N euroendocrinology 29 : 119-131 (1979)
Effects of Destruction of the Suprachiasmatic Nuclei on the Circadian Rhythms in Plasma Corticosterone, Body Temperature, Feeding and Plasma Thyrotropin1 K. Abe../. Kroning, M. A. Greer and V. Criiehlow Departments of Anatomy and Medicine, University of Oregon Health Sciences Center, Portland, Oreg.
Key Words. Circadian rhythms • Suprachiasmatic nuclei • Corticosterone • Temperature • Feeding • Thyrotropin
Because destruction of the suprachiasmatic nuclei (SCN) in rodents abolishes several circadian rhythms, it has been postulated that ' This work was supported by USPHS grants AM16794 and AM-01447 from the NIAMDD. NIH. Received: September 22, 1978 Accepted: January 22, 1979
these nuclei serve as a major pacemaker for the generation of such rhythms [I. 8], Moore and Eichler [9] were the first to report that ablation of the SCN in the rat disrupts the circadian rhythm in pituitary-adrenal func tion. However, these investigators used adre nal concentrations of corticosterone in groups of rats killed at four times during the 24-hour
Downloaded by: Kings's College London 137.73.144.138 - 10/18/2017 7:38:27 AM
Abstract. To study the role of the suprachiasmatic nuclei (SCN) in generating circadian rhythms in female rats, lesions were placed in the SCN or in the medial preoptic (PO) region. Serial blood sampling at 4-hour intervals at 3 and 13 weeks after surgery indicated that com plete SCN destruction abolished rhythmic fluctuations in plasma corticosterone levels in indi vidual rats. Partial destruction produced less interference, while medial PO lesions that spared the SCN were without effect. Similar effects were noted on daily changes in body temperature at 10 weeks after surgery; however, some rats showed evidence of dissociation of these two rhythmic functions in that some lesions appeared to affect one and not the other. In ancillary studies, it was found that all lesioned groups showed nocturnal feeding patterns similar to those of the controls and that the diurnal pattern in plasma thyrotropin (TSH) levels was altered by complete destruction of the SCN. These data suggest that the SCN are essential for the circadian rhythms in pituitary-adrenal function and body temperature and that separate pacemakers may be present in these nuclei for these two periodic functions. The SCN may also control rhythmic TSH secretion, but ihese nuclei and the medial PO region do not appear essential for nocturnal feeding.
light-dark cycle to assess the effects of the SCN lesions, so it was impossible to charac terize the nature of the deficit produced. Thus, it could not be determined whether the lesions abolished rhythms in individual rats or pro duced free-running rhythms that were asyn chronous between animals. The former would be expected if the influence of a master oscilla tor were eliminated and the latter if structures involved in entrainment were compromised. Subsequent to the studies of Moore and Eichler. Krieger [5] introduced a procedure for collecting serial blood samples for studying 24-hour patterns of plasma corticosterone levels in individual rats. This approach, which permits better characterization of the effects of experimental manipulationson this rhythm, was used to demonstrate that surgical isola tion of the medial basal hypothalamus [14] and placement of large lesions in the medial preoptic-anterior hypothalamic region [13] abolish the pituitary-adrenal rhythm, while optic enucleation [16] and exposure to con stant light [15] cause free-running. Krieger et at. [6] recently used serial blood sampling to demonstrate the loss of the normal 24-hour pattern of corticosterone levels after placement of SCN lesions, but they did not address the question of the effects of the lesions in indi vidual rats. Raisman and Brown-Grant [12] found that SCN lesions interfered with nor mal morning to evening differences in plasma corticosterone concentrations, but the effects on the rhythm in individual rats were not determined. The aims of the present experiment were to determine the nature of the disruption of rhythmic pituitary-adrenal function produced by ablation of the SCN and to compare the effects of such ablation with those produced by lesions placed in the adjacent medial preoptic (PO) region. To monitor the effects of
Abe K roning G reer-Critchlow
SCN lesions on other rhythmic functions, daily changes in food intake, body tempera ture and plasma concentrations of thyrotro pin (TSH) were also measured at various in tervals after surgery.
