Recovery of function following unilateral denervation, but not unilateral decentralization, of the pineal gland as indicated by measurements of pineal melatonin content and urinary melatonin metabolites

0306-4522/90$3.00+ 0.00 Pergamon Press plc 0 1990IBRO

Neuroscien~.?Vol. 37, No. 2, pp. 413-420, 1990 Printed in Great Britain

RECOVERY OF FUNCTION FOLLOWING UNILATERAL DENERVATION, BUT NOT UNILATERAL DECENTRALIZATION, OF THE PINEAL GLAND AS INDICATED BY MEASUREMENTS OF PINEAL MELATONIN CONTENT AND URINARY MELATONIN METABOLITES G.

A. KUCHEL,*~$ R. L. SHERMAN~and R. E. ZIGMOND~/~

TDepartment of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02115, U.S.A. $Division on Aging, Harvard Medical School; Department of Medicine, Beth Israel and Brigham and Women’s Hospitals, Boston, MA 02115, U.S.A. @e&ion of Analytical Biochemistry, Laboratory of Clinical Science, National Institute of Mental Health, National Institutes of Health, Bethesda, MD 20892, U.S.A. jlDepartment of Neurosciences, Case Western Reserve University, 2119 Abington Road, Cleveland, OH 44106. U.S.A. Abstract-The rat pineal gland is an attractive system for studies on the capacity of neural systems to recover following partial injury, allowing both for the creation of precise subtotal lesions and for the measurement of recovery of function at the cellular level. The pineal gland receives overlapping sympathetic innervation from the right and left internal carotid nerves from neurons whose cell bodies are located in the two superior cervical ganglia. This innervation regulates several aspects of pineal metabolism in a circadian fashion, with the most dramatic being a marked increase in the night-time activity of N-acetyltransferase, a key enzyme regulating the rate of melatonin synthesis. We have previously shown that a highly divergent pattern takes place in the night-time activity of this enzyme following two different unilateral lesions of the sympathetic innervation to the gland. Thus, following a unilateral lesion of the internal carotid nerve (unilateral denervation), there is an initial decline in ~-a~tyltransfera~ activity; however, normal activity is again seen during the second and subsequent nights. In contrast, a unilateral lesion of the cervical sympathetic trunk, the nerve that innervates the superior cervical ganglion (unilateral decentralization), results in “permanent” impairment of N-acetyltransferase activity. In the present study, we report that the functional capacity of the entire pathway for melatonin synthesis is similarly affected following these lesions, as reflected by the levels of melatonin and of its precursor N-aeetylserotonin in the pineal gland, as well as the levels of the main melatonin metabolite 6-hydroxymelatonin in the urine. N-Acetyltransferase activity was determined 1.5, 14 and 21 days after unilateral denervation and was found to be 97%, 122% and 98% of the night-time activity in sham-operated animals. At all three points, the pineal contents of N-acetylserotonin (84%, 129% and 105%) and of melatonin (90%, 110% and 115%) were also similar to the levels obtained in sham-operated animals. In contrast, after unilateral de~ntralization the activity of ~-a~tyltmnsfera~ (18%, 32% and 21%) and the content of both N-acetylserotonin (12%, 43% and 41%) and melatonin (35%, 55% and 47%) were lower than in sham-operated animals. Urinary measurements of 6-hydroxymelatonin, the main melatonin metabolite, showed a robust diurnal rhythm in control animals, with increased excretion at night. Measurements of this metabolite in the urine allowed assessment of pineal function in individual animals, before and after specific lesions. Following unilateral denervation, there was a return of normal trhydroxymelatonin levels by the second night after surgery. However, following unilateral d~ntr~i~tion, night-time 6-hydroxymelatonin measurements remained lower than prelesion values for at least 13 nights after the lesion. This study provides additional evidence for our hypothesis that, after certain lesions, intact, but electrophysiolo~~lly silent, nerve terminals of decentralized nerve cells may adversely affect the function of terminals from unoperated neurons and, thereby, prevent functional recovery.

*Brookdale National Fellow, The Brookdale Foundation. Present address: Mount Sinai School of Medicine, New York, NY 10029. U.S.A. ~TO whom correspondence should be addressed at: Case Western Reserve University. Abbreviurions: CST, cervical sympathetic trunk; EDTA, ethylenediaminetetra-acetic acid; HIOMT, hydroxyindole 0-methyltransferase; ICN, internal carotid nerve; NAT, N-acetyltransferase; NE, noreninenhrine: SCG. _. superior cervical ganglion. NSC 37/Z-E

413

The capacity

of neural

systems to recover from

partial injury is of great interest both to the scientist and to the clinician. While this problem has received

considerable study, only rarely has it been possible to assess functional recovery of an entire system in a quantitative fashion at the cellular level.-” The sympathetic innervation of the rat pineal gland presents

several attractive features that allow for the creation of highly precise and reproducible partial lesions and

