Brarn Rcsmwh Ruk/rn, Vol. 2X. pp. 629-63 I. I992 Printed in the USA. All nghts reserved
Copyright
0361.9230/92 $5.00 t .OO %I I992 Pergamon Press Ltd.
RAPID COMMUNICATION
Fluoxetine Shortens Circadian Period for Wheel Running Activity in Mice BERNARD
POSSIDENTE,’
AUGUSTUS
R. LUMIA,
Received
SARA
McELDOWNEY
AND
MARK
RAPP
30 July 199 1
B.. A. R. LUMIA. S. McELDOWNEY AND M. RAPP. E‘lrrowir~c shorlm\ c,ircudiun pcwod /or whcc~l running BRAIN RES BULL. 28(4) 629-631. 1992.-Fluoxetine is a potent and specific serotonin re-uptake inhibitor and an effective antidepressant drug. Male mice were treated with either fluoxetine (8 mg/kg body weight per day) or saline. Wheel running activity was monitored for 2 weeks in a 12: I2 LD cycle followed by 2 weeks in constant darkness (DD). Fluoxetlne significantly shortened free-running circadian period for wheel running activity (23.93 2 0.08 h for fluoxetine treated mice versus 24.17 * 0.07 h for saline treated mice; p < 0.03). These results are consistent with a role for serotonin in the regulation of circadian period in mice. POSSIDENTE.
ucIiuIy
VI mrcc
Ruoxetlne
Circadian
period
Mice
Serotonin
Wheel running
lines of evidence indicate a role for serotonin in the regulation of circadian rhythms. Serotonin can phase shift circadian rhythmicity in aplysia eye action potentials (5) and cockroach activity ( I I). Fluoxetine. a serotonin reuptake inhibitor (I) can phase shift a circadian activity rhythm in sparrows (2) and a serotonin agonist, quipazine. induces phase shifts in electrical activity of isolated rat SCN brain slices (14). Lesions of the Raphe nucleus, which is rich in serotonergic pathways and sends efferents to the SCN. and pharmacological treatments that alter serotonin metabolism have been shown to disrupt or modulate the expression of several circadian rhythms in mammals (8,9,16) to varying degrees. Clorgyline, a monoamine oxidase inhibitor, lengthens free-running period of a circadian rhythm for activity in hamsters, and increases the magnitude of phase shifts in response to light pulses (4). Shioiri et al. (15) showed that serotonin concentration in rat suprachiasmatic nuclei was negatively correlated with freerunning period for activity and positively correlated with mean activity level. Although the causal relationship among these variables was not clearly established, the possibilities included feedback effects of activity itself on circadian period through a nonphotic phase response mechanism (10). The evidence for serotonin involvement in circadian pacemaker function was most direct in aplysia, cockroaches, sparrows, and rat SCN tissue in vitro where serotonin administration was an independent experimental variable, and phase response of a circadian pacemaker was the variable assayed. Other evi-
dence for serotonergic regulation of circadian clock function was less direct and either does not demonstrate effects of serotonin manipulation on circadian period or phase-response or involved pharmacological treatments that are not specific to serotonin. The significance ofthis question, however. goes beyond basic neurochemical mechanisms regulating circadian system function, because serotonin deficiency is closely linked to depressive disorders (3) and quantitative disruptions of circadian rhythm expression are also associated with clinical depression ( 17). These associations make effects of antidepressant drugs, such as clorgyline. and fluoxetine on circadian rhythm expression of particular interest. Because fluoxetine is widely prescribed as the antidepressant drug Prozac and is more specific than other drugs in its effects on serotonin metabolism ( I ). we investigated effects of fluoxetine hydrochloride on the free-running circadian period for wheel-running activity in mice. and report here that fluoxetine significantly shortened the free-running circadian period in constant dark.
SEVERAL
’ Requests
tbr reprints
should he addressed
to Bernard
Possidente,
activity
MATERIALS
AND METHOD
Subjects were 22 male, outbred SWR strain house mice (Mus domestic~s), purchased from Taconic Labs (Germantown. NY) at 5 weeks ? 5 days old. The mice were housed individually in transluscent plastic cages 29 X I9 X I3 cm (Length X Width X Height) with ad lib food and water and a 12: I2 LD cycle.
