- Email: [email protected]
Direct in utero perception of light by the mammalian fetus
Developmental Brain Research, 47 (1989) 151-155 Elsevier
151
BRD 60309
Direct in utero perception of light by the mammalian fetus David R. Weaver and Steven M. Reppert Laboratory of Developmental Chronobiology, Children's Service, Massachusetts General Hospital and Department of Pediatrics and Program in Neuroscience, Harvard Medical School, Boston, MA 02114 (U.S.A.)
(Accepted 24 January 1989) Key words: Circadian rhythm; 2-Deoxyglucoseautoradiography; Retinohypothalamictract; Spiny mouse (Acomys cahirinus)
The primary neural pathway for entrainment of circadian rhythms in rodents is the retinohypothalamic tract, which conveys lighting information from the retina to a biological clock in the hypothalamic suprachiasmatic nuclei (SCN). In a precocious rodent species, the spiny mouse, the retinohypothalamic tract is present and functioning within the SCN on the day of birth, as assessed by HRP histochemistry and 14C-labeled 2-deoxyglucose autoradiography, respectively. Furthermore, direct perception of environmental lighting was observed in fetal spiny mice late in gestation. Direct, retina-mediated fetal light perception appears to be a less potent entraining agent than maternal cues communicatingtime-of-dayinformation to the fetal spiny mouse. Nevertheless, direct fetal light perception may reinforce maternal entraining signals during the prenatal period and therefore be of physiological significance for entrainment of circadian rhythmicity in the fetus. Circadian rhythms are endogenously generated rhythms with a cycle length of approximately 24 h. The suprachiasmatic nuclei (SCN) are the site of a biological clock in rodents and primates, generating a variety of circadian rhythms in physiology and behavior 12. In adult rodents, light is the most potent stimuius for synchronizing (entraining) circadian rhythms to environmental conditions. Photic information reaches the SCN via a direct retinohypothalamic tract and also via a retino-geniculo-hypothalamic pathway 13. In most rodent species, the retinohypothalamic tract develops during the postnatal period 3'9'16, so that direct, retina-mediated effects of light on circadian function only occur after birth 5'6. However, information about environmental lighting is perceived indirectly by the developing animal during this period. The mother acts as a transducer, sensing the lighting cycle and communicating time-of-day information to the fetus and neonate a4. Most of these studies of maternal-fetal communication of lighting information have used altricial rodent species. In more precocious species, however, pereep-
tion of light by the fetus may not be limited to indirect, maternally mediated mechanisms. Direct perception of light by the fetus requires that environmental lighting reaches the fetus in sufficient quantities and at wavelengths suitable for detection. Rodent circadian systems are very sensitive to light of wavelengths around 500 nm 1,2"4'1°,17. At 500 nm approximately 2% of incident light is transmitted through maternal tissue into the uterine lumen of rats and guinea pigs 7. It thus appears that light of biologically relevant wavelengths reaches the fetus. Direct perception of light by the fetus also requires that sensory and neural mechanisms are adequately developed. Fetal light perception is most likely to occur in species where the offspring are well-developed at birth. The spiny mouse (Acomys cahirinus) is a precocious rodent species appropriate for studies of fetal light perception. Litters of 1-4 pups are born after a gestation period of 38-40 days; on the day of birth, olfactory and auditory senses are functional, the eyes open, and the pups are mobile. We thus used this species to examine development of
Correspondence: D.R. Weaver, Laboratory of Developmental Chronobiology,Massachusetts General Hospital, Boston MA 02114, U.S.A.
