Physiology&Behavior,Vol. 49, pp. 533-537. ©PergamonPressplc, 1991. Printedin the U.S.A.
0031-9384/91 $3.00 + .00
Lateral Hypothalamic Regulation of Circadian Rhythm Phase N. GOODLESS-SANCHEZ,* R. Y. MOORE1" AND L. P. MORIN*:~ 1
*Department of Psychiatry ~-Department of Neurology and Department of Neurobiology and Behavior and SDepartment of Psychology, State University of New York, Stony Brook Received 30 July 1990
GOODLESS-SANCHEZ, N., R. Y. MOORE AND L. P. MORIN. Lateral hypothalamic regulation of circadian rhythmphase. PHYSlOL BEHAV 49(3) 533-537, 1991.--The suprachiasmaticnuclei generate a circadianrhythm which can be described by period, phase and amplitudevariables. Evidenceis accumulatingthat the three descriptors of circadianrhythmicitycan be modulated independentlyby several brain structures. This report describes the effects of lateral hypothalamic(LHA) damage on control of period, phase and amplitude of the hamster locomotor rhythm. Adult male hamsters received bilateral electrolytic lesions of the far lateral LHA. These lesions had no effect on the circadianperiod in constant dim, but significantlyadvancedthe onset of nocturnal wheel running and lengthenedthe duration of the activity phase. Rate of reentrainmentafter a 6-h phase advance or delay was not affected by the lesion. Rhythm amplitude, as indicatedby the numberof wheel revolutionsper day, was not affected by the lesions. The results support the view that differentbrain regions can exert independentmodulatorycontrol over the basic circadianrhythm generated by the suprachiasmaticnucleus. Electrolytic lesion
Locomotorrhythm
Entrainment Free-running period
CIRCADIAN rhythmicity is an expression of a fundamental adaptation of an animal's physiology and behavior to a fluctuating environment. This adaptation necessarily requires the circadian rhythm system to be responsive to periodic environmental cues. The most potent circadian rhythm synchronizing stimulus is the daily photoperiod. Exploration of the brain for the location of the 24-h clock was guided by the expectation that a direct neuronal connection existed between phototransducers in the retina and this clock. The hypothalamic suprachiasmatic nucleus (SCN) (12, 20, 22, 24) is known to function as a circadian clock. It receives photic information through a retinohypothalamic tract (RHT) (11,12) necessary for entrainment to the light-dark (LD) cycle (10). Emphasis has been placed on the role of the RHT in entrainment, but the retinal ganglion cells projecting through the RHT also project to the lateral geniculate region (15). It is now known that a secondary visual projection to the SCN arises from the intergeniculate leaflet (IGL) (2, 7, 25). The existence of this geniculohypothalamic tract (GHT) suggests that a pathway other than the RHT may alternatively or redundantly provide photoperiod cues to the circadian clock (6, 9, 13, 17). Each circadian rhythm can be described according to its parameters of period, phase, and amplitude. Period refers to the length, in units of time, of one circadian cycle. Phase refers to a particular position in the cycle. This is usually measured relative to an arbitrary reference point, such as lights off. Amplitude describes the magnitude of the rhythm, such as number of dally
Amplitude
Activityphase
wheel revolutions. Little effort has been taken to evaluate the relative degree of independent neural control of the three variables. In fact, rhythm regulation has been generally considered to be the province of the SCN with deviations from normality reflecting changes in SCN function. However, a number of experiments show that this is clearly not always the case. Rather, non-SCN areas which are part of a complex circuit modulating circadian rhythmicity may play an integral role in the generation and regulation of circadian rhythms. Destruction of the IGL blocks the period-lengthening effect of constant light (17), reduces the amplitude of phase shifts in response to a light pulse (17), induces advances in the phase of activity onset (9,16) and causes a modest reduction in the amount of daily running (8,9). When the dorsal raphe serotonergic projections to the basal forebraln are destroyed by a selective neurotoxin, phase of activity onset is advanced and the daily activity phase greatly expanded without an effect on the amount of dally running or on the circadian period in constant dark (23). In addition to the pathways described above, the RHT also sends projections to the lateral hypothalamic area (LHA) with the largest terminal plexus in the anterior hypothalamic component of the LHA (11,18). The functional significance of this connection is not known, but its presence suggests that the LHA may receive photoperiod information and thereby regulate one or more of the parameters by which a rhythm can be defined. Several studies support this suggestion. Amplitude and the duration of the active
~Requests for reprints should be addressed to Dr. Lawrence P. Morin, Department of Psychiatry, Health Science Center, SUNY, Stony Brook, NY 11794.