414

G. A.

KUCHEL et al

for the evaluation of functional recovery after such lesions. .?.Xi.ii The pineal gland is a midline structure innervated by noradrenergic sympathetic fibers of the internal carotid nerves (ICN) originating from the right and left superior cervical ganglia (SCG)? The two nerves each provide approximately 50% of the pineal innervation, distributed equally over the two halves of the gland.‘” These sympathetic nerves regulate a number of aspects of pineal biochemistry, with the most dramatic being the regulation of the synthesis of the hormone melatonin. Pineal parenchymal cells synthesize melatonin from serotonin by reactions catalyzed by two enzymes.“.‘3,2x The first, serotonin N-acetyltransferase (NAT; arylamine N-acetyltransferase. EC 2.3.1.87). catalyses the acetylation of serotonin to N-acetylserotonin. This compound is subsequently O-methylated to produce melatonin by a reaction catalyzed by the enzyme hydroxyindole 0-methyltrdnsferase (HIOMT; N-acetylserotonin 0-m~thyltransferase, EC 2.1.1.4). The sympathetic fibers to the pineal gland are the final component of a neural pathway by which changes in environmental lighting can influence the activity of NAT and HIOMT and the rate of synthesis of meIaton~n.‘3.~~,~9The onset of darkness results in a dramatic (40- to loo-fold) increase in the activity of NAT. This diurnal increase appears to be mediated through changes in the activity of the sympathetic noradrenerg~c fibers innervating the pineal gland and the release of the neurotransmitter norepinephrine (NE), since bilateral superior cervical ganglionectomy”~‘” or administration of the betaadrenergic antagonist propranolo14 lead to the abolition of this rhythm. In addition, electrical stimulation of the sympathetic innervation of the pineal gland during the daytime increases pineal NAT activity to levels comparable to night-time values.’ Thus, the level of NAT activity appears to reflect the extent of sympathetic stimulation of the pineal gland. Zigmond et al.“” discovered a dramatic difference in the recovery of night-time NAT activity following two types of subtotal lesions of the innervation of the pineal gland. Following unilateral superior cervical ganglionectomy or after cutting one ICN (UniICNX or unilaterat denervation; Fig. t), the activity of NAT was reduced by 15% during the first night after surgery, with enzyme activity returning to normal by the second night. Cutting one preganglionic cervical sympathetic trunk (CST), the nerve that provides the central innervation to the SCG (UniCSTX or unilateral decentralization; Fig. I), also resulted in a 75% reduction in NAT activity during the first night after the lesion. However, exanlination of the effects of this lesion at subsequent time points showed an absence of recovery up to 5 months later.3’ This study also revealed that “heteroneuronal uptake” of released NE by the intact, but decentralized, and. therefore. electrophysiological silent nerve terminals,

Controt

Fig. I. The sympathetic innervation of the pineal gland and the location of lesions. Diagram of the sympathetic innervation of the normal pineal gland (left) and of the location of the lesions to unilatera~y denervate (centre) or unilaterally decentralize (right) the gland. SCG, superior cervical ganglion; CST, cervical sympathetic trunk; ICN, internal carotid nerve; broken lines, anterograde degeneration resulting from the lesions. (Taken from Zigmond <“lal.3”)

appears to account for this “permanent” impairment in NAT activity.3’ In the present study, we expanded our examination of the degree of functional recovery of the metabolic pathway for the biosynthesis of melatonin by measuring the pineal concentrations of melatonin itself and of its immediate precursor, ~-acetylserotonin. This was of particular interest since three previous studies7.23,27 had indicated a lack of complete recovery of pineal melatonin content after unilateral ganglion~tomy, in contrast to our findings with NAT activity. In addition, we have examined the effects of sympathetic nerve lesions on the day/night pattern of urinary excretion of 6-hydroxymelatonin, the main metabolite of melatonin. These latter measurements have allowed us to examine pineal function in individual animals, both before and at various times after specific nerve lesions. A preliminary report of this study was made to the Society for Neuroscience.” EXPERIMENTAL

PROCEDURES

Animnls and treatmenrs

The animals used in these studies were Z-3 months old (300450 g) male Sprague-Dawley rats obtained from Charles River Breeding Laboratories (Wilmington. MA) and Zivic Miller (Allison Park, PA). Animals were housed in individual plastic cages under controlled lighting (I2 h light: I2 h darkness) with ad l~~j~~ access to Purina Rat Chow and water for at least 10 days before each experiment. All surgical procedures were performed at least 4 h into the light cycle. Animals were anesthetized with chlorat hydrate (735 mgikg, s.c.; Sigma Chemical, St Louis. MO). The ICN was cut near its entry into the carotid canal and again near the SCG, removing a small piece of the nerve (Fig. 1). In other animals, a piece of the CST (approximately 3 mm long) was removed proximal to the SCG. “Shamoperated” animals underwent anesthesia and superficial neck exploration only, without a nerve lesion. The success of ICN or CST nerve lesions was evaluated by the development of ptosis, which was found to be present in all of the lesioned animals studied. Animals were killed by decapitation, and their pineals were quickly removed. With the exception of the data obtained in Table 1, all animals were killed 6-9 h into the dark cycle under a dim red light2