0165-3806/89/$03.50 ~) 1989 Elsevier Science Publishers B.V. (Biomedical Division)
152 the neural substrate for entrainment, the retinohypothalamic tract, and to determine whether neonatal and fetal spiny mice can directly perceive environmental lighting. For all studies, spiny mice were housed within light-tight environmental compartments in a temperature and humidity controlled animal facility. The light-dark cycle within the compartments was automatically regulated, with cool white fluorescent lights (GE type F20C12 CW bulbs providing 600 lux (178/~W/cm 2) at mid-cage level) on from 04.00 to 18.00 h. Dim red light (>620 nm) remained on during both phases of the light-dark cycle. We first examined whether the primary neural substrate for entrainment to light, the retinohypothalamic tract, is present on the day of birth in spiny mice using horseradish peroxidase (HRP) histochemistry. Intraocular injections of HRP (Sigma type VI, 30%, 4 kd/eye) were performed under methoxyflurane anesthesia; spiny mice were injected within 5-26 h of birth. After a 12 to 22-h survival period, pups were anesthetized with methoxyflurane
and perfused with saline followed by fixative (2.5% glutaraldehyde, 1% paraformaldehyde in phosphate buffer) and 10% sucrose (in phosphate buffer). Free-floating 60- to 100-/~m coronal sections were reacted with tetramethylbenzidine to demonstrate HRP using standard methods 11. Sections were then mounted onto chrome-alum gel-coated slides, dehydrated in graded alcohols to xylene, coverslipped and examined by light microscopy. HRP reaction product was clearly present within the SCN in each of 4 pups receiving intraocular injection of HRP on the day of birth (Fig. 1). Reaction product was also observed in the primary visual system, including the optic chiasm, optic tract, and the lateral geniculate nuclei. The youngest pup examined was a maximum of 17 h old at the time of perfusion; labeling in the SCN of this animal was clear and did not differ from that observed in the other pups (ranging from a maximum of 29-48 h old at perfusion). Thus, the retinohypothalamic tract is present within the SCN on the day of birth, and may be present earlier.
Fig. 1. The retinohypothalamic tract is present within the SCN on the day of birth in spiny mice. Reaction product within the SCN (arrow) following bilateral, intraocular injection of HRP on the day of birth demonstrates the early development of the retinohypothalamic tract. This animal was a maximum of 26 h old at the time of injection; survival time was 22 h. III, third ventricle; OC, optic chiasm. Bar = 100/~m.
153 We next examined the effects of nocturnal light exposure on SCN metabolic activity in newborn spiny mice using 14C-labeled 2-deoxyglucose (DG) autoradiography. In a previous study, we showed that the metabolic activity of the SCN in fetal spiny mice is coordinated with the dam, with higher levels of metabolic activity in the fetal SCN during subjective day than at night TM. Exposure to light at night increases metabolic activity of the SCN in adult rats ~5, adult spiny mice (Weaver and Reppert, unpublished data), and in neonatal rats 6. Thus, photic activation of SCN metabolic activity is a functional measure of retinohypothalamic tract innervation of the SCN and provides a way to detect perception of light. 2-DG (2-deoxy-o [1-14C]-glucose, 2.5 /~Ci/pup; Amersham, 59 Ci/mol) was administered to newborn spiny mice (<36 h old) by i.p. injection. Pups were placed individually into polycarbonate cages and placed under a bank of cool white fluorescent lights providing ca. 1500 lux (ca 450/~W/cm 2) at the level of the animal (Gossen Luna-Pro electronic exposure meter, Gossen Div., Berkey Marketing, Woodside NY). Light exposure lasted for 20 min prior to 2-DG injection at 23.20 h. Animals remained in the light until decapitation 45 min later, at 00.05 h. As a control, littermates remained in dim red light throughout similar steps: they were placed in polycarbonate cages at 23.00 h, injected with 2-DG at 23.20 h, and decapitated at 00.05 h. Brains were removed, frozen in cooled 2-methylbutane (-20 °C), and processed for autoradiography as previously described TM. Metabolic activity of the SCN was assessed by determining the relative optical density (OD) of the SCN (OD of SCN/OD of adjacent hypothalamus). For each section, the OD of SCN and adjacent hypothalamus were determined in triplicate, and a single relative OD (ROD) value was calculated for each section. Mean (+ S.E.M.) ROD values were calculated for 4-7 sections from each animal. SCN metabolic activity was activated in each of 4 pups exposed to light at night on the day of birth; SCN ROD values for pups exposed to light at night were 1.39 + 0.04, 1.21 + 0.01, 1.17 + 0.02, and 1.09 + 0.01, while littermates kept in dim red light ('dark') had SCN ROD values of 1.01 + 0.02, 1.02 + 0.01, and 1.02 + 0.01. The increase in SCN ROD