533
534
GOODLESS-SANCHEZ, MOORE AND MORIN
phase of wheel-running behavior of hamsters have been reported to increase after lesions directed at the primary optic tract (POT) (19,21). These lesions could have included the LHA. When parasagittal knife cuts adjacent to the SCN damage the medial LHA, but not the POT, visual assessment of actogram records also indicates increased locomotor behavior. This increase in running behavior is more apparent than real, and is an artifact of the traditional actogram method used for rhythm visualization (8). In reality, there is a reduction in activity associated with an expansion of the time over which it is distributed. Phase of activity onset is advanced and more time is spent running, but at lower rates (8). Gladfelter and co-workers (3,4) have demonstrated that either damage to the medial LHA or destruction centered in the LHA produces profound reductions in rat locomotor behavior. Additional experiments showed that rats with far-lateral hypothalamic lesions do not exhibit permanent decreases in wheel-running activity. In fact, the relative levels of wheel-running activity of these rats tend to be somewhat greater than those of controls (5). Thus, period, phase and amplitude of the hamster's locomotor rhythm appear to be under multiple controls and the following study was undertaken to specifically address the role of the LHA in the regulation of these parameters of hamster locomotor rhythmicity. METHOD
Twenty-five adult male golden hamsters (Charles River) were individually housed in translucent polypropylene cages (48 x 27 x 20 cm), each equipped with a 17-cm diameter running wheel. Each revolution of the wheel closed a microswitch that was recorded by a computer. Wheel revolutions were accumulated in 5-min bins and plotted in standard actogram format. The cages were placed in light- and temperature-controlled rooms with food and water continuously available. All data were collected after the hamsters had been entrained to a 14-h light: 10-h dark cycle (lights off at 2000 h) for at least one month. Light intensity was about 10 ixW/cmz, which corresponds to about 29 lux.
Lesions At a mean body weight of 115 g, animals were anesthetized with sodium pentobarbital (10 mg/100 g b.wt.), placed in a stereotaxic instrument and given electrolytic lesions in the LHA. Lesions were made by passing a direct current of 1.5 mA for 7 s through an insect pin insulated with epoxylite except for the tip (0.3 mm). A ground electrode was attached to a needle which was subcutaneously inserted into the animals' flank. Approximately one week later, the procedure was repeated on the same animals (LHA group; N = 15), but the electrode was placed in the contralateral LHA. In addition, control animals (SH group; N = 10) each received a bilateral sham lesion during which the skull was opened, the dura incised and the electrode lowered to the LHA and retracted without passing current. Following the second surgery, animals remained in the original photoperiod for approximately 8 weeks. Animals were then placed in constant dim illumination (DD). The light intensity was about 0.2 la,W/cm2, which corresponds to about 1.2 lux. Approximately 6 weeks later, animals were returned to the original LD cycle for about 5 weeks. The photoperiod was then advanced 6 h (remaining LD 14:10). The animals were in this photoperiod for about 6 weeks, then were phase delayed for 6 h remaining in LD 14:10 for 5 additional weeks.
Histology At the experiment's end, all lesioned and sham-operated con-
trol animals were given an overdose of sodium pentobarbital and transcardially perfused with physiological saline followed by 454 paraformaldehyde containing sodium periodate and lysine. Brains were postfixed for 3 h and then equilibrated through an ascending series of sucrose solutions, ending at 30°k. Verification of lesion placement was determined from serial sections cut through the LHA at 30 txm on a freezing microtome. Adjacent sections were either stained with cresyl violet or retained to be processed for serotonin immunoreactivity. For the latter, sections were incubated in goat serum for 2 h (5% serum in phosphate buffered saline with 0.3% Triton X-100) and then incubated for 7 days at 4°C with antiserum to serotonin (Immunonuclear; 1:1500 in buffer with 0.3% Triton X-100 and 2% goat serum). Time delays imposed by difficulties with the antiserum-caused tissue deterioration reduced the intensity of the reaction product, limiting the usefulness of the serotonin immunohistochemistry.