415

Pineal function after partial sympathetic nerve lesions Table I. Diurnal variation in ~-acetyltransferase activity and ~-a~tyl~rotonin melatonin contents in the pineai gland

Day Night Night Night Night Night

hour hour hour hour hour hour

4 2 4 6 8 10

and

NAT activity (pmotlpg protein per 20 min)

N-Acetylserotonin content (pmoti pg protein)

Melatonin content (pmol/pg protein)

I.0 & 0.0 3.6 + 0.8 24.0 k I.1 38.6 + 3.8 35.9 + 0.9 32.1 + 6.8

7.8 + 0.6 67.7 + 31.9 86.4 + 20.0 135.5 f 14.8 L50.0& 5.5 109.7 + 32.7

0.5 * 0.4 29.0 * 1.1 31.2k3.1 38.4 f 4.7 45.5 & 3.2 38.6 & 9.7

All night-time groups included at least four animals, while the daytime group included three animals. Each data point represents the mean f S.E.M. for that group. One-way analysis of variance revealed significant diurnal changes in NAT activity (P < O.OOl), N-acetylserotonin (P < 0.05) and melatonin content (P < 0.01). The Tukey Honesty Sum Difference test showed that while NAT activity rose between hours 2 and 4 of the night, N-acetylserotonin and melatonin contents did not change significantly during that or later periods. Pineal neurochemical assays

When NAT activity was to be measured, pineal glands were immediately frozen on dry ice and stored at -80°C until assayed. NAT activity was determined by a moditication of a radiochemical assay developed by Deguchi and Axelrod,5 in which the rate of formation of [‘4C]N-acetyltryptamine from [‘4C]acetyl-CoA (New England Nuclear Corp., Boston, MA) and tryptamine is measured.’ Individual pineal glands were homogenized in 1OOfll of 0.1 M sodium phosphate buffer, pH 6.8, using glass-glass homogenizers. An aliquot of 40~1 was withdrawn for NAT assay. One hundred and twenty five microliters of a 165 mM trichloroacetic acid solution containing 50 nM epinephrine, as an internal standard, were added to another 40 ttl portion of the homogenate. Following protein precipitation, this aliquot was assayed for N-acetylserotonin and melatonin contents using high performance liquid chromatography. N-Acetylserotonin was eluted from a 25 cm Beckman Ultrasphere ODS 5 pm column with a mobile phase containing 50 mM H,PO, (pH 2.6), 0. t mM EDTA, 12% methanol and 0.4 mM octylsodium sulfate. Detection was accomplished electrochemically (using an ESA 5100A detector: ESA, Bedford, MA) with three electrodes set in series at +0.40 V, +0.05 V and -0.42 V. Melatonin was eluted from a 7.5 cm Beckman Ultrasphere ODS 5 tlrn column with a buffer containing 0.1 M-sodium citrate, 0.1 M sodium acetate (pH 3.8). 0.1 mM EDTA. 8% methanol and 8% acetonitrile (a modification of the method of Mefford er a(.?“). For melatonin detection, the three electrodes were set in series at +0.71 v, +0.37 V and -0.28 V. The protein content of the pineal gland was determined by the method of Lowry ef 01.‘~ Measurements

ofurinary melatonin metabolites

For measurement of urinary 6-hydroxymelatonin, animals were maintained in individual steel metabolic cages for 3 weeks prior to the beginning of an experiment. Lights were turned on from 7 a.m. to 7 p.m. Urine collections were made directly into plastic tubes placed on dry ice and subsequently stored at - 80°C. Urine was collected during the two nights and intervening day prior to surgery, immediately after surgery and I2 days after surgery. Collection tubes were changed 30min before the onset of darkness and 30min after the onset of light, resulting in an 21h “daytime” and 13 h “night-time” collection period. The ~hydroxymelatonin content of the urine was determined by gas chromatography-negative chemical ionization mass spectrometry. Tetradeuterated 6-hydroxymelatonin sulfate was added as an internal standard to a urine aliquot. This was followed by enzymatic (Glusuiase, Cal B&hem, San Diego, CA) hydrolysis, extraction of the freed Ghydroxymelatonin into

dichloromethane, a reaction to form a stable t-butyldimethylsilylpentafluoropropionyl derivative and column chromatography on a silica ge1.25Samples were then analyzed on a hybrid gas chromatograph-negative chemical ionization mass spectrometer consisting of a Varian 3700 capillary gas chromatograph fitted with a 0.25 pm narrow bore DB-5 capillary column and SpectrEL electronics (Extranuclear Laboratories, Inc., Pittsburgh, PA) controlling a Finnigan 3200 chemical ionization mass spectrometer. The data were recorded on a Teknivent data system. Slalislical analysis

