suggest very strongly that this figure needs to be reassessed.
How does melatonin get into the CSF? Early studies on humans and monkeys using lumbar or cisternal CSF suggested that melatonin entered the CSF from the blood (10, 11). To explain reports that melatonin concentrations in the lateral ventricle were higher than those in the plasma (17, 18), this hypothesis was subsequently refined to suggest that there was retrograde blood flow from the Galen vein to the choroid plexus (19, 23). Upon reaching the choroid plexus, melatonin would simply diffuse down its concentration gradient into the lateral ventricle. Anatomical evidence showing connections between the choroid plexus and the Galen vein lent support to this hypothesis (24, 25). Indeed, as melatonin increased in the jugular blood before it increased in the CSF at the transition to darkness, our second experiment appeared to support this proposal. However, our final study showed unequivocally that third ventricular levels of melatonin are 7-fold higher than lateral ventricle concentrations. This result suggests very strongly that the delay in the increase between CSF and plasma is due to the diffusion of melatonin from the pineal tissue to the third ventricle and not to the delay in melatonin entering first the blood and then the CSF.
How is it possible for lateral ventricle CSF melatonin levels to be elevated when the direction of CSF flow is away from the lateral ventricle? There are three possibilities, which may not be mutually exclusive. First, there may be some recirculation of CSF as previously proposed (23). However, as lumbar and cisternal CSF concentrations appear to be consistently low, such recirculation would not explain why levels are elevated above plasma concentrations. Second, there may well be some retrograde blood flow into the choroid plexus, resulting in elevated melatonin concentrations in this structure and allowing melatonin to diffuse into the lateral ventricle down its concentration gradient. Third, the flow of CSF may not be unidirectional in the ventricles. A characteristic of the ventricular system evident during surgery (our personal observation) is a very strong pulsatility. During each pulse, it is probable that at least some third ventricular CSF enters the lateral ventricle. In this respect it is noteworthy that lateral ventricle melatonin concentrations started to rise after an increase was noted in the third ventricle in all three ewes. This last hypothesis would not only explain the lower melatonin level in the lateral ventricle compared with that in the third ventricle, but would also explain why it is still higher than that in the plasma.
Does CSF melatonin have a function? The concept of humoral signals within the CSF was proposed many years ago (26), and interest in this hypothesis has recently resurfaced (27, 28). However, only indirect evidence exists to support the functional significance of CSF-borne factors (29). Of all the biological molecules found in CSF, melatonin may be the first to validate this hypothesis. First, it should be noted that our study concurs with the few other CSF studies that consistently show that the nocturnal increase in CSF melatonin, like that in plasma melatonin, accurately reflects the duration of the night. CSF melatonin, therefore, contains the chemical code of day length, which is an essential prerequisite if it is to play a role in transducing circadian and photoperiodic information. Second, there will be a huge chemical gradient between CSF and the surrounding tissue and plasma. It is of interest, therefore, that of all of the factors detected in the CSF, melatonin has the highest ratio with respect to plasma (30). Indeed, if plasma levels in the brain are reflected more accurately by carotid plasma concentrations and not jugular levels, then the CSF to plasma ratio is likely to be severalfold greater than we have reported in this study. This is because melatonin is degraded quite quickly (half-life in sheep is between 15–20 min) (31), and the melatonin concentration in jugular plasma represents levels before its dilution in the peripheral circulation. Third, melatonin is small (mol wt, 233) and highly lipophilic, properties that would facilitate its movement through tissue. Indeed, in an earlier study we found that melatonin diffused through neural tissue up to 2 mm away from an isolated 0.45-mm diameter microim-plant (32).
The physiologically important target sites of melatonin remain largely unidentified. The detection of a surfeit of melatonin receptors in the pars tuberalis of the adenohypophysis suggested that this region could be a crucial site of action. Although the pars tuberalis may have other functions, we have recently shown that it is not responsible for transducing the effects of melatonin to the neuroendocrine reproductive axis (20, 22, 32). Specifically, melatonin apposed to (22) or implanted directly in (32) the pars tuberalis has no effect, whereas melatonin microimplants in the hypothalamus (22, 32) and, more specifically, the premammilary hypothalamic region (20) potently affect reproductive functioning. This periventricular site extends laterally from the side of the third ventricle by only 2 mm. We believe strongly that melatonin will simply diffuse from the CSF in the third ventricle to its target in the premammilary hypothalamic region. This CSF hypothesis may appear at odds with studies showing that melatonin administered systemically to pin-ealectomized animals can entrain reproductive function (33, 34). These studies showed unequivocally that melatonin was the primary endocrine factor involved in seasonal reproductive entrainment, but no extrapolation about the route used by melatonin to get from its source to its target is possible. This is because the levels produced in these melatonin replacement studies replicated jugular concentrations and not carotid artery levels, which, as mentioned, are likely to be severalfold lower (19). Our laboratory is currently investigating the relationship among the carotid, jugular, and CSF compartments.