in pups exposed to light at night is likely due to retina-mediated photic activation of SCN metabolic activity via the retinohypothalamic tract. We next examined the effect of nocturnal light exposure on SCN metabolic activity in fetal spiny mice. Pregnant spiny mice were exposed to bright light at night on gestational day 38 (of a 38 to 40 day gestation; timed-pregnancies were obtained from postpartum matings). Light was provided by one bank of 4 cool white fluorescent tubes 30 cm above the dam (providing ca 3000 lux (ca 900 gW/cm 2) to her dorsal surface) and a second bank 12 cm below the dam (providing ca 5000 lux (ca 1500 gW/cm 2) to the ventral surface). To prevent the dams from perceiving light (and potentially communicating the presence of light to the fetuses), dams were enucleated under methoxyflurane anesthesia on the afternoon of gestational day 38. Light exposure began at 23.00 h, and at 23.20 h the dams were injected with DG (12.5 gCi, i.p.). At 00.05 h (45 min after injection), fetal and maternal brains were removed and processed for autoradiography as described above. The SCN metabolic activity of 3 of 4 fetuses from two litters was activated by light exposure in utero (Fig. 2); fetal SCN ROD values were 1.21 ___ 0.01, 1.11 + 0.01, 1.08 + 0.02, and 1.01 + 0.01. Neither dam's SCN was activated (e.g. SCN ROD values were < 1.05), confirming that enucleation rendered them insensitive to light. We have never observed fetal SCN ROD values > 1.05 in fetuses of dams injected at night without light exposure TM. Furthermore, neither fetus from a dam enucleated on
Fig. 2. 2-DG autoradiographs through the SCN from fetal spiny mice exposed to darkness (left) or light (right) at night on gestationalday 38. The arrow indicates the position of the metabolically active SCN (apparent as two dark spots) after light exposure.
154 gestational day 38 and injected with D G in the dark had activated SCN metabolic activity (fetal SCN R O D values were 0.93 + 0.01 and 0.93 + 0.02). These data suggest that fetal spiny mice are capable of perceiving environmental lighting in utero. Experiments to determine whether light perceived directly by the fetus can actually entrain circadian rhythmicity in the developing spiny mouse have been unsuccessful. When enucleated dams were placed in a reversed light-dark cycle for the last week of gestation, pup circadian rhythmicity was coordinated with the mother and not with the L D cycle. In addition, pups of enucleated dams exposed to a 6-h light extension each night for the last ca 7 nights of gestation were coordinated with the mother, as though light had no effect on their circadian phase. These studies confirm our earlier work demonstrating maternal-fetal communication of time-of-day information in spiny mice TM and suggest that light is less potent than maternal entraining cues during the late prenatal period. Despite being a less potent entraining stimulus than maternal cues, direct fetal light perception may reinforce maternal entraining signals during the prenatal period. These results demonstrate that the retinohypothalamic tract is present within the SCN on the day of birth in spiny mice. Furthermore, nocturnal light exposure can increase SCN metabolic activity in neonatal and fetal spiny mice. Thus, photic information can reach the spiny mouse SCN during the late prenatal period. However, we were unable to 1 Brainard, G.C., Richardson, B.A., King, T.S. and Reiter, R.J., The influence of different spectra on the suppression of pineal melatonin content in the Syrian hamster, Brain Res., 294 (1984) 333-339. 2 Brainard, G.C., Vaughan, M.K. and Reiter, R.J., Effect of light irradiance and wavelength on the Syrian hamster reproductive system, Endocrinology, 119 (1986) 648-654. 3 Burt, S.M., Lund, R.D. and Land, P.W., Prenatal development of the optic projection in albino and hooded rats, Dev. Brain Res., 6 (1983) 149-168. 4 Cardinali, D.P., Larin, F. and Wurtman, R.J., Control of the rat pineal gland by light spectra, Proc. Natl. Acad. Sci. U.S.A., 69 (1972) 2003-2007. 5 Duncan, M.J., Banister, M.J. and Reppert, S.M., Developmental appearance of light-dark entrainment in the rat, Brain Res., 369 (1986) 326-330. 6 Fuchs, J.L. and Moore, R.Y., Development of circadian rhythmicity and light responsiveness in the rat suprachiasmatic nucleus. A study using 2-deoxy-(1-14C)-glucose method, Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 12041208.