Biopsychology
629
Program.
Skidmore
College, Saratoga Springs. NY 12866.
630
POSSIDFNTE
El
,%I..
range were assigned as the circadian periods for LD and DD. Mean circadian periods of fluoxetine-treated and saline-treated mice did not differ significantly from 24 h in LD (23.95 + 0.07 h and 24.03 ? 0.06 h, respectively). All mice showed significant circadian periodicities in DD. Missing data due to equipment malfunction and absence of a significant circadian periodicity in LD resulted in omission of 2 saline-treated mice from the LD analysis. Actogram plots of wheel-running rhythms were printed from the Dataquest system. The actogram plots display daily histograms of activity plotted in a vertical column with successive days as the ordinate and clock time on the abscissa. Effects of fluoxetine were assessed with analysis of variance (ANOVAI. All means are reported + standard errors of the mean. RESULTS
FIG. 1. Representative double-plotted actograms for SAL and FLX treated mice. Fourteen days of activity in LD 12: I2 are followed by I4 days in DD. The photoperiod for the LD cycle is shown at the top of each actogram. The circadian period of the SAL-treated mouse represented here is 24.21 h and the period of the FLX-treated mouse is 23.96 h.
At approximately 370 days old, four days before activity monitoring began, the mice were acclimated to 47 X 27 X 20 cm (Length X Width X Height) transparent plastic wheel running cages. Each cage had an open wire-mesh floor, a 34 cm diameter stainless-steel activity wheel and ad-lib food and water. Each wheel was connected to a magnetic switch and each wheel-revolution was recorded with the Dataquest III software system (Mini-Mitter Co., Sun River, OR). Wheel revolution counts were organized into IO-min bins. Activity was monitored for 2 weeks in 12: 12 LD following the 4-day acclimation period and for the following 2 weeks in DD. Lights on in LD occurred at 0800 h and lights off at 2000 h. Cages were arranged side by side on open racks, alternating in sequence by drug treatment. Light intensity ranged from approximately 10 to 75 iux at the cage top throughout the room. Fluoxetine hydrochloride (Eli Lilly, Indianapolis, IN) was dissolved in physiological saline and injected intraperitoneally at a dose of 8 mdkg body weight, in approximately 0.1 ml of solution, at 1700-1800 h daily during the 14 days of activity monitoring in LD. Controls received a comparable volume of saline. Fluoxetine was given to 12 mice weighing an average of 45.32 + 1.58 g (SEM) and 10 mice weighing and average of 49.10 t 2.29 g received saline. The cosinor analysis from Dataquest III was used to scan the activity records for each mouse for circadian periodicities in the range of 22-26 h at 5-min intervals. The best-fit cosines in this
AND DISCUSSIOI\I
Fluoxetine treatment significantly shortened TAU(DD) (23.93 h f 0.08 vs. 24.17 F 0.07 h for fluoxetine and salinetreated mice, respectively, F( 1,18) = 5.9, p < 0.03) and there were no dramatic differences in activity patterns between the two treatment groups (Fig. I). The short duration of activity monitoring in DD does not allow us to distinguish between an effect of fluoxetine on the steady-state period of a circadian oscillator driving activity, versus transient changes in circadian period mediated through possible after-effects of the 12: I2 LD photoperiod, or Iluoxetine injections (I 2). The present study. nevertheless, reported evidence of an effect of fluoxetine on a free-running circadian period and is thus consistent with earlier studies that indicated a role for serotonin in the regulation of circadian clock function. Shiori et al.‘s (15) report of a negative correlation between serotonin levels in the SCN and circadian period suggested that manipulation of serotonin levels should alter circadian period in predictable ways. The present results are consistent with Shiori et al. in that fluoxetine should mimic increased serotonin levels as a result of its inhibition of serotonin reuptake and this should shorten circadian period if the correlation described by Shiori et al. reflects a causal relationship. This expectation is also consistent with reports that olfactory bulbectomy in rodents decreases serotonin levels or turnover in specific brain regions (6,7) and also lengthens circadian period (13). These observations warrant further investigation of the role of serotonin metabolism in circadian system function. ACKNOWLEDGEMENTS We thank Skidmore College for supporting this research and Lilly Research Laboratories for supplying the fluoxetine.