Statistical analysis was performed by Student’s f-test for two means (two-tails) or using one- or two-way analysis of variance, as noted. When analysis of variance was significant, it was followed by Tukey’s Honesty Sum Difference test, allowing for multiple comparisons of group pairs. RESULTS Neurochemical

results

The time course of changes in the night-time values of NAT activity and N-acetylserotonin and meiatonin content, respectively, determined in individual pineal glands from animals that had earlier undergone a sham operation, unilateral denervation or unilateral decentralization are represented in Fig. 2A, 3 and C. Bilateral denervation of the pineal gland resulted in nearly complete disappearance of these three parameters, indicating their dependence on sympathetic neural innervation (Fig. 2). In agreement with our earlier studies,30.3’ nighttime NAT activity was normal by the second night after unilateral denervation of the pineal gland (Fig. 2A) and remained at control values 14 and 21 days after this surgery. Also in agreement with these earlier experiments, unilateral decentralization resulted in an approximately 75% decrease in NAT activity at all three time points examined. In addition to con~~ing the earlier data for NAT activity, we found that after both pre- and postganglionic nerve lesions, the pineal content of Nacetylserotonin (Fig. 2B) and of melatonin (Fig. 2C) underwent changes qualitatively similar to those in NAT activity. Following unilaterai denervation,

416

G. A. KUCHELei of.

the pineal content of both ~-acetylseroton~n and melatonin was similar in sham-operated animals and in animals lesioned 1.5, 14 or 21 days earlier. As in the case of NAT activity, unilateral decentralization resulted in Lower levels of these two pineal indoles. However, there appeared to be quantitative differences in the three measures. Unilateral decentralization reduced night-time NAT activity to approximately 20-30% of that found in shamoperated animals. In the same lesioned animals, the levels of the pineal N-acetylserotonin and melatonin were proportionately higher, representing up to 40%

A. NAT ACTIVITY

B. N-ACE~LSEROTONIN T

125 2

100

2

75

S

50

C. MELAT~Nt~

and 50%, respectively, of the values obtained in the sham-operated animals. Urinary meiatonin metabafites

As previously described,” measurement of urinary 6-hydroxymelatonin revealed a robust diurnal rhythm, with night-time collections being S-&foId greater than those abtained during the daytime (Fig. 3). This rhythm was greater and the data less variable when expressed as ng of 6-hydroxymelatonin per collection period rather than per ml of urine, since there was significant inter-animal variability in the volume of urine produced (data not shown). The values obtained in all three experimental groups prior to the surgical procedures were similar. Urinary collections were made to evaluate functional recovery both at early (I st and 2nd nights), and later time points (12th and 13th nights) after the surgical procedures (Fig. 3). Normal high night-time Ghydroxymelatonin excretion and normal day/night differences in excretion were seen at all times examined subsequent to sham surgery. Following unilateral denervation, urinary levels were not statistically different on the first night as compared to the subsequent day. However, later nights were significantly different from their accampanying day values. In addition, the night-time values obtained after the surgery were not different from those obtained prior to surgery. In contrast, the night-time values obtained both at early and later time points after unilateral decentralization were significantly lower than those measured prior to this surgery (P < 0.05 for both comparisons). The mean postlesion night value was 67% of the mean prelesion night value. In addition, after this surgery, although there appeared to be a day/night rhythm, none of the night-time values differed significantly from their corresponding daytime vaiues using Tukey’s Honesty Sum Difference test, with the exception of night 13. Circadian changes in pineal neurochemical indices

0

0

P. , , , , * , t . . . P. 10

Days After

.

*

I

’

*

20

Surgery

Fig. 2. NAT activity, ,~-a~etyiserotonin and melatonin content following partial sympathetic nerve lesions. NAT activity, N-acetylserotonin and melatonin content were determined in individual pineal glands 1.5, 14 and 21 days after unilateral cutting of the internal carotid nerve (ICN) or the cervical sympatbeti~ trunk (CST). In addition, the effects of bilateral lesions of the ICN (0) were examined at I .5 and 14 days. Results are expressed as a percentage of the values obtained in animals undergoing sham operations at the same time intervals &S.E.M. At1 groups include at least five animals, with the exception of the two animals bilaterally lesioned 14 days earlier. Among the I5 sham-operated animals, mean values for NAT activity were 40.8 & 3.5 pmoliyg protein per 20 min, for N-acetylserotonin content, 120.3 4 20.3 pmol/yg protein and for melatonin content. 36.9 + 3.1 pmol/jfg protein. *P i 0.05; **P < 0.005 as compared to sham-operated animaIs in the same group.