Materials and Methods The adult female rats (Charles River, CD) used in these experiments weighed 240-320 g at surgery. The) were maintained under conditions of controlled light (04:00-18:00 h) and temperature (26±2 C) for at least 3 weeks prior to surgery. Purina Laboratory Chow and tap water were available at all times. Thi rats were housed 4/cage until 3 days before each stud) at which time they were placed in individual cages. One group of rats received small, anodal DC le sions aimed at the SCN. The lesions were made by passing 1,0 mA for 10 sec through a platinum electrodi shaped as an inverted T. The crossbar on the electrodi. was 1 mm long and uninsulated. The electrode was lowered through the midline and third ventricle to the dorsum of the optic chiasm with the crossbar in the sagittal plane. The electrode was then rotated 90 so that the crossbar was transversely placed in the region of the SCN. Lesions were made at 4 rostrocaudal por tions by moving the electrode caudally 0.4 mm between each lesion, starting at 2.6 mm anterior to the bregma with the tooth bar located 5.0 mm above the interaural plane. Another group of rats received bilateral DC lesions in the medial PO region with a platinum electrode that was insulated except for I mm at the tip. A current of 1.5 mA was applied for 20 sec with the electrode tip located 2.0 mm anterior to bregma and 0.5 mm on each side of midline. Intact and sham-lcsioned groups served as controls. The sham lesion procedure con sisted of lowering the electrode used for the medial PO lesions 6.0 mm below the cortical surface at 2.0 mm anterior to bregma and withdrawing it without passing current. Treatments were assigned according to a completely randomized design. Plasma Corticosterone Concentrations To determine the effects of the lesions on the circadian rhythm of pituitary-adrenal function, plasma corticosterone concentrations were measured in serial blood samples obtained from a tail vein, using a modi
Downloaded by: Kings's College London 137.73.144.138 - 10/18/2017 7:38:27 AM
120
Suprachiasm atic Nuclei and C ircadian R hythm s
Foot! Intake The daily patterns of food intake were determined at 5-6 weeks after surgery. The rats were placed in individual cages for I week prior to the study and food intake was measured at 6-hour intervals for 24 h, starting at 06:00 h. Body Temperature Body temperatures were monitored at 4-hour inter vals for 44 h at 10 weeks after surgery. As with the corticosterone experiments, the rats were placed in individual cages for 3 days and the animal room was locked 18 h before the study. A rectal Tele-thermom eter probe (Yellow Springs Instruments, Ohio) was inserted 5 cm and temperatures recorded 2.5 min later. Plasma TSH Serial blood samples were collected from tail veins as described above at 18 weeks after surgery for deter mining the effects of the lesions on plasma TSH levels. Because larger volumes of blood (0.8 ml) were needed for measuring TSH, only 3 blood samples were col lected over a period of 12 h, at 06:00, 12:00 and 18:00 h. Plasma TSH was measured by radioimmuno assay, as previously described [2], using materials sup plied by the Rat Pituitary Hormone Distribution Pro gram of the NIAMDD and Dr. A. Parlow. Intra- and interassay coefficients of variation were less than 15%.
The rats were autopsied at 20 weeks after surgery. The brains were removed, fixed and processed for histological examination of lesions. Adrenals, ovaries, uteri and pituitaries were removed and weighed. One- and two-way analyses of variance (ANOVA) for repeated measures [17] were used for analyzing group data derived from serial sampling. Because the number of samples and the duration of sampling were appropriate in the studies involving corticosterone levels and body temperature, the criteria of Wilson and Greer [15] were used with minor modification to evaluate the presence or absence of rhythmicity in individual rats. These criteria offer an objective method for determining whether 24-hour patterns are repeated on 2 consecutive days of observation, i.e.. whether there is evidence of rhythmicity. These criteria are: (I) the highest values on both days occur at the same time of day plus or minus 4 h: (2) there is a positive correlation coefficient between values on both days with time: (3) the correlation coefficient is within 2 SD of the mean coefficient of the intact control group, and (4) the ratio of the maximum to minimum values is within 2 SD of the mean ratio demonstrated by intact controls on both days. The fourth criterion was modified slightly in order to remove the restraint on the extent to which an excursion in a value might ex ceed those of the intact controls. Thus, the maximum to minimum ratio could be more, but not less, than 2 SD above the mean for the intact controls and still be judged rhythmic. This modification prevented label ing the patterns of some rats as arrhythmic because the amplitudes of the daily fluctuations were unusually high. A pattern which satisfied all four criteria was considered rhythmic. Organ and body weights were compared with one-way ANOVA and the NewmanKeuls test.
