Brain Research Bulletin, Vol. 26, pp. 677-682.
m Pergamon Press pk. 1991. Printed in the U.S.A.
0361-9230/91
$3.00 + 40
Effects of Tetrodotoxin on the Circadian Pacemaker Mechanism in Suprachiasmatic Explants in Vitro DAVID J. EARNEST,’
SUSAN M. DIGIORGIO
AND CELIA D. SLADEK
University of Rochester Medical Center, Departments of NeurobiologylAnatomy and Neurology, Rochester, NY I4642 Received
12 November
1990
EARNEST, D. J., S. M. DIGIORGIO AND C. D. SLADEK. Effects of retrodotoxin on the circadian pacemaker mechanism in suprachiasmatic explants in vitro. BRAIN RES BULL 26(5) 677-682, 1991. -Using perifused explants of the rat suprachiasmatic nucleus (SCN), the effects of tetrodotoxin (TTX) on vasopressin (VP) release and its circadian profile were studied at various times throughout the circadian cycle. VP release from SCN explants was consistently attenuated during l’TX treatment, with the amplitude of this effect depending on the time of administration. In addition, the effect of ‘ITX on the circadian pattern of VP release was also time-dependent, such that treatment during the late subjective day was followed by a disruption of circadian rhythmicity in which peptide output remained at basal levels without notable variation whereas treatment at all other times caused no measurable perturbation in the circadian VP rhythm in succeeding cycles. In SCN explants experiencing this TTX-induced arrest of circadian VP output, subsequent exposure to KC1 induced acute increases in VP release, suggesting that VP neurons retain the capacity to actively release peptide in spite of this effect of ‘ITX. These results indicate that the interruption of electrical impulses at a critical phase may compromise the circadian function or output of the pacemaker in the SCN. In addition, the present observations provide further evidence that the overt expression of circadian rhythmicity is dependent on sodium-generated action potentials and that disruption of these electrical signals does not alter the precision of the SCN pacemaker, at least in instances where the phase of the VP rhythm can be determined after treatment. Circadian
rhythms
Suprachiasmatic
nucleus
Vasopressin
Pacemaker
Sodium channel
Action potentials
Organ culture
Although the precise relationship between the electrical activity and VP output of SCN neurons is not known, the temporal coordination of their circadian patterns suggests that electrical impulses may serve to regulate VP release from SCN neurons. Consequently, the purpose of the present study was to determine whether electrical impulses are necessary for the expression of circadian VP release by SCN neurons and for the function of the pacemaker mechanism underlying this rhythm. Since tetrodotoxin (‘FIX) reversibly blocks the generation of action potentials in SCN neurons in vitro (21) and abolishes the expression of SCN-driven circadian rhythms in vivo without altering the underlying pacemaker mechanism (15), experiments were conducted to determine whether the administration of this sodium channel blocker at different times alters VP release and its circadian profile from the SCN in vitro. Explants of the rat SCN maintained under perifusion conditions were utilized to study circadian VP release from the SCN in vitro because this method provides a basis for monitoring the VP rhythm over at least four cycles with sufficient temporal resolution to analyze the phase-shifting effects of neuroactive agents on this rhythm (3). In view of results indicating that explants expressed basal VP output without sign of circadian variation after TTX treatment during a specific phase of the cycle, our analysis was ex-
IN mammals, the generation of circadian rhythms in a variety of biochemical, physiological and behavioral activities is governed by an internal biological clock located in the suprachiasmatic nucleus (SCN) of the anterior hypothalamus (11, 22, 26). Consistent with its function as a circadian pacemaker, various indices of endogenous neural activity expressed by the SCN oscillate on a circadian basis. In this regard, a number of studies have demonstrated that when isolated in vitro or in vivo, the SCN is intrinsically capable of generating circadian patterns of neuronal electrical activity and vasopressin (VP) secretion. Following surgical isolation in a hypothalamic island, the SCN continues to generate a circadian rhythm in multiunit activity while this electrical oscillation is abolished in areas outside of the island (9,lO). SCN neurons also play a unique role in the generation and expression of the circadian VP rhythm in cerebrospinal fluid (16). Moreover, circadian rhythms in single unit firing rate and VP secretion are expressed for one or more cycles by SCN neurons after isolation in hypothalamic slice or explant preparations (2, 3, 6-8). Comparison of these functional markers of SCN pacemaking activity over time reveals that their circadian patterns in vivo and in vitro closely parallel each other, with both SCN electrical activity and VP release rising late in the subjective night, and reaching peak levels early in the subjective day. ‘Requests for reprints should be addressed to Dr. David J. Earnest, Anatomy, 601 Ehnwood Avenue, Box 603, Rochester, NY 14642.