Pineal NAT activity, as well as ~-acetyiserotonin and melatonin content, were studied in control animals at a number of time points in the circadian cycle (Table 1). This was done to evaluate the extent to which these measures changed in parallel during the night-time in control animals. In the case of NAT activity

and

obtained

melatonin

were markedly

content.

night-time

values

higher than

the single corresponding daytime value. As found earlier,’ NAT activity 2 h into the dark cycle was less than 10% of the ultimate maximum value, which was not reached until 6 h into the dark phase. In contrast, both melatonin and ~-acetyIserotonin underwent more dramatic their

early

maximal

increases, night-time

reaching

50%

or more

of

value 2 h into the dark cycle. Between 2 and 6 h into the night, NAT activity increased IO-fold, while melatonin levels did not change

signjficantIy.

417

Pineal function after partial sympathetic nerve lesions DISCUSSION The rat pineai

gland

as a model system

The pineal gland has proven to be a useful model system for studying neuronal plasticity and the capacSham Surgery T

4001

300

200

100

0

Unilateral ICNx

400

1

Unilateral CSTx

ity of a retatively simple neural system to recover from partial injury. In addition to allowing for the creation of highly specific and reproducible lesions, it also permits the quantitative evaluation of functional recovery at the cellular level. Previous studies from this laboratory have utilized the night-time activity of the enzyme NAT as an index of functional recovery.30,3’These studies established that the extent of recovery of pineal function differed dramatically after unilateral denervation from that after unilateral decentralization. Both lesions resulted in decreased function during the first night after surgery. However, while there was full recovery by the second night following unilateral denervation, function remained impaired after unilateral decentralization. This impairment appears to be mediated by “heteroneuronal uptake”,” a mechanism that involves the uptake of the NE released from ele~trophysiologically active nerve terminals by nearby nerve terminals that are decentralized and, therefore, electrophysiologically silent. Part of the evidence for this hypothesis comes from the observation that the systemic administration of desmethylimipramine, a specific inhibitor of neuronal NE uptake, could reverse the transient impairment in night-time NAT activity 8-9 h after unilateral denervation,)’ as well as the “permanent” impairment which occurs after a unilateral decentralization lesion.3”.3’ Thus, nerve terminals of deafferented nerve cells, which do not release neurotransmitter, may adversely affect the function of other nearby normal nerve terminals. Functional

Fig. 3. Diurnal rhythm in urinary 6-hydroxymelatonin, the main metabolite of melatonin, prior to and following a sham, unilateral denervation (Unilateral ICNx) or unilateral decentralization (Unilateral CSTx) operation. The approximate midpoint of the period during which surgery was performed is indicated by the arrows. Dark and clear bars refer to urinary night-time and daytime collections respectively. with the values representing the means for five animals (&S.E.M.) in each group. Collections were performed during the two nights and one day preceding the day of the surgery. Subsequent collections took place both at early (1st and 2nd nights and 1st day) and later (12th and 13th nights and 12th day) time points after surgery. In all three groups, one-way analysis of variance followed by Tukey’s Honesty Sum Difference test revealed a diurnal rhythm prior to surgery. Following sham or unilateral denervation surgery (Unilateral ICNx), each night-time value was higher than its corresponding daytime value (P < 0.05) with the exception of the 1st night after unilateral denervation. Following unilateral decentralization (Unilateral CSTx), none of the night values were significantly higher than the corresponding daytime results with the exception of the 13th night. Two-way analysis of variance followed by Tukey’s Honesty Sum Difference test showed that night-time values obtained both at early (P < 0.05) and at later (P < 0.05) time points after unilateral decentralization were lower than those prior to the lesion.

recovery

in pineal

neurochemical

indices

While NAT activity is a useful index of pineal function, it is possible that changes in its activity are not fully representative of changes in the production of the end product of the biosynthetic pathway, melatonin. Full recovery of NAT activity occurred 1.5 days after unilateral superior cervical ganglionectomy or ICN lesion. 30.3’However, Dornay et al.’ have stated that pineal melatonin does not completely recover to normal, even by 7 days following unilateral ganglionectomy. Interestingly, in their studies’ and in our own,‘6.3’ collateral neuronal sprouting was found to occur after unilateral ganglionectomy or after a unilateral lesion of the ICN, However, while sprouting could be detected 3 days after these lesions, it could not be detected at 1.5 days. Dornay et a/.’ concluded that though NAT recovery can occur without collateral sprouting (e.g. by 1.5 days after unilateral gangtionectomy), full recovery of melatonin biosynthesis cannot take place until collateral sprouting occurs. They speculated that collateral sprouting may lead to a stimulation of HIOMT activity. To date, the effects of unilateral denervation or decentralization on HIOMT activity have not been evaluated directly. Our data (Fig. 2B, C), however, indicate that the levels in the pineal gland of melatonin and of its immediate precursor N-acetylserotonin are the same in animals 1.5, I4 and 21 days

418

G. A.

KUCHEL ef al.