demonstrate effects of prenatal light perception on circadian rhythmicity in this species. In fetuses exposed to light at night, activation of SCN metabolic activity was not observed in all animals. Fetal position within the uterus may affect the ability of light to reach the fetal eye. Alternatively, the retinohypothalamic tract may become functional only immediately before birth in this species. Slight variation in the rate of development could produce variability in response to light exposure in fetuses and could contribute to the failure of prenatal light perception to entrain fetal spiny mice. Examination of species in which neurological development is even more advanced prior to birth may be necessary to show physiological effects of prenatal light perception on the circadian timing system. In the present report we have focused on circadian effects of light. Prenatal perception of light may also influence the visual system, as development of the retinohypothalamic tract is delayed relative to other visual projections 3"9'16. While the physiological relevance of prenatal light perception remains to be determined, the potential for direct, retina-mediated light perception by the fetus should be considered when studying development of visual systems. This work was supported by N I H Grant HD-14427 to S . M . R . D . R . W . is the recipient of N R S A award HD-09676. S.M.R. is an Established Investigator of the American Heart Association. Portions of this work have been reported in abstract form 8. 7 Jacques, S.L., Weaver, D,R. and Reppert, S.M., Penetration of light into the uterus of pregnant mammals, Photochem. Photobiol., 45 (1987) 637-641. 8 Jacques, S.L., Weaver, D.R. and Reppert, S.M., Precocious spiny mice as a model to assess the potential for retina-mediated light perception in utero, Soc. Neurosci. Abstr., 13 (1987) 864. 9 Lenn, N.J., Beebe, G.M. and Moore, R.Y., Postnatal development of the suprachiasmatic nucleus in the rat, Cell Tissue Res., 178 (1977) 463-475. 10 McGuire, R.A., Rand, W.M. and Wurtman, R.J., Entrainment of the temperature rhythm in rats: effect of color and intensity of environmental light, Science, 181 (19731 956--957. 11 Mesulam, M.M., Tetramethyl benzidine for horseradish peroxidase neurohistochemistry. A noncarcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents, J. Histochem. Cytochem, 26 (1978) 106-117. 12 Moore, R.Y., Organization and function of a central nervous system oscillator: the suprachiasmatic hypotha-
155 lamic nucleus, Fed. Proc., 42 (1983) 2783-2789. 13 Moore, R.Y. and Card, J.P., Visual pathways and the entrainment of circadian rhythms, Ann. N.Y. Acad. Sci., 453 (1985) 123-133. 14 Reppert, S.M. and Weaver, D.R., Maternal transduction of light-dark information for the fetus. In W.P. Smotherman and S.R] Robinson (Eds.), Behavior of the Fetus, Telford, Caldwell, NJ, 1989, pp. 119-139. 15 Schwartz, W.J. and Gainer, H., Suprachiasmatic nucleus: use of 14C-labeled deoxyglucose uptake as a functional marker, Science, 197 (1977) 1089-1091.
16 Stanfield, B. and Cowan, W.M., Evidence for a change in the retinohypothalamic projection in the rat following early removal of one eye, Brain Res., 104 (1976) 129-136. 17 Takahashi, J.S., DeCoursey, P.J., Bauman, L. and Menaker, M., Spectral sensitivity of a novel photoreceptive system mediating entrainment of mammalian circadian rhythms, Nature (Lond.), 308 (1984) 186-188. 18 Weaver, D.R. and Reppert, S.M., Maternal-fetal communication of circadian phase in a precocious rodent, the spiny mouse, Am. J. Physiol., 253 (1987) E401-407.
151
BRD 60309
Direct in utero perception of light by the mammalian fetus David R. Weaver and Steven M. Reppert Laboratory of Developmental Chronobiology, Children's Service, Massachusetts General Hospital and Department of Pediatrics and Program in Neuroscience, Harvard Medical School, Boston, MA 02114 (U.S.A.)