REFERENCES Benfield, P.; Hell, R. C.; Lewis, S. P. Fluoxetine. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic efficacy in depressive illness. Drugs 32:481-508; 1986. Cassone, V. M.; Menaker, M. Circadian rhythms of house sparrows are phase shifted by pharmacological manipulation of brain serotonin. J. Comp. Physiol. 156:145-152; 1985. Coccaro, E. F.; Siever, L. J.; KIar, H. M.; Maurer, G.; Cochrane, K.; Cooper, T. B.; Mohs, R. C.; Davis, K. L. Serotonergic studies in patients with affective and personality disorders: Correlates with suicidal and impulsive aggressive behavior. Arch. Gen. Psychiatry 46:587-599; 1989. Duncan, W. C.; Tamarkin, L.; Sokolove, P. G.; Wehr, T. Chronic clorgyline treatment of Syrian hamsters: An analysis of effects on the circadian pacemaker. J. Biol. Rhythms 3:305-322; 1988.
5. Eskin, A.; Corrent, G.; Lin, C.-Y.; McAdoo, D. J. Mechanism for shifting the phase of a circadian rhythm by serotonin: Involvement of CAMP. Proc. Natl. Acad. Sci. 79~660-664; 1982. 6. Hirsch, J. D. The neurochemical sequelae of olfactory bulbectomy. Life Sci. 26:1551-1559; 1980. 7. Lumia, A. R.; Teicher, M. H.; Salchli, F.; Possidente, B. Olfactory bulbectomy as a model for agitated hyposerotonergic depression. Brain Res. Under Review. 8. Mitchell, P. J.; Redfem, P. H.; Stolz, J. Effects of the antidepressant drugs clomipramine, fluoxetine and mianserin on free-running activity rhythms in the rat. Ann. Rev. Chronophamtacology 5:143146: 1988. 9. Morin, L. P.; Michels, K. M.; SmaIe, L.; Moore, R. Y. Serotonin regulation of circadian rhythmicity. Ann. NY Acad. Sci. 600:4 18426; 1990.
FLUOXETINE
AND CIRCADIAN
PERIOD
IO. Mrovsky. N.; Salmon, P. A. A behavioral method for accelerating
re-entrainment of rhythms to new light-dark cycles. Nature 330: 372-373; 1987. 11. Page. T. L. Serotonin phase-shifts the circadian rhythm of locomotor activity in the cockroach. J. Biol. Rhythms 2:23-34; 1987. 13. Pittendrigh. c‘. S.: Daan. S. A functional analysis of circadian pacemakers in nocturnal rodents. I. The stability and lability of spontaneous frequency. 3, Comp. Physiol. 106:223-252; 1976. 13. Possidente. B. P.; Lumia. A. R.; McGinnis, M. Y.; Teicher, M. H.: DeLemos. E.; Sterner, L.: Deros, 1.. Olfactory bulb control of circadian period in mice. Brain Res. S 13:325-328: 1990.
631
14. Prosser, R. A.; Miller, J. D.; Heller, C. A serotonin agonist phaseshifts the circadian clock in the suprachiasmatic nuclei in vitro. Brain Res. 534:336-339: 1990. 15. Shioiri, T.: Kiyohisa, T.: Yamada, N.: Takahashi, S. Motor activity correlates negatively with free-running period, while positively with serotonin contents m SCN in free-running rats. Physiol. PCBehav. 49:779-786: 199 1. 16. Smale, K. M.: Moore, R. Y .; Morin. L. P. Destruction of the hamster serotonergic system by 5,7-DHT: effects on circadian rhythm phase, entrainment and response to triazolam. Brain Res. 5 l5:9-19: 1990. 17. Wehr, T. A.: Goodwin, F. K. Circadian rhythms in psychiatry. Pacific Grove. CA: Boxwood Press: 1983.