after unilateral denervation as those found in shamoperated animals. While not examined in the present study, we have found, in connection with a different study. that both N-acetylserotonin and melatonin decrease by 70% and 42%. respectively, 8 h after unilateral dcnervation of the pineal glands (Kuchel and Zigmond, unpublished observations). Thus, our data indicate that all three components of the system involved in melatonin synthesis that we have studied decrease after unilateral denervation, but recover to control values within I .5 days. Therefore, the possible role played in the recovery of pineal function by collateral sprouting, which only occurs after 1.5days. remains to be established. Our findings concerning pineal melatonin are also in apparent contradiction to those of two other studies. Reiter rt d.” examined rat pineal melatonin content 4 and 8 h into the night 2 days after unilateral ganglionectomy and found it to be 40-50% lower than that of control animals. Even more surprisingly, they found that NAT activity also failed to fully recover after such a lesion, in contrast to our findings in the present and in two previous studies.30.” In a second paper, Vassilieff rf a/.?’ reported that 8 days after unilateral ganglionectomy the melatonin content measured 4 h into the night was 6575% of that found in sham-operated animals. However, it was not stated whether this difference was statistically significant, nor were later time points in the dark period evaluated. With respect to the comparisons between these studies. it should be noted that we have previously found that the time course ofchanges in NAT activity is similar after unilateral ganglionectomy and after unilateral lesion of the ICN.‘” Functiotid

recorrr.~* in urinary

melatonin

mrtabolites

In the rat, the main metabolite of melatonin, 6-hydroxymelatonin, is excreted primarily in the urineI and shows a robust diurnal rhythm in its urinary levels, with high levels seen at night.” The urinary collection of this compound allows for the study of the time course of recovery of pineal function in an individual animal. The rhythm in urinary 6hydroxymelatonin reflects the melatonin synthesized by the pineal gland since pinealectomy, bilateral denervat~on or bilateral decentralization all lead to nearly complete abolition of the night-time elevations.” The diurnal differences in melatonin production may be even larger than is represented by these measurements. The daytime values may be somewhat overestimated, since it is possible that not all of the 6hydroxymelatonin produced and excreted during a given night is voided during that night and that some is voided during the subsequent day. On the other hand, the fact that in our study night-time collections included urine collected during 1 h of daytime probably had little effect on the night-time values since the data are expressed as ng per collection period.

Sham surgery did not disrupt the normal nighttime 6-hydroxymelatonin excretion at any of the time points examined. After unilateral denervation, the excretion of 6-hydroxymelatonin was low in two of the five animals studied during the first night. However, values in the other three animals were similar to the prelesion measurements. Previous work evaluating NAT activity after unilateral denervation showed greatly reduced activity during the first night3”.” It must be kept in mind that these measurements of NAT activity, unlike those of 6-hydroxymelatonin. were made at a single moment in time, 8 h after surgery. At this time, the nerve terminals of the axotomized nerve are still intact, and thus they may be able to inhibit pineal function through heteroneuronal uptake of NE as described above for the terminals of unilaterally decentralized nerve cells,” It is known that approximately I S--l5 h after ganglionectomy, a process referred to as “degeneration contraction” occurs in the rat eyelid’Y and as “degeneration secretion” in the salivary glands,” with these events lasting for about 12 h. These phenomena are believed to result from the degeneration of postganglionic sympathetic fibers, which leads to the release of NE and a subsequent activation of postsynaptic adrenergic receptors. A similar phenomenon appears to occur in the pineal gland, resulting in an elevation of pineal NAT activityI 20-24 h and elevated pineal melatonin contenti 16-24 h after bilateral ganglionectomy. Interestingly, pineal NE content begins to decrease 12 h after removal of both SCG, with complete disappearance by 27 h.“,‘” It is likely, then. that during degeneration of nerve terminals in the pineal gland, increased NE “release” occurs, resulting in increased melatonjn synthesis and secretion during the latter portion of the first night’s urinary collection. Thus, the urine collected during the first night after unilateral denervation may have included urine both from an early period of impaired and a late period of normal or higher melatonin production. Measurements during subsequent nights following unilateral denervation, times by which degeneration of noradrenergic nerve terminals is complete, show normal night-time excretion of Ghydroxymelatonin, consistent with a full recovery of normal melatonin synthesis. In contrast. unilateral decentralization results in night-time 6-hydroxymelatonin excretion both at early and at later time points after the surgery that is lower than the prelesion night-time values. These latter data support OUT conclusion from n~easurements of pineal indoles that there is an overall functional defect of the system for melatonin synthesis following unilateral decentralization.