CSF melatonin has recently been proposed to play an interesting and potentially critical role in Alzheimer’s disease (35, 36). According to this hypothesis, the lower levels of CSF melatonin observed in Alzheimer’s disease patients compared with normal controls (36, 37) mediate at least in part the loss of memory and other functions associated with this disease. This speculation is based on the recent report that melatonin is a powerful free radical scavenger and antioxidant (7, 8, 38). Thus, reduced melatonin in CSF could lead to an increase in neuronal damage and death. It is worth noting, therefore, that although no experimental data exist to substantiate this proposal, recent studies on individuals suffering from Alzheimer’s disease showed an improvement of several functions after melatonin replacement (9, 39, 40). This novel area of research warrants significant further investigation.
A final point of note is the question of the physiological vs. pharmacological action of melatonin, which was recently reviewed very elegantly by Brzezinski (41). Until now, this comparison has been made on the basis of the concentration detected in the peripheral circulation and how exogenous melatonin elevates this level. Our study provides compelling evidence that this question needs to be reevaluated, particularly with respect to neurally mediated functions.
In summary, we report for the first time that melatonin concentrations in the third ventricle are 20 times greater than those in the peripheral circulation and have a robust circadian profile, increasing and decreasing in accordance with darkness and light, respectively. Although melatonin increases in the plasma before it rises in the CSF, third ventricular melatonin does not appear to be derived from the peripheral circulation or the choroid plexus as previously proposed. Rather, CSF melatonin seems to originate directly from the pineal gland and may simply diffuse from this source into the third ventricle. The potential importance of CSF melatonin has recently been highlighted by the discovery of periventricular melatonin sites of action and the suggestion that melatonin may have a role in neurological diseases such as Alzheimer’s.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received April 7, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4/4 genotype. J Clin Endocrinol Metab 84:323–327This article has been cited by other articles:
 |
|
 |
|
  T. Lialiaris, E. Lyratzopoulos, F. Papachristou, M. Simopoulou, C. Mourelatos, and N. Nikolettos Supplementation of melatonin protects human lymphocytes in vitro from the genotoxic activity of melphalan Mutagenesis, May 22, 2008; (2008) gen020v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Bechtold and A. S. I. Loudon Hypothalamic Thyroid Hormones: Mediators of Seasonal Physiology Endocrinology, August 1, 2007; 148(8): 3605 - 3607. [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. Fischer, T. W. Sweatman, I. Semak, R. M. Sayre, J. Wortsman, and A. Slominski Constitutive and UV-induced metabolism of melatonin in keratinocytes and cell-free systems FASEB J, July 1, 2006; 20(9): 1564 - 1566. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Ma, J. R. Idle, K. W. Krausz, and F. J. Gonzalez METABOLISM OF MELATONIN BY HUMAN CYTOCHROMES P450 Drug Metab. Dispos., April 1, 2005; 33(4): 489 - 494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Reiter, D.-x. Tan, J. Leon, U. Kilic, and E. Kilic When Melatonin Gets on Your Nerves: Its Beneficial Actions in Experimental Models of Stroke Experimental Biology and Medicine, February 1, 2005; 230(2): 104 - 117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. W Rajaratnam, B. Middleton, B. M Stone, J. Arendt, and D.-J. Dijk Melatonin advances the circadian timing of EEG sleep and directly facilitates sleep without altering its duration in extended sleep opportunities in humans J. Physiol., November 15, 2004; 561(1): 339 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. W. Rajaratnam, D.-J. Dijk, B. Middleton, B. M. Stone, and J. Arendt Melatonin Phase-Shifts Human Circadian Rhythms with No Evidence of Changes in the Duration of Endogenous Melatonin Secretion or the 24-Hour Production of Reproductive Hormones J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4303 - 4309. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Nakahara, M. Nakamura, M. Iigo, and H. Okamura Bimodal circadian secretion of melatonin from the pineal gland in a living CBA mouse PNAS, August 5, 2003; 100(16): 9584 - 9589. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Wolfler, P. M. Abuja, W. Linkesch, K. Schauenstein, P. M. Liebmann, Vijayalaxmi, C. R. Thomas Jr, R. J. Reiter, and T. S. Herman Questionable Benefit of Melatonin for Antioxidant Pharmacologic Therapy J. Clin. Oncol., October 1, 2002; 20(19): 4127 - 4129. [Full Text] [PDF]
|
|
|
|
|
|
Vijayalaxmi, C. R. Thomas Jr, R. J. Reiter, and T. S. Herman Melatonin: From Basic Research to Cancer Treatment Clinics J. Clin. Oncol., May 15, 2002; 20(10): 2575 - 2601. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Tricoire, A. Locatelli, P. Chemineau, and B. Malpaux Melatonin Enters the Cerebrospinal Fluid through the Pineal Recess Endocrinology, January 1, 2002; 143(1): 84 - 90. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. MacGibbon, R. S. Walls, and A. V. Everitt An Age-Related Decline in Melatonin Secretion Is Not Altered by Food Restriction J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2001; 56(1): 21B - 26. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