Copyright
0361.9230/92 $5.00 t .OO %I I992 Pergamon Press Ltd.
RAPID COMMUNICATION
Fluoxetine Shortens Circadian Period for Wheel Running Activity in Mice BERNARD
POSSIDENTE,’
AUGUSTUS
R. LUMIA,
Received
SARA
McELDOWNEY
AND
MARK
RAPP
30 July 199 1
B.. A. R. LUMIA. S. McELDOWNEY AND M. RAPP. E‘lrrowir~c shorlm\ c,ircudiun pcwod /or whcc~l running BRAIN RES BULL. 28(4) 629-631. 1992.-Fluoxetine is a potent and specific serotonin re-uptake inhibitor and an effective antidepressant drug. Male mice were treated with either fluoxetine (8 mg/kg body weight per day) or saline. Wheel running activity was monitored for 2 weeks in a 12: I2 LD cycle followed by 2 weeks in constant darkness (DD). Fluoxetlne significantly shortened free-running circadian period for wheel running activity (23.93 2 0.08 h for fluoxetine treated mice versus 24.17 * 0.07 h for saline treated mice; p < 0.03). These results are consistent with a role for serotonin in the regulation of circadian period in mice. POSSIDENTE.
ucIiuIy
VI mrcc
Ruoxetlne
Circadian
period
Mice
Serotonin
Wheel running
lines of evidence indicate a role for serotonin in the regulation of circadian rhythms. Serotonin can phase shift circadian rhythmicity in aplysia eye action potentials (5) and cockroach activity ( I I). Fluoxetine. a serotonin reuptake inhibitor (I) can phase shift a circadian activity rhythm in sparrows (2) and a serotonin agonist, quipazine. induces phase shifts in electrical activity of isolated rat SCN brain slices (14). Lesions of the Raphe nucleus, which is rich in serotonergic pathways and sends efferents to the SCN. and pharmacological treatments that alter serotonin metabolism have been shown to disrupt or modulate the expression of several circadian rhythms in mammals (8,9,16) to varying degrees. Clorgyline, a monoamine oxidase inhibitor, lengthens free-running period of a circadian rhythm for activity in hamsters, and increases the magnitude of phase shifts in response to light pulses (4). Shioiri et al. (15) showed that serotonin concentration in rat suprachiasmatic nuclei was negatively correlated with freerunning period for activity and positively correlated with mean activity level. Although the causal relationship among these variables was not clearly established, the possibilities included feedback effects of activity itself on circadian period through a nonphotic phase response mechanism (10). The evidence for serotonin involvement in circadian pacemaker function was most direct in aplysia, cockroaches, sparrows, and rat SCN tissue in vitro where serotonin administration was an independent experimental variable, and phase response of a circadian pacemaker was the variable assayed. Other evi-
dence for serotonergic regulation of circadian clock function was less direct and either does not demonstrate effects of serotonin manipulation on circadian period or phase-response or involved pharmacological treatments that are not specific to serotonin. The significance ofthis question, however. goes beyond basic neurochemical mechanisms regulating circadian system function, because serotonin deficiency is closely linked to depressive disorders (3) and quantitative disruptions of circadian rhythm expression are also associated with clinical depression ( 17). These associations make effects of antidepressant drugs, such as clorgyline. and fluoxetine on circadian rhythm expression of particular interest. Because fluoxetine is widely prescribed as the antidepressant drug Prozac and is more specific than other drugs in its effects on serotonin metabolism ( I ). we investigated effects of fluoxetine hydrochloride on the free-running circadian period for wheel-running activity in mice. and report here that fluoxetine significantly shortened the free-running circadian period in constant dark.
SEVERAL
’ Requests
tbr reprints
should he addressed
to Bernard
Possidente,
activity
MATERIALS
AND METHOD
Subjects were 22 male, outbred SWR strain house mice (Mus domestic~s), purchased from Taconic Labs (Germantown. NY) at 5 weeks ? 5 days old. The mice were housed individually in transluscent plastic cages 29 X I9 X I3 cm (Length X Width X Height) with ad lib food and water and a 12: I2 LD cycle.