Data Analysis Three observers, blind to the experimental treatment of the animals, independently evaluated the actograms with regard to the phase angle of entrainment and the duration of the activity phase. Analysis was based on the average of the three estimates. The average activity onsets and offsets were determined by a line visually fitted to 20 days of actogram data beginning 25 days before surgery and 10 days following the second surgery. Both activity onset and activity offset are expressed relative to the time of lights off. Duration of the activity phase was calculated as the interval from activity onset to activity offset. The rate of reentrainment was evaluated for the phase advance and the phase delay manipulations. A line was visually fitted through the stable activity onsets preceding each phase shift and the average number of days to accomplish a 6-h shift relative to the preshift phase of each animal was calculated. The period of the free-running rhythm in constant dim illumination was calculated from a line eye-fitted through 20 days of activity onsets beginning about 10 days after the change to constant dim. The effect of the LHA lesion on the amount of wheel-running behavior was evaluated by calculating the number of wheel revolutions per day for a 7-day period beginning 10 days prior to surgery and comparing this value with the average daily number of revolutions during a 7-day period beginning 20 days following the second surgery. RESULTS
Histology Histological examination of serial sections through each lesion-bearing brain revealed three subsets (N =5 each) that differed slightly according to lesion placement. One group had lesions that were bilaterally symmetrical immediately above the POT (Fig. 1A). The second group had bilaterally symmetrical lesions that were dorsal to the sites of group 1 (Fig. 1B), The third group had lesions that were asymmetric either in the rostrocaudal direction or the dorsoventral direction, tending to be a mixture of the other two lesion types. The typical symmea'ical ventral lesion (Fig. IA) included a region boundexl by the zona incerta dorsally, the internal capsule laterally and the fornix medially. They were centered in the rostral tuberal hypothalamus and extended rostrally into the anterior hypothalamic region and caudally to the level of the mid-ventromedial nucleus. The more dorsal symmetrical lesions were located in approximately the same rosa:o-caudal position, but were centered on the medial zona incerta with damage extending into the ventromedial thalamic nucleus dorsally and into the LHA ventrally (Fig. 1B).
LATERAL HYPOTHALAMUS AND RHYTHMS
535
0.3
j,j
0.2
Ii-
"10
m
J9
--, Gel
-48
v
L..
r-
J J
// // //
v
0.1 I-Ld I/) Z -0.0 O
/ / J f J f JJ
I
i.-
.i7
tu ol
J6
o
/,j
"I" - 0 . 1 n _J
,ec
-15
/ /
-0.2
-,4
/ / J J //
-0.5
-r ._.1
//
"~
-,3
J i //
FIG. 2. Left: Time of alpha (activity phase) onset measured relative to lights off. Right: Time of alpha offset. Open bars--sham-operated group; hatched bars--LHA-lesioned group. *The two groups differ, p
0031-9384/91 $3.00 + .00
Lateral Hypothalamic Regulation of Circadian Rhythm Phase N. GOODLESS-SANCHEZ,* R. Y. MOORE1" AND L. P. MORIN*:~ 1
*Department of Psychiatry ~-Department of Neurology and Department of Neurobiology and Behavior and SDepartment of Psychology, State University of New York, Stony Brook Received 30 July 1990
GOODLESS-SANCHEZ, N., R. Y. MOORE AND L. P. MORIN. Lateral hypothalamic regulation of circadian rhythmphase. PHYSlOL BEHAV 49(3) 533-537, 1991.--The suprachiasmaticnuclei generate a circadianrhythm which can be described by period, phase and amplitudevariables. Evidenceis accumulatingthat the three descriptors of circadianrhythmicitycan be modulated independentlyby several brain structures. This report describes the effects of lateral hypothalamic(LHA) damage on control of period, phase and amplitude of the hamster locomotor rhythm. Adult male hamsters received bilateral electrolytic lesions of the far lateral LHA. These lesions had no effect on the circadianperiod in constant dim, but significantlyadvancedthe onset of nocturnal wheel running and lengthenedthe duration of the activity phase. Rate of reentrainmentafter a 6-h phase advance or delay was not affected by the lesion. Rhythm amplitude, as indicatedby the numberof wheel revolutionsper day, was not affected by the lesions. The results support the view that differentbrain regions can exert independentmodulatorycontrol over the basic circadianrhythm generated by the suprachiasmaticnucleus. Electrolytic lesion
Locomotorrhythm
Entrainment Free-running period
