E L S E V I E R Neuroscience Letters 176 (1994) 80-84
NEUROSCIENCE fETTERS
Host resets phase of grafted SCN: influence of implant site, tissue specificity and pineal secretion
Jacques Servi6re a'*, Ghislaine Gendrota~ Joseph LeSauter b, Rae Silver b
aLaboratoire de Physiologie Sensorielle-INRA, 78352 Jouy-en-Josas, France ~Barnard College of Columbia University, 3009 Broadwa), New York, N Y 10027, USA
Received 20 December 1993: Revised version received 10 May 1994; Accepted 20 May 1994
Abstract The suprachiasmatic nuclei (SCN) have daily fluctuations in energy consumption with glucose utilization high in the day, and low
at night. In a previous study, we used [~4C]2-deoxyglucose (2-DG) uptake to index the phase of the SCN, and found that in intact hamsters bearing SCN grafts in the third ventricle (3V), the native and grafted SCN, which were initially 12 h out of phase, became synchronized to the phase of the host clock [32]. In the present study, adult males (host animals) and pregnant females (source of donor tissue) were housed in opposite light,lark cycles. Host animals were sacrificed 14 days after implantation with either (1) SCN grafts into the lateral ventricle (LV), or (2) cortical grafts into 3V, or (3) SCN grafts in 3V of pinealectomized hamster. The results indicate that rhythms of 2-DG uptake are not induced in cortical tissue grafts, that the host SCN does not entrain SCN grafts located in the LV, and that entrainment of the grafted SCN (located in 3V) by the host circadian system occurs in the absence of pineal secretions.
Many lines of evidence indicate that the suprachias- matic nuclei (SCN) of the anterior hypothalamus serve as pacemakers regulating circadian rhythmicity [21,29,33]. Circadian rhythms in endocrine and behav- ioral responses are abolished following SCN lesions [22,28]. Rhythms of electrical activity are seen within the SCN both in vivo [15] and in vitro [13]. Studies of meta- bolic activity using [~4C]2-deoxyglucose (2-DG) indicate that the SCN exhibit a circadian rhythm of glucose util -~ ization with maximum energy consumption during the subjective day and minimum during subjective night, both in vivo [31] and in vitro [23]. Transplants of anterior hypothalamic tissue containing the SCN restore cir- cadian rhythmicity when implanted in the third ventricle (3V) of SCN lesioned animals (reviewed in [18]). The restored period is that of the donor animal, rather than that of the host [25], indicating control of circadian rhythmicity by the grafted SCN. When SCN grafts are
implanted in the lateral ventricle (LV) restoration of cir- cadian rhythmicity does not occur [7].
It has been reported that pinealectomy has little effect on circadian rhythmicity [24,26]. Nevertheless, the SCN are known to contain melatonin receptors in some ro- dents [16], and several lines of evidence suggest that melatonin alters circadian rhythmicity by acting on the circadian clock [4]. In the rat, (i) daily injections at the onset of locomotor activity period can entrain the free- running rhythm of intact males, (ii) acute administration of melatonin in the late subjective day reduces the SCN 2-DG uptake [6], (iii) micro-injections of melatonin at projected CT12 shift the peak of multi-unit activity of SCN in vitro [20].
In a previous study [32], we demonstrated that SCN grafts implanted in 3V became synchronized with the host SCN within 14 days, and that the graft assumed the phase of the host. In the present study we addressed three issues: first, the influence of site of implantation was assessed by implanting SCN into the lateral ventricle (LV); second, the possibility that metabolic rhythms of
J. Servibre et al. / Neuroscience Letters 176 (1994) 80-84 81
non-SCN tissue could be synchronized to the host SCN was assessed by implanting occipital cortex in the 3V; finally, the role of pineal secretions in synchronizing the rhythms of the graft and host SCN was assessed by implanting SCN in the 3V of pinealectomized hosts. Donor and host animals were 12 h out of phase and we assessed synchronization of metabolic activity in trans- planted and native SCN using 2-DG uptake as an index of phase.
Three experimental groups of SCN intact, adult male hamsters (Mesocricetus auratus) were studied. Half the animals in each group were sacrificed at CT05, and the other half at CT14. Fetal SCN grafts were implanted in LV (n = 9), cortical tissue was implanted in 3V (n = I0), SCN grafts were implanted in the 3V of pinealectomized hamsters (n = 18). Pinealectomy was performed 2-3 weeks prior to grafting surgery.