The reductions of the night-time levels of pineal melatonin and its principal metabolite, 6-hydroxymelatonin~ suggest that heteroneuronal

Pineal function

after partial

uptake, as discussed above, may result in an inhibition of melatonin synthesis through its effect on NAT activity. However, there are some interesting quantitative differences in the recovery of the three parameters. While all three are lower following unilateral decentralization, the relative recovery is less for NAT activity (2&30% of shamoperated animals) as opposed to melatonin content (40-50%) or 6-hydroxymelatonin excretion (60&70%). Several potential explanations exist for these quantitative differences. NAT has been said to be the main rate-limiting enzyme in the synthesis of melatonin, since it undergoes a dramatic increase in activity by more than 40-fold at night, with HIOMT undergoing little, if any, day/night change.“,” In general, it has been assumed that this dramatic night-time increase in NAT activity is responsible for increased synthesis of N-acetylserotonin, which is then rapidly converted to melatonin by HIOMT through a simple mass action effect.’ However, as shown in Table 1, following adrenergic stimulation, peak melatonin levels are reached before NAT activity has reached its maximum value. These findings suggest that relatively early in the night, following a small initial increase in activity, NAT activity may exceed that of HIOMT, resulting in a build-up of N-acetylserotonin, and the rate of melatonin production later during the night may be limited by some other process, for example by the activity of HIOMT.X.“.29 Thus, peak night NAT activity in control animals may be greater than required to produce normal night-time melatonin synthesis and, following nerve lesions, a 25% recovery in peak night NAT activity may result in a proportionately greater recovery of melatonin synthesis, as indicated by our data (Figs 2C and 3). An additional important consideration is that NAT activity, as measured in vitro here, may not be entirely reflective of the rate of NAT activity in L&O. norepinephrine

sympathetic Implications

419

nerve lesions of findings

for

other neural

systems

The functional recovery in the pineal gland in response to partial sympathetic denervation may represent a useful model for examining the capacity of the central nervous system to recover from losses of nerve cells through disease, injury or aging. Similarly the lack of functional recovery after partial decentralization may be a useful model for examining functional deficits under these conditions. It remains to be established, however, whether heteroneuronal neurotransmitter uptake, such as that which seems to be important after partial decentralization, plays a role in the central nervous system after neural damage. It will also be of interest to determine if heteroneuronal uptake is important, both in the central and peripheral nervous systems in certain intact neural pathways. For example, anatomical proximity of noradrenergic and dopaminergic nerve terminals may exist in the brain, since the neocortex and such structures as the amygdala, the lateral septal nucleus and the interstitial nucleus of the stria terminalis receive both dense noradrenergic and dopaminergic innervations. *’ Interestingly, there is significant overlap in the specificity of the uptake systems of different catecholaminergic nerve terminals for their respective neurotransmitters. Noradrenergic nerve terminals have been shown to be able to actively transport dopamine and dopaminergic nerve terminals can take up norepinephrine.24 Thus, the postsynaptic effectiveness of transmitter released by one of the systems could be modulated by transmitter uptake by terminals of the other system. Acknowledgements-We

thank

Dr John

W. Rowe and Dr

Sanford P. Markey for their support and encouragement. We would also like to thank Claire Baldwin, Hilary Hyatt and Dr Chauncey Bowers for helpful comments on the manuscript. This research was supported by NIH grant NS 17512. R.E.Z. is a recipient of an NIMH Research Scientist Award (MH 00162).

REFERENCES

1. Bowers C. W. and Zigmond R. E. (1980) Electrical stimulation of the cervical sympathetic trunks mimics the effects of darkness on the activity of serotonin: N-acetyltransferase in the rat pineal. Brain Res. 185, 435-O. 2. Bowers C. W. and Zigmond R. E. (1982) The influence of the frequency and pattern of sympathetic nerve activity on serotonin N-acetyltransferase in the rat pineal gland. J. Physiol. 330, 279-296. 3. Bowers C. W., Baldwin C. and Zigmond R. E. (1984) Sympathetic reinnervation of the pineal gland after postganglionic nerve lesion does not restore normal pineal function. J. Neurosci. 4, 2010-2015. 4. Deguchi T. and Axelrod J. (1972) Control of circadian change of seiotonin N-acetyltransferase activity in the pineal organ by the beta-adrenergic receptor. Proc. natn. Acad. Sci. U.S.A. 69, 2547-2550. N-acetyltransferase activity in the pineal. Analyf. 5. Deguchi T. and Axelrod J. (1972) Sensitive assay for serotonin Biochem. 50, 1744179. 6. Delfs U. and Emmelin N. (1974) Degeneration secretion of saliva in the rat following sympathectomy. J. Physiol. 239, 623430. 7. Dornay M., Gilad V. H. and Gilad G. M. (1985) Compensatory changes in contralateral sympathetic neurons of the superior cervical ganglion and in their terminals in the pineal gland following unilateral ganglionectomy. J. Neurosci. 5, 1522-1526. 8. Klein D. C. and Weller J. L. (1970) Indole metabolism in the pineal gland: a circadian rhythm in N-acetyltransferase. Science 169, 1093-1095. 9. Klein D. C. and Rowe J. (1970) Pineal gland in organ culture. I. Inhibition by harmine of serotonin-‘4C oxidation, accompanied by stimulation of melatonin-14C production. Molec. Pharmac. 6, 164171. 10. Klein D. C., Weller J. L. and Moore R. Y. (1971) Melatonin metabolism: neural regulation of pineal serotonin:acetylcoenzyme A N-acetyltransferase activity. Proc. nafn. Acad. Sci. U.S.A. 68, 3107T3110.