Biopsychology
629
Program.
Skidmore
College, Saratoga Springs. NY 12866.
630
POSSIDFNTE
El
,%I..
range were assigned as the circadian periods for LD and DD. Mean circadian periods of fluoxetine-treated and saline-treated mice did not differ significantly from 24 h in LD (23.95 + 0.07 h and 24.03 ? 0.06 h, respectively). All mice showed significant circadian periodicities in DD. Missing data due to equipment malfunction and absence of a significant circadian periodicity in LD resulted in omission of 2 saline-treated mice from the LD analysis. Actogram plots of wheel-running rhythms were printed from the Dataquest system. The actogram plots display daily histograms of activity plotted in a vertical column with successive days as the ordinate and clock time on the abscissa. Effects of fluoxetine were assessed with analysis of variance (ANOVAI. All means are reported + standard errors of the mean. RESULTS
FIG. 1. Representative double-plotted actograms for SAL and FLX treated mice. Fourteen days of activity in LD 12: I2 are followed by I4 days in DD. The photoperiod for the LD cycle is shown at the top of each actogram. The circadian period of the SAL-treated mouse represented here is 24.21 h and the period of the FLX-treated mouse is 23.96 h.
At approximately 370 days old, four days before activity monitoring began, the mice were acclimated to 47 X 27 X 20 cm (Length X Width X Height) transparent plastic wheel running cages. Each cage had an open wire-mesh floor, a 34 cm diameter stainless-steel activity wheel and ad-lib food and water. Each wheel was connected to a magnetic switch and each wheel-revolution was recorded with the Dataquest III software system (Mini-Mitter Co., Sun River, OR). Wheel revolution counts were organized into IO-min bins. Activity was monitored for 2 weeks in 12: 12 LD following the 4-day acclimation period and for the following 2 weeks in DD. Lights on in LD occurred at 0800 h and lights off at 2000 h. Cages were arranged side by side on open racks, alternating in sequence by drug treatment. Light intensity ranged from approximately 10 to 75 iux at the cage top throughout the room. Fluoxetine hydrochloride (Eli Lilly, Indianapolis, IN) was dissolved in physiological saline and injected intraperitoneally at a dose of 8 mdkg body weight, in approximately 0.1 ml of solution, at 1700-1800 h daily during the 14 days of activity monitoring in LD. Controls received a comparable volume of saline. Fluoxetine was given to 12 mice weighing an average of 45.32 + 1.58 g (SEM) and 10 mice weighing and average of 49.10 t 2.29 g received saline. The cosinor analysis from Dataquest III was used to scan the activity records for each mouse for circadian periodicities in the range of 22-26 h at 5-min intervals. The best-fit cosines in this
AND DISCUSSIOI\I
Fluoxetine treatment significantly shortened TAU(DD) (23.93 h f 0.08 vs. 24.17 F 0.07 h for fluoxetine and salinetreated mice, respectively, F( 1,18) = 5.9, p < 0.03) and there were no dramatic differences in activity patterns between the two treatment groups (Fig. I). The short duration of activity monitoring in DD does not allow us to distinguish between an effect of fluoxetine on the steady-state period of a circadian oscillator driving activity, versus transient changes in circadian period mediated through possible after-effects of the 12: I2 LD photoperiod, or Iluoxetine injections (I 2). The present study. nevertheless, reported evidence of an effect of fluoxetine on a free-running circadian period and is thus consistent with earlier studies that indicated a role for serotonin in the regulation of circadian clock function. Shiori et al.‘s (15) report of a negative correlation between serotonin levels in the SCN and circadian period suggested that manipulation of serotonin levels should alter circadian period in predictable ways. The present results are consistent with Shiori et al. in that fluoxetine should mimic increased serotonin levels as a result of its inhibition of serotonin reuptake and this should shorten circadian period if the correlation described by Shiori et al. reflects a causal relationship. This expectation is also consistent with reports that olfactory bulbectomy in rodents decreases serotonin levels or turnover in specific brain regions (6,7) and also lengthens circadian period (13). These observations warrant further investigation of the role of serotonin metabolism in circadian system function. ACKNOWLEDGEMENTS We thank Skidmore College for supporting this research and Lilly Research Laboratories for supplying the fluoxetine.