Animals in the colony were housed in LD 16:8 (lights on 0500-2100). Males to be used as host animals were transferred to LD 12:12 (lights on 0500; off 1700) at about the time of weaning. Females which were to pro- vide pups for donor SCN tissue were transferred to LD 12:12 (lights on 1700; off0500) two weeks before mating. Thus, host and donor animals were in reversed circadian phases at the time of implantation. After grafting surgery at 2-3 months of age, males were transferred to constant darkness (DD) until they were sacrificed.
Wheel-running activity of the host was monitored using a chart recorder that reset each 24 h, from the day of grafting until the animals were sacrificed, so that con- secutive days were plotted on the vertical axis. Locomo- tor activity of the mother was used to assess circadian time of postnatal pups providing donor tissue. The free- running period of hamsters in this colony is 24.19 h _ 0.06.
The grafting procedure and method for localizing the SCN within the fetal hypothalamus has been previously described [17]. For LV implants, the following coordi- nates were used: 1.1 mm anterior to bregma, 1.3 mm right of midline, 3.0 mm below dura. Cortical grafts of approximately 0.5 mm 3 were taken from occipital cortex.
In hamsters, anesthesia increases the difference in 2-DG uptake in the SCN relative to the surrounding hypothalamus [30]. Thus, animals were given Zoletil, a short lasting anesthetic (5 mg/100 g), then injected intra- muscularly with [~4C]2-deoxyglucose (125 mCi/kg, spec. act. 51.9 mCi/mmol, Dositek France), either during their subjective day, at the time when local cerebral glucose utilization (LCGU) values were high (CT 05), or during their subjective night when LCGU values were low (CT 14), and sacrificed 1 h later under deep anesthesia with an overdose of sodium pentobarbital. Grafted animals were sacrificed at 14 days after SCN implantation.
For 2-DG measurements, brains were quickly re- moved and frozen at -35°C. Coronal brain sections
1.50 1
t m
.~ 1.00
e r
0.75
SCN in LV
l k / r =
05 14
• H~SCN I~ Gn~
1.25
.~ 1.0(2
0.75
Cortex in 3V
'k
05 14
~ 1 . ~
o.~
SCN in 3V, PNx host
0 5 1 4
Circadian Time
Fig. I. 2-DG uptake based on ROD measurements of host and grafted SCN made on day 14 after transplantation in animals implanted with SCN tissue in LV (first panel), cortical grafts in 3V (middle panel), and SCN grafts in 3V of pinealectomized animals (bottom panel). Circadian times shown in the table refer to the host animal. At the time of transplantation, the subjective day of the host corresponds to the sub- jective night of the donor, as described in the text. ROD (SCN ODIOD for adjacent hypothalamus) at CT 05 differs from ROD at CTI4 with *P < 0.05 or with **P < 0.01, Tukey's test after significant ANOVA.
(20 pm thick) were cut on a cryostat at -20°C, picked up on numbered glass slides and dried on a hotplate at 50°C. Autoradiographs of the brain sections and ~4C- radioactive standards (Amersham micro-scales RPA 504) were prepared by exposure to X-ray film (Kodak SB) for 7 days. After autoradiography, all sections were stained using Cresyl violet for delineation of host SCN and grafted tissue.
The gray levels of the autoradiographs were measured
82 J. ServiOre et al. / Neuroscience Letters 176 ( 1994 ) 80~84
by an observer blind to the experimental status of the animal, and averaged over the entire anteroposterior extent of the graft using a computer image processing system (Biocom 500, France) providing a 'semi-quantita- tive' method [10]. Relative optical density (ROD) meas- urements were calculated in 2 ways: (1) by dividing the value for the OD of the SCN by the OD of the adjacent hypothalamus and (2) by dividing the value of the OD of the SCN by the OD of radiolabeled standards. Den- sitometric measurements were analyzed statistically using a 2-factor ANOVA of the difference scores in OD between SCN in the graft and in the host animal, and between circadian times (CT 05 and CT 14) (Table 1).
A summary of the results is presented in Fig. 1. The data are expressed as ROD (SCN OD/adjacent hypo- thalamus OD). Results were also calculated using SCN OD/OD of the radiolabeled standards. The results were in the same direction with both methods.
The present results demonstrate that when a grafted SCN is placed in LV, the host SCN exhibits its character- istic circadian rhythm. There is no difference however, between subjective day and subjective night in metabolic activity of the grafted tissue. When an occipital cortex graft is implanted in 3V, the grafted tissue does not ex- hibit a fluctuation in 2-DG uptake even though the host SCN expresses the anticipated circadian rhythm. In pine- alectomized animals, the grafted SCN implanted in 3V expresses a circadian rhythm of 2-DG uptake in syn- chrony with that of the host SCN, with the expected high values during subjective day and lower values during subjective night.