CIRCADIAN rhythmicity is an expression of a fundamental adaptation of an animal's physiology and behavior to a fluctuating environment. This adaptation necessarily requires the circadian rhythm system to be responsive to periodic environmental cues. The most potent circadian rhythm synchronizing stimulus is the daily photoperiod. Exploration of the brain for the location of the 24-h clock was guided by the expectation that a direct neuronal connection existed between phototransducers in the retina and this clock. The hypothalamic suprachiasmatic nucleus (SCN) (12, 20, 22, 24) is known to function as a circadian clock. It receives photic information through a retinohypothalamic tract (RHT) (11,12) necessary for entrainment to the light-dark (LD) cycle (10). Emphasis has been placed on the role of the RHT in entrainment, but the retinal ganglion cells projecting through the RHT also project to the lateral geniculate region (15). It is now known that a secondary visual projection to the SCN arises from the intergeniculate leaflet (IGL) (2, 7, 25). The existence of this geniculohypothalamic tract (GHT) suggests that a pathway other than the RHT may alternatively or redundantly provide photoperiod cues to the circadian clock (6, 9, 13, 17). Each circadian rhythm can be described according to its parameters of period, phase, and amplitude. Period refers to the length, in units of time, of one circadian cycle. Phase refers to a particular position in the cycle. This is usually measured relative to an arbitrary reference point, such as lights off. Amplitude describes the magnitude of the rhythm, such as number of dally
Amplitude
Activityphase
wheel revolutions. Little effort has been taken to evaluate the relative degree of independent neural control of the three variables. In fact, rhythm regulation has been generally considered to be the province of the SCN with deviations from normality reflecting changes in SCN function. However, a number of experiments show that this is clearly not always the case. Rather, non-SCN areas which are part of a complex circuit modulating circadian rhythmicity may play an integral role in the generation and regulation of circadian rhythms. Destruction of the IGL blocks the period-lengthening effect of constant light (17), reduces the amplitude of phase shifts in response to a light pulse (17), induces advances in the phase of activity onset (9,16) and causes a modest reduction in the amount of daily running (8,9). When the dorsal raphe serotonergic projections to the basal forebraln are destroyed by a selective neurotoxin, phase of activity onset is advanced and the daily activity phase greatly expanded without an effect on the amount of dally running or on the circadian period in constant dark (23). In addition to the pathways described above, the RHT also sends projections to the lateral hypothalamic area (LHA) with the largest terminal plexus in the anterior hypothalamic component of the LHA (11,18). The functional significance of this connection is not known, but its presence suggests that the LHA may receive photoperiod information and thereby regulate one or more of the parameters by which a rhythm can be defined. Several studies support this suggestion. Amplitude and the duration of the active
~Requests for reprints should be addressed to Dr. Lawrence P. Morin, Department of Psychiatry, Health Science Center, SUNY, Stony Brook, NY 11794.
533
534
GOODLESS-SANCHEZ, MOORE AND MORIN
phase of wheel-running behavior of hamsters have been reported to increase after lesions directed at the primary optic tract (POT) (19,21). These lesions could have included the LHA. When parasagittal knife cuts adjacent to the SCN damage the medial LHA, but not the POT, visual assessment of actogram records also indicates increased locomotor behavior. This increase in running behavior is more apparent than real, and is an artifact of the traditional actogram method used for rhythm visualization (8). In reality, there is a reduction in activity associated with an expansion of the time over which it is distributed. Phase of activity onset is advanced and more time is spent running, but at lower rates (8). Gladfelter and co-workers (3,4) have demonstrated that either damage to the medial LHA or destruction centered in the LHA produces profound reductions in rat locomotor behavior. Additional experiments showed that rats with far-lateral hypothalamic lesions do not exhibit permanent decreases in wheel-running activity. In fact, the relative levels of wheel-running activity of these rats tend to be somewhat greater than those of controls (5). Thus, period, phase and amplitude of the hamster's locomotor rhythm appear to be under multiple controls and the following study was undertaken to specifically address the role of the LHA in the regulation of these parameters of hamster locomotor rhythmicity. METHOD
Twenty-five adult male golden hamsters (Charles River) were individually housed in translucent polypropylene cages (48 x 27 x 20 cm), each equipped with a 17-cm diameter running wheel. Each revolution of the wheel closed a microswitch that was recorded by a computer. Wheel revolutions were accumulated in 5-min bins and plotted in standard actogram format. The cages were placed in light- and temperature-controlled rooms with food and water continuously available. All data were collected after the hamsters had been entrained to a 14-h light: 10-h dark cycle (lights off at 2000 h) for at least one month. Light intensity was about 10 ixW/cmz, which corresponds to about 29 lux.