Results Histology o f Lesions Histological examination showed that most of the lesions aimed for the SCN produced incomplete destruction of these nuclei. In ad dition, some of those aimed for the medial PO region destroyed the SCN, some involved the SCN partially while others spared these nuclei completely. Therefore, the following
Downloaded by: Kings's College London 137.73.144.138 - 10/18/2017 7:38:27 AM
fication of the procedure of Krieger [6] described pre viously [14]. The samples were collected at 4-hour intervals for 44 h at 3 and 13 weeks after surgery, and the rats were unanesthetized but gently restrained dur ing sampling. The animals were placed in individual cages 3 days before these studies, and the room was locked 18 h before sampling began to minimize in advertent environmental disturbances. The rats were removed from their cages and taken individually to an adjacent laboratory where 0.3 ml of blood was col lected less than 3 min from the time of cage opening. Blood was collected in EDTA-rinsed pipettes and centrifuged; the separated plasma was stored (1 5 C) for eventual determination of corticosterone concen trations using the microfluorometric method of Click el al. [4]. Correction for residual fluorescence, approxi mately 6 //g 100 ml of plasma, was not made. Sensitiv ity of the assay is such that 0.5 ng of corticosterone can be detected. The intraassay coefficient of variation in the normal range of the assay was 6% and the inter assay variation was 5%.
121
122
Abe K roning/G rcer/C ritchlcnv
shows photomicrographs of two lesions which completely destroyed the SCN. one which involved primarily these nuclei and one which included portions of the medial PO region. Partial SCN Destruction. 16 rats had le sions which left portions of the SCN morpho logically intact. This group was the most heterogenous in that there was considerable variation in the extent to which these nuclei were involved. However, all rats in this group had portions of the caudal half of the SCN present unilaterally or bilaterally.
Fig. 1. Photomicrographs of two lesions which completely destroyed the SCN. A A small lesion which involved primarily the SCN and optic chiasm. B A larger lesion which included the medial PO region in addition to the SCN.
new experimental groups were formed based on the extent to which the SCN were involved. Complete SCN Destruction. 6 rats had le sions which completely destroyed the SCN: 3 of these lesions were small and minimally involved adjacent brain structures, while 3 were larger and impinged upon the adjacent medial PO and anterior hypothalamic regions. Most of these lesions and many of those in the other groups damaged the optic chiasm, but such damage did not seem to be related to changes in the endpoints studied. Figure I
Plasma Corticosterone Levels The 44-hour plasma corticosterone patterns for the groups at 3 weeks after surgery are shown in figure 2. Two-way ANOVA indi cated that there was no overall effect of le sions on the steroid patterns, but there were effects attributable to time of day ( p < 0.005) and interaction (p
Effects of Destruction of the Suprachiasmatic Nuclei on the Circadian Rhythms in Plasma Corticosterone, Body Temperature, Feeding and Plasma Thyrotropin1 K. Abe../. Kroning, M. A. Greer and V. Criiehlow Departments of Anatomy and Medicine, University of Oregon Health Sciences Center, Portland, Oreg.
Key Words. Circadian rhythms • Suprachiasmatic nuclei • Corticosterone • Temperature • Feeding • Thyrotropin
Because destruction of the suprachiasmatic nuclei (SCN) in rodents abolishes several circadian rhythms, it has been postulated that ' This work was supported by USPHS grants AM16794 and AM-01447 from the NIAMDD. NIH. Received: September 22, 1978 Accepted: January 22, 1979
these nuclei serve as a major pacemaker for the generation of such rhythms [I. 8], Moore and Eichler [9] were the first to report that ablation of the SCN in the rat disrupts the circadian rhythm in pituitary-adrenal func tion. However, these investigators used adre nal concentrations of corticosterone in groups of rats killed at four times during the 24-hour
Downloaded by: Kings's College London 137.73.144.138 - 10/18/2017 7:38:27 AM
Abstract. To study the role of the suprachiasmatic nuclei (SCN) in generating circadian rhythms in female rats, lesions were placed in the SCN or in the medial preoptic (PO) region. Serial blood sampling at 4-hour intervals at 3 and 13 weeks after surgery indicated that com plete SCN destruction abolished rhythmic fluctuations in plasma corticosterone levels in indi vidual rats. Partial destruction produced less interference, while medial PO lesions that spared the SCN were without effect. Similar effects were noted on daily changes in body temperature at 10 weeks after surgery; however, some rats showed evidence of dissociation of these two rhythmic functions in that some lesions appeared to affect one and not the other. In ancillary studies, it was found that all lesioned groups showed nocturnal feeding patterns similar to those of the controls and that the diurnal pattern in plasma thyrotropin (TSH) levels was altered by complete destruction of the SCN. These data suggest that the SCN are essential for the circadian rhythms in pituitary-adrenal function and body temperature and that separate pacemakers may be present in these nuclei for these two periodic functions. The SCN may also control rhythmic TSH secretion, but ihese nuclei and the medial PO region do not appear essential for nocturnal feeding.