We previously showed the occurrence of synchroniza- tion of metabolic activity between host SCN and grafted fetal tissue in animals in which host and donor were initially 12 h out of phase. More specifically, SCN grafts placed in the 3V became synchronized to the host SCN within 2 weeks of implantation [32]. The present study evaluated the importance of site of implantation, tissue type, and melatonin secretion on such synchronization.
SCN grafts placed in the caudal LV did not become synchronized to the host SCN. This result might be ex- plained by inability of a coupling signal from the host to reach grafted SCN (whether such a signal is a diffusible or synaptic). Alternatively the coupling signal might Table 1
CT05 C T I 4 t-test F-test
2 - D G day 14 p inea lec tomized host (PNX) SCN 1.07 0.87 P = 0.003 Graf t 0.96 0.82 P -- 0.072
P = 0.0065
2 - D G cortex imp lan t SCN 1.01 0.90 P = 0.015 Gra f t 0.78 0.85 P = 0.35
P = 0.5508
2 -DG SCN graf t in la tera l ven t r icu la r SCN 1.05 0.89 P -- 0.0012
Gra f t 0.94 0.94 P = 0.987
P = 0.227
reach the SCN graft in LV but the signal strength might be too weak to produce coupling within a two week period. Such a result is unlikely since LV grafts of the SCN do not restore locomotor rhythmicity in SCN le- sioned hamsters monitored for at least 6 weeks after transplantation [7].
It should be noted that the results do not address the question of whether the SCN graft within the LV is itself rhythmic. We know that the tissue is viable by virtue of Cresyl violet staining and 2-DG uptake. Furthermore, SCN grafts express VIP by 7-10 and VP by 10-16 days after transplantation [27], well before the age at which the grafts were examined.
The absence of synchrony in 2-DG uptake between graft and host indicates that the graft is not controlled by the host SCN. It is likely that pacemaker cells of the graft were free-running and that LV grafts in different animals were out of phase with each other. In sampling only two time points, the trough of glucose consumption is especially easy to miss since, in hamster SCN [12] it is restricted to a narrow temporal window (in comparison to the rat [31]).
The results indicate that circadian rhythms of meta- bolic activity of the grafted and host SCN were synchro- nized and in phase with the host SCN. In contrast, corti- cal grafts in the 3V were not entrained by the host SCN and did not exhibit a circadian fluctuation of glucose uptake. Thus, entrainment is not a passive response of fetal tissue grafts, and cells of the SCN are necessary for synchronization with the host clock.
This result is consistent with behavioral and physio- logical effects of extra-SCN grafts. Circadian rhythms of drinking behavior in rats [11], and of locomotor behav- ior in hamsters [17], are not restored in SCN-lesioned hosts following implantation of occipital cortical grafts, or other non-SCN tissue such as the supraoptic nucleus or the sub-paraventricular hypothalamus [19]. Similarly, it has been shown in the rat that grafts containing the precursor population of SCN neurons re-established the circadian rhythm of vasopressin levels in the cerebrospi- nal fluid, while extra-SCN hypothalamic grafts were in- effective [9]. SCN grafts in 3V were in synchrony with the host SCN 6 weeks after implantation [1], while neither anterior hypothalamic grafts lacking identifiable SCN- like structures, nor posterior hypothalamic grafts showed day-night differences in glucose utilization or single-unit activity. In this latter study, the initial phase of the host and grafted tissue were not controlled, though one might anticipate that the phase of the graft and the host would be far apart in the absence of entrainment at 6 weeks after implantation.
When implanted in the 3V of pinealectomized host animals, the graft exhibited a circadian fluctuation of 2-DG uptake in synchrony with that of the host, suggest- ing that pineal melatonin secretions do not constitute a necessary coupling signal. This is consistent with previ-
J. Servibre et al./ Neuroscience Letters 176 (1994) 80-84 83
ously reported evidence that in Syrian hamsters, mela- tonin injections do not entrain the locomotor activity [14, unpublished work cited in [3]), and that pinealec- tomy does not alter the phase response curve to light pulses [2]. Furthermore, melatonin receptors are absent or of low affinity in the adult [35] consequent to a reduc- tion of specific binding sites around posnatal day 10-12 [8]. In the present experiments, the grafted SCN was 14 days old at the time of sacrifice; nevertheless the absence of pineal melatonin did not interfere with the process of synchronization. Using quantitative measurements of glucose utilization [12] we have shown that acute mela- tonin injection at the end of subjective day does not decrease glucose consumption in Syrian hamster. Simi- larly, Vitte et al. [34] reported a lack of effect of mela- tonin injections upon quantitative glucose utilization of 98 brain areas, including the SCN of the rat, while Cas- sone et al [5,6] reported a decrease in glucose uptake in the rat SCN folowing melatonin injection using relatix~e optical density measurements.