Lesions At a mean body weight of 115 g, animals were anesthetized with sodium pentobarbital (10 mg/100 g b.wt.), placed in a stereotaxic instrument and given electrolytic lesions in the LHA. Lesions were made by passing a direct current of 1.5 mA for 7 s through an insect pin insulated with epoxylite except for the tip (0.3 mm). A ground electrode was attached to a needle which was subcutaneously inserted into the animals' flank. Approximately one week later, the procedure was repeated on the same animals (LHA group; N = 15), but the electrode was placed in the contralateral LHA. In addition, control animals (SH group; N = 10) each received a bilateral sham lesion during which the skull was opened, the dura incised and the electrode lowered to the LHA and retracted without passing current. Following the second surgery, animals remained in the original photoperiod for approximately 8 weeks. Animals were then placed in constant dim illumination (DD). The light intensity was about 0.2 la,W/cm2, which corresponds to about 1.2 lux. Approximately 6 weeks later, animals were returned to the original LD cycle for about 5 weeks. The photoperiod was then advanced 6 h (remaining LD 14:10). The animals were in this photoperiod for about 6 weeks, then were phase delayed for 6 h remaining in LD 14:10 for 5 additional weeks.
Histology At the experiment's end, all lesioned and sham-operated con-
trol animals were given an overdose of sodium pentobarbital and transcardially perfused with physiological saline followed by 454 paraformaldehyde containing sodium periodate and lysine. Brains were postfixed for 3 h and then equilibrated through an ascending series of sucrose solutions, ending at 30°k. Verification of lesion placement was determined from serial sections cut through the LHA at 30 txm on a freezing microtome. Adjacent sections were either stained with cresyl violet or retained to be processed for serotonin immunoreactivity. For the latter, sections were incubated in goat serum for 2 h (5% serum in phosphate buffered saline with 0.3% Triton X-100) and then incubated for 7 days at 4°C with antiserum to serotonin (Immunonuclear; 1:1500 in buffer with 0.3% Triton X-100 and 2% goat serum). Time delays imposed by difficulties with the antiserum-caused tissue deterioration reduced the intensity of the reaction product, limiting the usefulness of the serotonin immunohistochemistry.
Data Analysis Three observers, blind to the experimental treatment of the animals, independently evaluated the actograms with regard to the phase angle of entrainment and the duration of the activity phase. Analysis was based on the average of the three estimates. The average activity onsets and offsets were determined by a line visually fitted to 20 days of actogram data beginning 25 days before surgery and 10 days following the second surgery. Both activity onset and activity offset are expressed relative to the time of lights off. Duration of the activity phase was calculated as the interval from activity onset to activity offset. The rate of reentrainment was evaluated for the phase advance and the phase delay manipulations. A line was visually fitted through the stable activity onsets preceding each phase shift and the average number of days to accomplish a 6-h shift relative to the preshift phase of each animal was calculated. The period of the free-running rhythm in constant dim illumination was calculated from a line eye-fitted through 20 days of activity onsets beginning about 10 days after the change to constant dim. The effect of the LHA lesion on the amount of wheel-running behavior was evaluated by calculating the number of wheel revolutions per day for a 7-day period beginning 10 days prior to surgery and comparing this value with the average daily number of revolutions during a 7-day period beginning 20 days following the second surgery. RESULTS
Histology Histological examination of serial sections through each lesion-bearing brain revealed three subsets (N =5 each) that differed slightly according to lesion placement. One group had lesions that were bilaterally symmetrical immediately above the POT (Fig. 1A). The second group had bilaterally symmetrical lesions that were dorsal to the sites of group 1 (Fig. 1B), The third group had lesions that were asymmetric either in the rostrocaudal direction or the dorsoventral direction, tending to be a mixture of the other two lesion types. The typical symmea'ical ventral lesion (Fig. IA) included a region boundexl by the zona incerta dorsally, the internal capsule laterally and the fornix medially. They were centered in the rostral tuberal hypothalamus and extended rostrally into the anterior hypothalamic region and caudally to the level of the mid-ventromedial nucleus. The more dorsal symmetrical lesions were located in approximately the same rosa:o-caudal position, but were centered on the medial zona incerta with damage extending into the ventromedial thalamic nucleus dorsally and into the LHA ventrally (Fig. 1B).
LATERAL HYPOTHALAMUS AND RHYTHMS
535
0.3
j,j
0.2
Ii-
"10
m
J9
--, Gel
-48
v
L..
r-
J J
// // //
v
0.1 I-Ld I/) Z -0.0 O
/ / J f J f JJ
I
i.-
.i7
tu ol
J6
o
/,j
"I" - 0 . 1 n _J
,ec
-15
/ /
-0.2
-,4
/ / J J //
-0.5
-r ._.1
//
"~
-,3
J i //
FIG. 2. Left: Time of alpha (activity phase) onset measured relative to lights off. Right: Time of alpha offset. Open bars--sham-operated group; hatched bars--LHA-lesioned group. *The two groups differ, p