light-dark cycle to assess the effects of the SCN lesions, so it was impossible to charac terize the nature of the deficit produced. Thus, it could not be determined whether the lesions abolished rhythms in individual rats or pro duced free-running rhythms that were asyn chronous between animals. The former would be expected if the influence of a master oscilla tor were eliminated and the latter if structures involved in entrainment were compromised. Subsequent to the studies of Moore and Eichler. Krieger [5] introduced a procedure for collecting serial blood samples for studying 24-hour patterns of plasma corticosterone levels in individual rats. This approach, which permits better characterization of the effects of experimental manipulationson this rhythm, was used to demonstrate that surgical isola tion of the medial basal hypothalamus [14] and placement of large lesions in the medial preoptic-anterior hypothalamic region [13] abolish the pituitary-adrenal rhythm, while optic enucleation [16] and exposure to con stant light [15] cause free-running. Krieger et at. [6] recently used serial blood sampling to demonstrate the loss of the normal 24-hour pattern of corticosterone levels after placement of SCN lesions, but they did not address the question of the effects of the lesions in indi vidual rats. Raisman and Brown-Grant [12] found that SCN lesions interfered with nor mal morning to evening differences in plasma corticosterone concentrations, but the effects on the rhythm in individual rats were not determined. The aims of the present experiment were to determine the nature of the disruption of rhythmic pituitary-adrenal function produced by ablation of the SCN and to compare the effects of such ablation with those produced by lesions placed in the adjacent medial preoptic (PO) region. To monitor the effects of
Abe K roning G reer-Critchlow
SCN lesions on other rhythmic functions, daily changes in food intake, body tempera ture and plasma concentrations of thyrotro pin (TSH) were also measured at various in tervals after surgery.
Materials and Methods The adult female rats (Charles River, CD) used in these experiments weighed 240-320 g at surgery. The) were maintained under conditions of controlled light (04:00-18:00 h) and temperature (26±2 C) for at least 3 weeks prior to surgery. Purina Laboratory Chow and tap water were available at all times. Thi rats were housed 4/cage until 3 days before each stud) at which time they were placed in individual cages. One group of rats received small, anodal DC le sions aimed at the SCN. The lesions were made by passing 1,0 mA for 10 sec through a platinum electrodi shaped as an inverted T. The crossbar on the electrodi. was 1 mm long and uninsulated. The electrode was lowered through the midline and third ventricle to the dorsum of the optic chiasm with the crossbar in the sagittal plane. The electrode was then rotated 90 so that the crossbar was transversely placed in the region of the SCN. Lesions were made at 4 rostrocaudal por tions by moving the electrode caudally 0.4 mm between each lesion, starting at 2.6 mm anterior to the bregma with the tooth bar located 5.0 mm above the interaural plane. Another group of rats received bilateral DC lesions in the medial PO region with a platinum electrode that was insulated except for I mm at the tip. A current of 1.5 mA was applied for 20 sec with the electrode tip located 2.0 mm anterior to bregma and 0.5 mm on each side of midline. Intact and sham-lcsioned groups served as controls. The sham lesion procedure con sisted of lowering the electrode used for the medial PO lesions 6.0 mm below the cortical surface at 2.0 mm anterior to bregma and withdrawing it without passing current. Treatments were assigned according to a completely randomized design. Plasma Corticosterone Concentrations To determine the effects of the lesions on the circadian rhythm of pituitary-adrenal function, plasma corticosterone concentrations were measured in serial blood samples obtained from a tail vein, using a modi
Downloaded by: Kings's College London 137.73.144.138 - 10/18/2017 7:38:27 AM
120
Suprachiasm atic Nuclei and C ircadian R hythm s
Foot! Intake The daily patterns of food intake were determined at 5-6 weeks after surgery. The rats were placed in individual cages for I week prior to the study and food intake was measured at 6-hour intervals for 24 h, starting at 06:00 h. Body Temperature Body temperatures were monitored at 4-hour inter vals for 44 h at 10 weeks after surgery. As with the corticosterone experiments, the rats were placed in individual cages for 3 days and the animal room was locked 18 h before the study. A rectal Tele-thermom eter probe (Yellow Springs Instruments, Ohio) was inserted 5 cm and temperatures recorded 2.5 min later. Plasma TSH Serial blood samples were collected from tail veins as described above at 18 weeks after surgery for deter mining the effects of the lesions on plasma TSH levels. Because larger volumes of blood (0.8 ml) were needed for measuring TSH, only 3 blood samples were col lected over a period of 12 h, at 06:00, 12:00 and 18:00 h. Plasma TSH was measured by radioimmuno assay, as previously described [2], using materials sup plied by the Rat Pituitary Hormone Distribution Pro gram of the NIAMDD and Dr. A. Parlow. Intra- and interassay coefficients of variation were less than 15%.