In summary, the results show a close parallel between in vivo (present results) and in vitro [1] studies of SCN grafts implanted into intact hosts. In both instances, the graft and host SCN are synchronized in their rhythms of 2-DG uptake. Furthermore, there is a parallel between the effectiveness of grafts in restoring behavioral rhyth- micity, and their ability to express metabolic rhythms of glucose uptake. Extra-SCN and LV grafts neither fluctu- ate in their glucose uptake, nor restore behvioral rhyth- micity. SCN grafts in 3V restore behavioral rhythmicity irrespective of attachment site in or near the 3V, but are ineffective when implanted in the LV.
This research was supported by INRA (J.S.), AFOSR Grant F49620-92-J-0195DEF (R.S.) and by a NATO grant supporting travel.
[i I Aguilar-Roblero, R., Shibata, S., Speh, J.C., Drucker-Colin, R. and Moore R.Y., Morphological and functional development of the suprachiasmatic nucleus in transplanted fetal hypothalamus, Brain Res., 580 (1992) 288-296.
[2] Aschoff J., Gerecke, U., von Goetz, Chr, Groos, G.A. and Turek, F., Phase responses and characteristics of free-running activity rhythms in the golden hamster: independance of the pineal gland. In J. Aschoff, S. Daan and G. Groos (Eds.), Vertebrate Circadian Systems, Springer, Berlin, 1982, pp. 129-140.
13] Armstrong, S.M. and Redman, J., Melatonin administation: ef- fects on rodent circadian rhythms Photoperiodism, melatonin and the pineal, Ciba Foundation Symposium 117, Pitman, London, 1983, pp. 188 207.
[4] Cassone, V.M., Effects of melatonin on vertebrate circadian sys- tems, Trends Neurosci., 13 (1990)457470.
[5] Cassone, V.M., Roberts, M.H. and Moore, R.Y., Melatonin in- hibits metabolic activity in the rat suprachiasmatic nuclei, Neu- rosci. Lett., 81 (1987) 29-34.
[6] Cassone, V.M., Roberts, M.H. and Moore R.Y., Effects of mela- tonin on 2-deoxy-[1-~4C]glucose uptake within rat suprachiasmatic nucleus, Am. J. Physiol., 255 (1988) R332-R337.
[7] DeCoursey, P.J. and Buggy, J., Tissue and site specificity factors
in restoration of circadian rhythmicity by fetal brain transplants to arrhythmic hosts: neuroendocrine aspects. In T. Hiroshige, S. Fujimoto and K. Honma (Eds.), Endocrine Chronobiology, Hokkaido University Press, Sapporo, Japan, 1992, pp. 46-63.
[8] Duncan, M.J. and Davis, EC., Developmental appearance and age related changes in specific 2-[~25I]iodomelatonin binding sites in the suprachiasmatic nuclei of female Syrian hamsters, Dev. Brain Res., 73 (1993) 205-212.
[9] Earnest, D.J., Sladek, C.D., Gash, D.M. and Wiegand, S.J., Spec- ificity of circadian function in transplants of the fetal suprachias- matic nucleus, J. Neurosci., 9 (1989) 2671-2677.
[10] Gallistel, C.R., Piner, CT., Allen, T.O., Adler, N.T., Yaclin, E. and Negin, M., Computer assisted analysis of 2DG autoradiographs, Neurosci. Behav. Rev., 6 (1982) 148-156.
[11] Garcia-Hernandez, F. Aguilar-Roblero, R. and Drucker-Colin, R., Transplantation of the fetal occipital cortex to the third ventri- cle of SCN-lesioned rats induces a diurnal rhythm in drinking behavior, Brain Res., 418 (1987) 193-197.
[12] Gendrot, G., Martinet, L. and Servi6re, J., Effects of melatonin injections on the circadian rhythm of local cerebral glucose utiliza- tion in the Syrian hamster brain: a quantitative study with [14C]- 2-deoxyglucose. In ¥. Touitou and P. Pevet (Eds.), Proceedings of the International Symposium on Melatonin and the Pineal Gland, Elsevier, Geneva, 1993, pp. 269-272.
[13] Gillette, M.U., SCN electrophysiolegy in vitro: rhythmic activity and endogenous clock properties. In D. Klein, R.Y. Moore and S.M. Reppert (Eds.), The Suprachiasmatic Nucleus: The Mind's Clock, Oxford University Press, New York, 1991, pp. 349374.