University
677
of Rochester
Medical
Center,
Department
of Neurobiology
and
678
EARNEST,
tended to address the possibility that this arrest of the circadian pattern of VP release may be a consequence of a chronic disruption of active peptide release. Accordingly, additional experiments were conducted to determine whether exposure to potassium chloride (KCl) could stimulate VP release from SCN explants during this TTX-induced arrest of rhythmic VP output. METHOD
Donor Animals and Explant Dissection Procedures Male Sprague-Dawley rats (Charles River Laboratories) weighing 100-125 g were used to obtain the specific brain tissue required in the present analysis. All animals were housed for 7-8 days prior to experimentation in a temperature- and light-controlled room (LD 12:12; lights-on at 0600 h) with access to food and water ad lib. Using a paradigm established in previous studies (2,3), explants were dissected from the hypothalami of decapitated animals so as to include only the bilateral suprachiasmatic nuclei, their rostra1 projections to the organum vasculosum of the lamina terminalis and the underlying optic chiasm. Culture Conditions Following dissection, SCN explants were repeatedly rinsed with sterile culture medium (Nutrient Mixture Fi2; GIBCO) containing 100 U/ml penicillin and 100 &ml streptomycin and then placed in individual culture chambers which, in turn, were mounted in a perifusion cell culture system (Endotronics Inc.). Culture medium (Nutrient Mixture F,, with sodium chloride reduced to yield a final osmolarity of -290-300 mosmol/kg H,O) supplemented with 10% fetal calf serum, 2 kg/ml glucose and 292 Kg/ml L-glutamine was perifused over the explants (rate= 0.4 ml/h) via a fluid conduit from the medium reservoir, through a multichannel peristaltic pump, to and from the culture chambers. Incoming medium was warmed using a circulating water bath heater and gassed in the heat/gas exchanger with 95% 0,/S% CO,. Lighting conditions, oxygen levels and the thermal environment were established and maintained through the design of the perifusion culture system; throughout all experiments, conditions of constant temperature (37 + 0. 1°C) and darkness as well as the maintenance of oxygen levels were accomplished by enclosing the culture chambers within an insulated compartment. Following a 6-hour period of equilibration, serial samples of the perifusate were collected at 2-hour intervals for 4 days with a fraction collector cooled to 45°C. To retard the enzymatic degradation of VP (19), 20 ~1 of bacitracin dissolved in dH,O was added a priori to the sample tubes so as to effectively yield a final concentration of 1.7 X 1O-4 M upon dilution in the total volume of each media sample. At the conclusion of each experiment, all samples of the perifusate were frozen and stored at - 20°C until radioimmunoassay. Experimental Protocol and Analysis Experiments were conducted in which SCN explants (n = 36) were exposed to culture medium containing lTX (peak cont. = 3.1 x lo-’ M) on day 2 in culture for 6 or 12 hours at various times throughout the circadian cycle. TTX inhibits the generation of action potentials by selectively and reversibly blocking sodium channels, preventing the requisite increase in sodium conductance (5,12). VP output from individual explants during the period of ‘ITX treatment was compared with that observed in untreated control explants (n= 18) during the same interval of time. Statistical analysis of the effect of ‘ITX treatment on VP
DIGIORGIO
AND SLADEK