The rats were autopsied at 20 weeks after surgery. The brains were removed, fixed and processed for histological examination of lesions. Adrenals, ovaries, uteri and pituitaries were removed and weighed. One- and two-way analyses of variance (ANOVA) for repeated measures [17] were used for analyzing group data derived from serial sampling. Because the number of samples and the duration of sampling were appropriate in the studies involving corticosterone levels and body temperature, the criteria of Wilson and Greer [15] were used with minor modification to evaluate the presence or absence of rhythmicity in individual rats. These criteria offer an objective method for determining whether 24-hour patterns are repeated on 2 consecutive days of observation, i.e.. whether there is evidence of rhythmicity. These criteria are: (I) the highest values on both days occur at the same time of day plus or minus 4 h: (2) there is a positive correlation coefficient between values on both days with time: (3) the correlation coefficient is within 2 SD of the mean coefficient of the intact control group, and (4) the ratio of the maximum to minimum values is within 2 SD of the mean ratio demonstrated by intact controls on both days. The fourth criterion was modified slightly in order to remove the restraint on the extent to which an excursion in a value might ex ceed those of the intact controls. Thus, the maximum to minimum ratio could be more, but not less, than 2 SD above the mean for the intact controls and still be judged rhythmic. This modification prevented label ing the patterns of some rats as arrhythmic because the amplitudes of the daily fluctuations were unusually high. A pattern which satisfied all four criteria was considered rhythmic. Organ and body weights were compared with one-way ANOVA and the NewmanKeuls test.
Results Histology o f Lesions Histological examination showed that most of the lesions aimed for the SCN produced incomplete destruction of these nuclei. In ad dition, some of those aimed for the medial PO region destroyed the SCN, some involved the SCN partially while others spared these nuclei completely. Therefore, the following
Downloaded by: Kings's College London 137.73.144.138 - 10/18/2017 7:38:27 AM
fication of the procedure of Krieger [6] described pre viously [14]. The samples were collected at 4-hour intervals for 44 h at 3 and 13 weeks after surgery, and the rats were unanesthetized but gently restrained dur ing sampling. The animals were placed in individual cages 3 days before these studies, and the room was locked 18 h before sampling began to minimize in advertent environmental disturbances. The rats were removed from their cages and taken individually to an adjacent laboratory where 0.3 ml of blood was col lected less than 3 min from the time of cage opening. Blood was collected in EDTA-rinsed pipettes and centrifuged; the separated plasma was stored (1 5 C) for eventual determination of corticosterone concen trations using the microfluorometric method of Click el al. [4]. Correction for residual fluorescence, approxi mately 6 //g 100 ml of plasma, was not made. Sensitiv ity of the assay is such that 0.5 ng of corticosterone can be detected. The intraassay coefficient of variation in the normal range of the assay was 6% and the inter assay variation was 5%.
121
122
Abe K roning/G rcer/C ritchlcnv
shows photomicrographs of two lesions which completely destroyed the SCN. one which involved primarily these nuclei and one which included portions of the medial PO region. Partial SCN Destruction. 16 rats had le sions which left portions of the SCN morpho logically intact. This group was the most heterogenous in that there was considerable variation in the extent to which these nuclei were involved. However, all rats in this group had portions of the caudal half of the SCN present unilaterally or bilaterally.
Fig. 1. Photomicrographs of two lesions which completely destroyed the SCN. A A small lesion which involved primarily the SCN and optic chiasm. B A larger lesion which included the medial PO region in addition to the SCN.
new experimental groups were formed based on the extent to which the SCN were involved. Complete SCN Destruction. 6 rats had le sions which completely destroyed the SCN: 3 of these lesions were small and minimally involved adjacent brain structures, while 3 were larger and impinged upon the adjacent medial PO and anterior hypothalamic regions. Most of these lesions and many of those in the other groups damaged the optic chiasm, but such damage did not seem to be related to changes in the endpoints studied. Figure I
Plasma Corticosterone Levels The 44-hour plasma corticosterone patterns for the groups at 3 weeks after surgery are shown in figure 2. Two-way ANOVA indi cated that there was no overall effect of le sions on the steroid patterns, but there were effects attributable to time of day ( p < 0.005) and interaction (p