(Accepted 24 January 1989) Key words: Circadian rhythm; 2-Deoxyglucoseautoradiography; Retinohypothalamictract; Spiny mouse (Acomys cahirinus)
The primary neural pathway for entrainment of circadian rhythms in rodents is the retinohypothalamic tract, which conveys lighting information from the retina to a biological clock in the hypothalamic suprachiasmatic nuclei (SCN). In a precocious rodent species, the spiny mouse, the retinohypothalamic tract is present and functioning within the SCN on the day of birth, as assessed by HRP histochemistry and 14C-labeled 2-deoxyglucose autoradiography, respectively. Furthermore, direct perception of environmental lighting was observed in fetal spiny mice late in gestation. Direct, retina-mediated fetal light perception appears to be a less potent entraining agent than maternal cues communicatingtime-of-dayinformation to the fetal spiny mouse. Nevertheless, direct fetal light perception may reinforce maternal entraining signals during the prenatal period and therefore be of physiological significance for entrainment of circadian rhythmicity in the fetus. Circadian rhythms are endogenously generated rhythms with a cycle length of approximately 24 h. The suprachiasmatic nuclei (SCN) are the site of a biological clock in rodents and primates, generating a variety of circadian rhythms in physiology and behavior 12. In adult rodents, light is the most potent stimuius for synchronizing (entraining) circadian rhythms to environmental conditions. Photic information reaches the SCN via a direct retinohypothalamic tract and also via a retino-geniculo-hypothalamic pathway 13. In most rodent species, the retinohypothalamic tract develops during the postnatal period 3'9'16, so that direct, retina-mediated effects of light on circadian function only occur after birth 5'6. However, information about environmental lighting is perceived indirectly by the developing animal during this period. The mother acts as a transducer, sensing the lighting cycle and communicating time-of-day information to the fetus and neonate a4. Most of these studies of maternal-fetal communication of lighting information have used altricial rodent species. In more precocious species, however, pereep-
tion of light by the fetus may not be limited to indirect, maternally mediated mechanisms. Direct perception of light by the fetus requires that environmental lighting reaches the fetus in sufficient quantities and at wavelengths suitable for detection. Rodent circadian systems are very sensitive to light of wavelengths around 500 nm 1,2"4'1°,17. At 500 nm approximately 2% of incident light is transmitted through maternal tissue into the uterine lumen of rats and guinea pigs 7. It thus appears that light of biologically relevant wavelengths reaches the fetus. Direct perception of light by the fetus also requires that sensory and neural mechanisms are adequately developed. Fetal light perception is most likely to occur in species where the offspring are well-developed at birth. The spiny mouse (Acomys cahirinus) is a precocious rodent species appropriate for studies of fetal light perception. Litters of 1-4 pups are born after a gestation period of 38-40 days; on the day of birth, olfactory and auditory senses are functional, the eyes open, and the pups are mobile. We thus used this species to examine development of
Correspondence: D.R. Weaver, Laboratory of Developmental Chronobiology,Massachusetts General Hospital, Boston MA 02114, U.S.A.
0165-3806/89/$03.50 ~) 1989 Elsevier Science Publishers B.V. (Biomedical Division)
152 the neural substrate for entrainment, the retinohypothalamic tract, and to determine whether neonatal and fetal spiny mice can directly perceive environmental lighting. For all studies, spiny mice were housed within light-tight environmental compartments in a temperature and humidity controlled animal facility. The light-dark cycle within the compartments was automatically regulated, with cool white fluorescent lights (GE type F20C12 CW bulbs providing 600 lux (178/~W/cm 2) at mid-cage level) on from 04.00 to 18.00 h. Dim red light (>620 nm) remained on during both phases of the light-dark cycle. We first examined whether the primary neural substrate for entrainment to light, the retinohypothalamic tract, is present on the day of birth in spiny mice using horseradish peroxidase (HRP) histochemistry. Intraocular injections of HRP (Sigma type VI, 30%, 4 kd/eye) were performed under methoxyflurane anesthesia; spiny mice were injected within 5-26 h of birth. After a 12 to 22-h survival period, pups were anesthetized with methoxyflurane
and perfused with saline followed by fixative (2.5% glutaraldehyde, 1% paraformaldehyde in phosphate buffer) and 10% sucrose (in phosphate buffer). Free-floating 60- to 100-/~m coronal sections were reacted with tetramethylbenzidine to demonstrate HRP using standard methods 11. Sections were then mounted onto chrome-alum gel-coated slides, dehydrated in graded alcohols to xylene, coverslipped and examined by light microscopy. HRP reaction product was clearly present within the SCN in each of 4 pups receiving intraocular injection of HRP on the day of birth (Fig. 1). Reaction product was also observed in the primary visual system, including the optic chiasm, optic tract, and the lateral geniculate nuclei. The youngest pup examined was a maximum of 17 h old at the time of perfusion; labeling in the SCN of this animal was clear and did not differ from that observed in the other pups (ranging from a maximum of 29-48 h old at perfusion). Thus, the retinohypothalamic tract is present within the SCN on the day of birth, and may be present earlier.