release was performed on the mean values using a pooled t-test. To determine whether TTX treatment shifted the circadian rhythm of VP release, analysis focused on a phase reference point in individual records of SCN explants corresponding to the first sampling interval after a period of basal release in which the increment in VP output was at least 25% of the difference between the trough and the succeeding peak of the VP rhythm. As demonstrated previously (3), this reference point provides a reliable marker for determining the period length of the VP rhythm and for analyzing treatment-induced changes in the period and phase of the rhythm. In the present study, control and ‘ITX-treated explants were compared with regard to the period between consecutive phase reference points during the last three cycles in culture. Two-way analysis of variance with repeated measures was used to determine if the period between reference points during these cycles differed between control and ‘ITXtreated explants, and the differences were tested post hoc for significance using the simple main effects test and the NewmanKeuls test. In a second series of experiments, SCN explants (n = 6) were exposed to medium containing ‘ITX (peak cont. = 3.1 x lo-’ M) for 6 hours on day 2 in culture during the late subjective day (i.e., when TTX treatment had been observed to consistently disrupt circadian VP release) and then challenged with KCl-supplemented medium (peak cont. =41 mM) for 1 hour on day 4 in culture during the subjective day. VP release by an individual explant during exposure to KC1 was compared to VP output by the explant during the preceding sampling interval. Statistical analyses were performed on the raw data using a paired t-test to determine the effect of KC1 administration on VP release. VP Radioimmunoassay Determinations of VP concentration in duplicate 200 p,l aliquots of culture medium from each sampling interval were performed by radioimmunoassay (20), using an antiserum provided by Drs. Lindheimer and Durr of the University of Chicago. The assay has a minimum sensitivity of 0.25 pg of VP/assay tube. and crossreactivity with oxytocin is less than 0.001%. Interassay variation is 15.6%. Vasopressin concentration in all samples from any given explant was determined in a single assay. RESULTS
In 16 of the 18 untreated control explants, circadian rhythmicity was clearly evident in the temporal pattern of VP release throughout the period of analysis. The circadian pattern of VP output was homogeneous among individual control explants, although the levels of release during a given time interval showed individual variation, especially during the first and last days of sampling. Signs of rhythmic VP release were similarly apparent prior to experimental manipulation (i.e., day 1 in culture) in 35 of the 36 TTX-treated explants and persisted after treatment, except as noted below. The circadian rhythms in VP release generated by these control and TTX-treated explants were consistently phase-coordinated such that peptide release increased late in the subjective night, reached peak levels near the middle of the subjective day and after a precipitous decline remained at basal levels throughout the late subjective day and most of the subjective night. The aforementioned control and experimentally treated explants that failed to clearly generate circadian rhythms in VP release were excluded from further analysis. VP release from SCN explants was consistently depressed
EFFECTS OF TTX ON SCN PACEMAKER
619
FUNCTION
‘-““-‘*
+z
UNTREATED
_“’
-
TTX
CONTROC (t&S)
TTX-TREATED
(N.12)
100
*
20 +_g 2
PO B0
a:
50
ulae cn0-l
d
. . 0
i D
>
0
CONTROL
12.SD
6.ESD
B-LSD
12.SN
l-TX TREATMENT
B/“:
FIG. 1. Effect of ‘lTX on peak and basal levels of VP release from SCN explants. Explants were exposed to TTX on day 2 in culture for either 12 h during the subjective day (12~SD; N= 12), 6 h early in the subjective day (6-ESD; N= ll), 6 h late in the subjective day (6-LSD; N=6), or 12 h during the subjective night (12-SN; N=6). For each group of ‘ITX-treated explants, VP release during treatment is expressed as a percentage (mean k SEM) of VP release from their respective control explants (CONTROL) over the same interval. This analysis is based on comparisons utilizing all explants within the control and TTX-treated groups that contributed to the temporal profiles shown in Fig. 2. In all cases, TTX exposure yielded a significant decrease (*p
m Pergamon Press pk. 1991. Printed in the U.S.A.