[14] Hastings, M.H., Mead, S.M., Vindlacheruvu, R.R., Ebling, F.J.P., Maywood, E.S. and Grosse, J., Non-photic phase shifting of the circadian activity rhythm of Syrian hamsters: the relative potency of arousal and melatonin, Brain Res., 591 (1992) 20-26.
[15] Inouye, S.T. and Kawamura, H., Characteristics of a circadian pacemaker in the suprachiasmatic nucleus, J. Comp. Physiol., 146 (1982) 153-160.
[17] Lehman, M.N., Silver, R., Gladstone, W.R., Kahn, R.M., Gibson, M. and Bittman, E.L., Circadian rhythmicity restored by neural transplant. Immunocytochemical characterization of the graft and its integration with the host brain, J. Neurosci., 7 (1987) 1626- 1638.
[18] Lehman, M.N. and Silver, R., Restoration of circadian rhythms by neural transplants. In J. Marwah, H. Teitelbaum and K.N. Prasad (Eds.), Neurotransplantation, CNS Injury, and Regenera- tion, CRC Press, New York, 1994, pp. 141-160.
[19] LeSauter, J. and Silver, R., SCN transplants are sufficient to sus- tain circadian locomotor rhythms in hamsters, Soc. Neurosci. Abstr., 19 (1993) 1056.
[20] McArthur, A.J., Gillette, M.U. and Prosser, R.A., Melatonin di- rectly resets the rat suprachiasmatic circadian clock in vitro, Brain Res., 565 (1991) 158-161.
[21] Moore, R.Y., Organization and function of a central nervous system circadian oscillator in the suprachiasmatic hypothalamic nucleus, Fed. Proc., 42 (1983) 2783-2789.
[22] Moore, R.Y. and Eichler, V.B., Loss of a circadian adrenal corti- costerone rhythm following suprachiasmatic lesions in the rat, Brain Res., 42 (1972) 201-206.
[23] Newman, G., SCN metabolic activity in vitro. In D. Klein, R.Y. Moore, and S.M. Reppert (Eds.), The Suprachiasmatic Nucleus: The Mind's Clock, Oxford University Press, New York, 1991, pp. 157-176.
[24] Quay, W., Individuation and lack of pineal effect in the rat's circadian locomotor rhythm, Physiol. Behav., 3 (1968) 109-118.
84 J. Servikre et al./Neuroscience Letters 176 (1994) 80-84
[26] Redman, J.R., Circadian effects of melatonin in rats: an update. In Y. Touitou and P. Pevet (Eds.), Proceedings of the International Symposium on Melatonin and the Pineal Gland, Elsevier, Geneva, 1993, pp. 143-150.
[27] Romero, M.-T. and Silver, R., Time course of peptidergic expres- sion in fetal suprachiasmatic nucleus transplanted into adult ham- ster, Dev. Brain Res., 57 (1990) 1-6.
[28] Rusak, B., The role of the suprachiasmatic nuclei in the generation of circadian rhythms in the golden hamster, Mesocrecitus auratus, J. Comp. Physiol., 118 (1977) 145-164.
[29] Rusak, B. and Zucker, I., Neural regulation of circadian rhythms, Physiol. Rev., 59 (1979) 449-526.
[30] Schwartz, W.J., Davidsen, L.C. and Smith, C.B., In vivo metabolic activity of a putative circadian oscillator, the rat suprachiasmatic nucleus, J. Comp. Neurol., 189 (1980) 157-167.
[31] Schwartz, W.J., SCN metabolic activity in vivo. In D. Klein, R.Y.
Moore and S.M. Reppert (Eds.), The Suprachiasmatic Nucleus: The Mind's Clock, Oxford University Press, New York, 1991, pp. 144-156.
[32] Servi6re, J., Gendrot, G. LeSauter, J. and Silver, R., Host resets phase of grafted SCN: a [~4C]-2-deoxyglucose study of time course of synchronization, Soc. Neurosci. Abstr., 18 (1992) 1223.
[33] Stephan, F.K. and Zucker, I, Circadian rhythms in behavior and locomotive activity of rats are eliminated by hypothalamic lesions, Proc. Natl. Acad. Sci. USA, 69 0972) 1583-1586.
[34] Vitte, EA., Brun, J., Lestage, P., Claustrat, B. and Bobillier, E, The effects of melatonin and pinealectomy upon local cerebral glucose utilization in awake unrestrained rats are restricted to a few specific regions, Brain Res., 489 (1989) 273-282.
[35] Weaver D.R., Rivkees S.A. and Reppert S.M., Localization and characterization of melatonin receptors in rodent brain by in vitro autoradiography, J. Neurosci., 9 (1989) 2581-2590.