420

G. A. KU~HELrf ai.

II Klein D. C., Auerbach D., Namboodiri M. A. A. and Wheler G. H. T. (1981) lndole metabolism in the mammalian pineal gland. In Pineui Giand: I. Anatomy and Biochetnbtry (ed. Reiter R.), pp. 199-227. CRC Press, Boca Raton, FL. 12. Klein D. C., Smoot R., Weller J. L., Higa S., Markey S. I?., Creed G. J. and Jacobowitz D. M. (1983) Lesions of the paraventricular nucleus area of the hypothalamus disrupt the suprachiasmatic-spinal cord circuit in the meiatonin rhythm generating system. Brain Res. Bull. 10, 647452. 13. Klein D. C. (1985) Photoneuronal regulation of the mammalian pineal gland. In Photoperiodism. Melatonin and the Pineal. Ciba Foundation Symposium 117. Pitman, London. 14. Kopin I. J.. Pare C. M. B.. Axelrod J. and Weissbach H. (1961) The fate of melatonin in animals. J. biol. Chem. 236, 3072.-3075. 15. Kuchel G. A., Sherman R. L. and Zigmond R. E. (I 989) Recovery to normal of pineal melatonin content and urinary melatonin metabolites following unilateral denervation but not unilateral decentralization of the pineal gland. SOC. Neurosci. Abstr. 15, 781. 16. Lingappa J. R. and Zigmond R. E. (1987) A histochemical study of the adrenergic innervation of the rat pineal gland: evidence for overlap of the innervation from the two superior cervical ganglia and for sprouting following unilateral denervation. Neuroscience 21, 893-902. 17. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. L. (1951) Protein measurement with the Fohn phenol reagent. J. biol Chem. 193, 265-275. 18. Luchelli-Fortis M. A., St&no F. J. E. and Perec C. J. (1982) ~gene~ation activity of the pineal gland after sympathetic denervation. Nuunyn-Scfimiedeberg’.F Arch, Pharmac. 321, 298-301. 19. Lundberg D. (1969) Adrenergic neuron blockers and transmitter release after sympathetic denervation studied in the conscious rat. Acra physiol. scond. 75, 415 426. 70. Mefford I. N., Chang P.. Klein D. C., Namboodiri M. A. A.. Sugden D. and Barchas J. (1983) Reciprocal day/night relationship between serotonin oxidation and ~-acety~ation products in the rat pineal gland. Endocrinology 113, )582--l 586. 21. Moore R. Y. and Bloom F. E. (1979) Central catecholamine neuron systems: anatomy and physiology of the norepinephrine and epinephrine systems. A. Rev. Neurowi. 2, 113-168. 12. Morgan W. W. and Hansen J. T. (1978) Time course of the disappearance of pineal noradrenaline following superior cervical ganglionectomy. E.rpi Brain Rex. 32, 429434. 23. Reiter R. J., Rudeen P. K., Banks A. F. and Rollag M. D. (1979) Acute effects of unilateral or bilateral superior cervical ganglionectomy on rat pineal N-acetyltransferase activity and melatonin content. Experientia 35, 691-692. 24. Snyder S. H. and Coyle J. T. (1969) Regional differences in [3H]norepinephrine and [lH]dopamine uptake into rat brain homogenates. J. Pharmac. erp. Ther. 165, 78 86. 2s. Tetsuo M., Markey S. P., Colburn R. W. and Kopin I. J. (198f) Quantitative analysis of 6-hydroxymelatonin in human urine by gas chromatography -negative chemical ionization spectrometry. Anaiyt. Biochem. 110, 208-215. 26. Vacas M. 1.. Keller Sarmiento M. I., Pereyra E. and Cardinali D. P. (1982) Early changes in CAMP and melatonin levels of rat pineal gland after superior cervical ganglionectomy. Neuroendocrinol. Lett. 4, 267-271. 27. Vassilieff V. S., Sourkes T. L. and Anderson G. M. (1982) Superior cervical ganglionectomy: effect on indolic compounds in rat pineal gland. J. Neural Trans. 55, 45-52. 28. Wurtman R. J., Axeirod J. and Kelly D. E. (1968) The Pineal Gland. Academic Press, New York. 29. Wurtman R. J. and Moskowitz M. A. (1977) The pineal organ. New Engl. J. Med. 296, 1329-1333, 1383-1386. 30. Zigmond R. E.. Baldwin C. and Bowers C. W. (1981) Rapid recovery of function after partial denervation of the rat pineal gland suggests a novel mechanism for neural plasticity. Proc. natn. Acad. Sci. U.S.A. 78, 3959-3963. 31. Zigmond R. E., Baldwin C. and Bowers C. W. (1985) Rapid recovery of function after partial denervation: a possible role for heteroneuronal uptake of transmitter in modulating synaptic efficacy. J. Xeurosci. 5, 142-150. (Accepted 23 April 1990)