Fig. 1. The retinohypothalamic tract is present within the SCN on the day of birth in spiny mice. Reaction product within the SCN (arrow) following bilateral, intraocular injection of HRP on the day of birth demonstrates the early development of the retinohypothalamic tract. This animal was a maximum of 26 h old at the time of injection; survival time was 22 h. III, third ventricle; OC, optic chiasm. Bar = 100/~m.
153 We next examined the effects of nocturnal light exposure on SCN metabolic activity in newborn spiny mice using 14C-labeled 2-deoxyglucose (DG) autoradiography. In a previous study, we showed that the metabolic activity of the SCN in fetal spiny mice is coordinated with the dam, with higher levels of metabolic activity in the fetal SCN during subjective day than at night TM. Exposure to light at night increases metabolic activity of the SCN in adult rats ~5, adult spiny mice (Weaver and Reppert, unpublished data), and in neonatal rats 6. Thus, photic activation of SCN metabolic activity is a functional measure of retinohypothalamic tract innervation of the SCN and provides a way to detect perception of light. 2-DG (2-deoxy-o [1-14C]-glucose, 2.5 /~Ci/pup; Amersham, 59 Ci/mol) was administered to newborn spiny mice (<36 h old) by i.p. injection. Pups were placed individually into polycarbonate cages and placed under a bank of cool white fluorescent lights providing ca. 1500 lux (ca 450/~W/cm 2) at the level of the animal (Gossen Luna-Pro electronic exposure meter, Gossen Div., Berkey Marketing, Woodside NY). Light exposure lasted for 20 min prior to 2-DG injection at 23.20 h. Animals remained in the light until decapitation 45 min later, at 00.05 h. As a control, littermates remained in dim red light throughout similar steps: they were placed in polycarbonate cages at 23.00 h, injected with 2-DG at 23.20 h, and decapitated at 00.05 h. Brains were removed, frozen in cooled 2-methylbutane (-20 °C), and processed for autoradiography as previously described TM. Metabolic activity of the SCN was assessed by determining the relative optical density (OD) of the SCN (OD of SCN/OD of adjacent hypothalamus). For each section, the OD of SCN and adjacent hypothalamus were determined in triplicate, and a single relative OD (ROD) value was calculated for each section. Mean (+ S.E.M.) ROD values were calculated for 4-7 sections from each animal. SCN metabolic activity was activated in each of 4 pups exposed to light at night on the day of birth; SCN ROD values for pups exposed to light at night were 1.39 + 0.04, 1.21 + 0.01, 1.17 + 0.02, and 1.09 + 0.01, while littermates kept in dim red light ('dark') had SCN ROD values of 1.01 + 0.02, 1.02 + 0.01, and 1.02 + 0.01. The increase in SCN ROD
in pups exposed to light at night is likely due to retina-mediated photic activation of SCN metabolic activity via the retinohypothalamic tract. We next examined the effect of nocturnal light exposure on SCN metabolic activity in fetal spiny mice. Pregnant spiny mice were exposed to bright light at night on gestational day 38 (of a 38 to 40 day gestation; timed-pregnancies were obtained from postpartum matings). Light was provided by one bank of 4 cool white fluorescent tubes 30 cm above the dam (providing ca 3000 lux (ca 900 gW/cm 2) to her dorsal surface) and a second bank 12 cm below the dam (providing ca 5000 lux (ca 1500 gW/cm 2) to the ventral surface). To prevent the dams from perceiving light (and potentially communicating the presence of light to the fetuses), dams were enucleated under methoxyflurane anesthesia on the afternoon of gestational day 38. Light exposure began at 23.00 h, and at 23.20 h the dams were injected with DG (12.5 gCi, i.p.). At 00.05 h (45 min after injection), fetal and maternal brains were removed and processed for autoradiography as described above. The SCN metabolic activity of 3 of 4 fetuses from two litters was activated by light exposure in utero (Fig. 2); fetal SCN ROD values were 1.21 ___ 0.01, 1.11 + 0.01, 1.08 + 0.02, and 1.01 + 0.01. Neither dam's SCN was activated (e.g. SCN ROD values were < 1.05), confirming that enucleation rendered them insensitive to light. We have never observed fetal SCN ROD values > 1.05 in fetuses of dams injected at night without light exposure TM. Furthermore, neither fetus from a dam enucleated on
Fig. 2. 2-DG autoradiographs through the SCN from fetal spiny mice exposed to darkness (left) or light (right) at night on gestationalday 38. The arrow indicates the position of the metabolically active SCN (apparent as two dark spots) after light exposure.