0361-9230/91
$3.00 + 40
Effects of Tetrodotoxin on the Circadian Pacemaker Mechanism in Suprachiasmatic Explants in Vitro DAVID J. EARNEST,’
SUSAN M. DIGIORGIO
AND CELIA D. SLADEK
University of Rochester Medical Center, Departments of NeurobiologylAnatomy and Neurology, Rochester, NY I4642 Received
12 November
1990
EARNEST, D. J., S. M. DIGIORGIO AND C. D. SLADEK. Effects of retrodotoxin on the circadian pacemaker mechanism in suprachiasmatic explants in vitro. BRAIN RES BULL 26(5) 677-682, 1991. -Using perifused explants of the rat suprachiasmatic nucleus (SCN), the effects of tetrodotoxin (TTX) on vasopressin (VP) release and its circadian profile were studied at various times throughout the circadian cycle. VP release from SCN explants was consistently attenuated during l’TX treatment, with the amplitude of this effect depending on the time of administration. In addition, the effect of ‘ITX on the circadian pattern of VP release was also time-dependent, such that treatment during the late subjective day was followed by a disruption of circadian rhythmicity in which peptide output remained at basal levels without notable variation whereas treatment at all other times caused no measurable perturbation in the circadian VP rhythm in succeeding cycles. In SCN explants experiencing this TTX-induced arrest of circadian VP output, subsequent exposure to KC1 induced acute increases in VP release, suggesting that VP neurons retain the capacity to actively release peptide in spite of this effect of ‘ITX. These results indicate that the interruption of electrical impulses at a critical phase may compromise the circadian function or output of the pacemaker in the SCN. In addition, the present observations provide further evidence that the overt expression of circadian rhythmicity is dependent on sodium-generated action potentials and that disruption of these electrical signals does not alter the precision of the SCN pacemaker, at least in instances where the phase of the VP rhythm can be determined after treatment. Circadian
rhythms
Suprachiasmatic
nucleus
Vasopressin
Pacemaker
Sodium channel
Action potentials
Organ culture
Although the precise relationship between the electrical activity and VP output of SCN neurons is not known, the temporal coordination of their circadian patterns suggests that electrical impulses may serve to regulate VP release from SCN neurons. Consequently, the purpose of the present study was to determine whether electrical impulses are necessary for the expression of circadian VP release by SCN neurons and for the function of the pacemaker mechanism underlying this rhythm. Since tetrodotoxin (‘FIX) reversibly blocks the generation of action potentials in SCN neurons in vitro (21) and abolishes the expression of SCN-driven circadian rhythms in vivo without altering the underlying pacemaker mechanism (15), experiments were conducted to determine whether the administration of this sodium channel blocker at different times alters VP release and its circadian profile from the SCN in vitro. Explants of the rat SCN maintained under perifusion conditions were utilized to study circadian VP release from the SCN in vitro because this method provides a basis for monitoring the VP rhythm over at least four cycles with sufficient temporal resolution to analyze the phase-shifting effects of neuroactive agents on this rhythm (3). In view of results indicating that explants expressed basal VP output without sign of circadian variation after TTX treatment during a specific phase of the cycle, our analysis was ex-
IN mammals, the generation of circadian rhythms in a variety of biochemical, physiological and behavioral activities is governed by an internal biological clock located in the suprachiasmatic nucleus (SCN) of the anterior hypothalamus (11, 22, 26). Consistent with its function as a circadian pacemaker, various indices of endogenous neural activity expressed by the SCN oscillate on a circadian basis. In this regard, a number of studies have demonstrated that when isolated in vitro or in vivo, the SCN is intrinsically capable of generating circadian patterns of neuronal electrical activity and vasopressin (VP) secretion. Following surgical isolation in a hypothalamic island, the SCN continues to generate a circadian rhythm in multiunit activity while this electrical oscillation is abolished in areas outside of the island (9,lO). SCN neurons also play a unique role in the generation and expression of the circadian VP rhythm in cerebrospinal fluid (16). Moreover, circadian rhythms in single unit firing rate and VP secretion are expressed for one or more cycles by SCN neurons after isolation in hypothalamic slice or explant preparations (2, 3, 6-8). Comparison of these functional markers of SCN pacemaking activity over time reveals that their circadian patterns in vivo and in vitro closely parallel each other, with both SCN electrical activity and VP release rising late in the subjective night, and reaching peak levels early in the subjective day. ‘Requests for reprints should be addressed to Dr. David J. Earnest, Anatomy, 601 Ehnwood Avenue, Box 603, Rochester, NY 14642.