REFERENCES Benfield, P.; Hell, R. C.; Lewis, S. P. Fluoxetine. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic efficacy in depressive illness. Drugs 32:481-508; 1986. Cassone, V. M.; Menaker, M. Circadian rhythms of house sparrows are phase shifted by pharmacological manipulation of brain serotonin. J. Comp. Physiol. 156:145-152; 1985. Coccaro, E. F.; Siever, L. J.; KIar, H. M.; Maurer, G.; Cochrane, K.; Cooper, T. B.; Mohs, R. C.; Davis, K. L. Serotonergic studies in patients with affective and personality disorders: Correlates with suicidal and impulsive aggressive behavior. Arch. Gen. Psychiatry 46:587-599; 1989. Duncan, W. C.; Tamarkin, L.; Sokolove, P. G.; Wehr, T. Chronic clorgyline treatment of Syrian hamsters: An analysis of effects on the circadian pacemaker. J. Biol. Rhythms 3:305-322; 1988.
5. Eskin, A.; Corrent, G.; Lin, C.-Y.; McAdoo, D. J. Mechanism for shifting the phase of a circadian rhythm by serotonin: Involvement of CAMP. Proc. Natl. Acad. Sci. 79~660-664; 1982. 6. Hirsch, J. D. The neurochemical sequelae of olfactory bulbectomy. Life Sci. 26:1551-1559; 1980. 7. Lumia, A. R.; Teicher, M. H.; Salchli, F.; Possidente, B. Olfactory bulbectomy as a model for agitated hyposerotonergic depression. Brain Res. Under Review. 8. Mitchell, P. J.; Redfem, P. H.; Stolz, J. Effects of the antidepressant drugs clomipramine, fluoxetine and mianserin on free-running activity rhythms in the rat. Ann. Rev. Chronophamtacology 5:143146: 1988. 9. Morin, L. P.; Michels, K. M.; SmaIe, L.; Moore, R. Y. Serotonin regulation of circadian rhythmicity. Ann. NY Acad. Sci. 600:4 18426; 1990.
FLUOXETINE
AND CIRCADIAN
PERIOD
IO. Mrovsky. N.; Salmon, P. A. A behavioral method for accelerating
re-entrainment of rhythms to new light-dark cycles. Nature 330: 372-373; 1987. 11. Page. T. L. Serotonin phase-shifts the circadian rhythm of locomotor activity in the cockroach. J. Biol. Rhythms 2:23-34; 1987. 13. Pittendrigh. c‘. S.: Daan. S. A functional analysis of circadian pacemakers in nocturnal rodents. I. The stability and lability of spontaneous frequency. 3, Comp. Physiol. 106:223-252; 1976. 13. Possidente. B. P.; Lumia. A. R.; McGinnis, M. Y.; Teicher, M. H.: DeLemos. E.; Sterner, L.: Deros, 1.. Olfactory bulb control of circadian period in mice. Brain Res. S 13:325-328: 1990.
631
14. Prosser, R. A.; Miller, J. D.; Heller, C. A serotonin agonist phaseshifts the circadian clock in the suprachiasmatic nuclei in vitro. Brain Res. 534:336-339: 1990. 15. Shioiri, T.: Kiyohisa, T.: Yamada, N.: Takahashi, S. Motor activity correlates negatively with free-running period, while positively with serotonin contents m SCN in free-running rats. Physiol. PCBehav. 49:779-786: 199 1. 16. Smale, K. M.: Moore, R. Y .; Morin. L. P. Destruction of the hamster serotonergic system by 5,7-DHT: effects on circadian rhythm phase, entrainment and response to triazolam. Brain Res. 5 l5:9-19: 1990. 17. Wehr, T. A.: Goodwin, F. K. Circadian rhythms in psychiatry. Pacific Grove. CA: Boxwood Press: 1983.