University
677
of Rochester
Medical
Center,
Department
of Neurobiology
and
678
EARNEST,
tended to address the possibility that this arrest of the circadian pattern of VP release may be a consequence of a chronic disruption of active peptide release. Accordingly, additional experiments were conducted to determine whether exposure to potassium chloride (KCl) could stimulate VP release from SCN explants during this TTX-induced arrest of rhythmic VP output. METHOD
Donor Animals and Explant Dissection Procedures Male Sprague-Dawley rats (Charles River Laboratories) weighing 100-125 g were used to obtain the specific brain tissue required in the present analysis. All animals were housed for 7-8 days prior to experimentation in a temperature- and light-controlled room (LD 12:12; lights-on at 0600 h) with access to food and water ad lib. Using a paradigm established in previous studies (2,3), explants were dissected from the hypothalami of decapitated animals so as to include only the bilateral suprachiasmatic nuclei, their rostra1 projections to the organum vasculosum of the lamina terminalis and the underlying optic chiasm. Culture Conditions Following dissection, SCN explants were repeatedly rinsed with sterile culture medium (Nutrient Mixture Fi2; GIBCO) containing 100 U/ml penicillin and 100 &ml streptomycin and then placed in individual culture chambers which, in turn, were mounted in a perifusion cell culture system (Endotronics Inc.). Culture medium (Nutrient Mixture F,, with sodium chloride reduced to yield a final osmolarity of -290-300 mosmol/kg H,O) supplemented with 10% fetal calf serum, 2 kg/ml glucose and 292 Kg/ml L-glutamine was perifused over the explants (rate= 0.4 ml/h) via a fluid conduit from the medium reservoir, through a multichannel peristaltic pump, to and from the culture chambers. Incoming medium was warmed using a circulating water bath heater and gassed in the heat/gas exchanger with 95% 0,/S% CO,. Lighting conditions, oxygen levels and the thermal environment were established and maintained through the design of the perifusion culture system; throughout all experiments, conditions of constant temperature (37 + 0. 1°C) and darkness as well as the maintenance of oxygen levels were accomplished by enclosing the culture chambers within an insulated compartment. Following a 6-hour period of equilibration, serial samples of the perifusate were collected at 2-hour intervals for 4 days with a fraction collector cooled to 45°C. To retard the enzymatic degradation of VP (19), 20 ~1 of bacitracin dissolved in dH,O was added a priori to the sample tubes so as to effectively yield a final concentration of 1.7 X 1O-4 M upon dilution in the total volume of each media sample. At the conclusion of each experiment, all samples of the perifusate were frozen and stored at - 20°C until radioimmunoassay. Experimental Protocol and Analysis Experiments were conducted in which SCN explants (n = 36) were exposed to culture medium containing lTX (peak cont. = 3.1 x lo-’ M) on day 2 in culture for 6 or 12 hours at various times throughout the circadian cycle. TTX inhibits the generation of action potentials by selectively and reversibly blocking sodium channels, preventing the requisite increase in sodium conductance (5,12). VP output from individual explants during the period of ‘ITX treatment was compared with that observed in untreated control explants (n= 18) during the same interval of time. Statistical analysis of the effect of ‘ITX treatment on VP
DIGIORGIO
AND SLADEK
release was performed on the mean values using a pooled t-test. To determine whether TTX treatment shifted the circadian rhythm of VP release, analysis focused on a phase reference point in individual records of SCN explants corresponding to the first sampling interval after a period of basal release in which the increment in VP output was at least 25% of the difference between the trough and the succeeding peak of the VP rhythm. As demonstrated previously (3), this reference point provides a reliable marker for determining the period length of the VP rhythm and for analyzing treatment-induced changes in the period and phase of the rhythm. In the present study, control and ‘ITX-treated explants were compared with regard to the period between consecutive phase reference points during the last three cycles in culture. Two-way analysis of variance with repeated measures was used to determine if the period between reference points during these cycles differed between control and ‘ITXtreated explants, and the differences were tested post hoc for significance using the simple main effects test and the NewmanKeuls test. In a second series of experiments, SCN explants (n = 6) were exposed to medium containing ‘ITX (peak cont. = 3.1 x lo-’ M) for 6 hours on day 2 in culture during the late subjective day (i.e., when TTX treatment had been observed to consistently disrupt circadian VP release) and then challenged with KCl-supplemented medium (peak cont. =41 mM) for 1 hour on day 4 in culture during the subjective day. VP release by an individual explant during exposure to KC1 was compared to VP output by the explant during the preceding sampling interval. Statistical analyses were performed on the raw data using a paired t-test to determine the effect of KC1 administration on VP release. VP Radioimmunoassay Determinations of VP concentration in duplicate 200 p,l aliquots of culture medium from each sampling interval were performed by radioimmunoassay (20), using an antiserum provided by Drs. Lindheimer and Durr of the University of Chicago. The assay has a minimum sensitivity of 0.25 pg of VP/assay tube. and crossreactivity with oxytocin is less than 0.001%. Interassay variation is 15.6%. Vasopressin concentration in all samples from any given explant was determined in a single assay. RESULTS
In 16 of the 18 untreated control explants, circadian rhythmicity was clearly evident in the temporal pattern of VP release throughout the period of analysis. The circadian pattern of VP output was homogeneous among individual control explants, although the levels of release during a given time interval showed individual variation, especially during the first and last days of sampling. Signs of rhythmic VP release were similarly apparent prior to experimental manipulation (i.e., day 1 in culture) in 35 of the 36 TTX-treated explants and persisted after treatment, except as noted below. The circadian rhythms in VP release generated by these control and TTX-treated explants were consistently phase-coordinated such that peptide release increased late in the subjective night, reached peak levels near the middle of the subjective day and after a precipitous decline remained at basal levels throughout the late subjective day and most of the subjective night. The aforementioned control and experimentally treated explants that failed to clearly generate circadian rhythms in VP release were excluded from further analysis. VP release from SCN explants was consistently depressed
EFFECTS OF TTX ON SCN PACEMAKER
619
FUNCTION
‘-““-‘*
+z
UNTREATED
_“’
-
TTX
CONTROC (t&S)
TTX-TREATED
(N.12)
100
*
20 +_g 2
PO B0
a:
50
ulae cn0-l
d
. . 0
i D
>
0
CONTROL
12.SD
6.ESD
B-LSD
12.SN
l-TX TREATMENT
B/“:
FIG. 1. Effect of ‘lTX on peak and basal levels of VP release from SCN explants. Explants were exposed to TTX on day 2 in culture for either 12 h during the subjective day (12~SD; N= 12), 6 h early in the subjective day (6-ESD; N= ll), 6 h late in the subjective day (6-LSD; N=6), or 12 h during the subjective night (12-SN; N=6). For each group of ‘ITX-treated explants, VP release during treatment is expressed as a percentage (mean k SEM) of VP release from their respective control explants (CONTROL) over the same interval. This analysis is based on comparisons utilizing all explants within the control and TTX-treated groups that contributed to the temporal profiles shown in Fig. 2. In all cases, TTX exposure yielded a significant decrease (*p