154 gestational day 38 and injected with D G in the dark had activated SCN metabolic activity (fetal SCN R O D values were 0.93 + 0.01 and 0.93 + 0.02). These data suggest that fetal spiny mice are capable of perceiving environmental lighting in utero. Experiments to determine whether light perceived directly by the fetus can actually entrain circadian rhythmicity in the developing spiny mouse have been unsuccessful. When enucleated dams were placed in a reversed light-dark cycle for the last week of gestation, pup circadian rhythmicity was coordinated with the mother and not with the L D cycle. In addition, pups of enucleated dams exposed to a 6-h light extension each night for the last ca 7 nights of gestation were coordinated with the mother, as though light had no effect on their circadian phase. These studies confirm our earlier work demonstrating maternal-fetal communication of time-of-day information in spiny mice TM and suggest that light is less potent than maternal entraining cues during the late prenatal period. Despite being a less potent entraining stimulus than maternal cues, direct fetal light perception may reinforce maternal entraining signals during the prenatal period. These results demonstrate that the retinohypothalamic tract is present within the SCN on the day of birth in spiny mice. Furthermore, nocturnal light exposure can increase SCN metabolic activity in neonatal and fetal spiny mice. Thus, photic information can reach the spiny mouse SCN during the late prenatal period. However, we were unable to 1 Brainard, G.C., Richardson, B.A., King, T.S. and Reiter, R.J., The influence of different spectra on the suppression of pineal melatonin content in the Syrian hamster, Brain Res., 294 (1984) 333-339. 2 Brainard, G.C., Vaughan, M.K. and Reiter, R.J., Effect of light irradiance and wavelength on the Syrian hamster reproductive system, Endocrinology, 119 (1986) 648-654. 3 Burt, S.M., Lund, R.D. and Land, P.W., Prenatal development of the optic projection in albino and hooded rats, Dev. Brain Res., 6 (1983) 149-168. 4 Cardinali, D.P., Larin, F. and Wurtman, R.J., Control of the rat pineal gland by light spectra, Proc. Natl. Acad. Sci. U.S.A., 69 (1972) 2003-2007. 5 Duncan, M.J., Banister, M.J. and Reppert, S.M., Developmental appearance of light-dark entrainment in the rat, Brain Res., 369 (1986) 326-330. 6 Fuchs, J.L. and Moore, R.Y., Development of circadian rhythmicity and light responsiveness in the rat suprachiasmatic nucleus. A study using 2-deoxy-(1-14C)-glucose method, Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 12041208.
demonstrate effects of prenatal light perception on circadian rhythmicity in this species. In fetuses exposed to light at night, activation of SCN metabolic activity was not observed in all animals. Fetal position within the uterus may affect the ability of light to reach the fetal eye. Alternatively, the retinohypothalamic tract may become functional only immediately before birth in this species. Slight variation in the rate of development could produce variability in response to light exposure in fetuses and could contribute to the failure of prenatal light perception to entrain fetal spiny mice. Examination of species in which neurological development is even more advanced prior to birth may be necessary to show physiological effects of prenatal light perception on the circadian timing system. In the present report we have focused on circadian effects of light. Prenatal perception of light may also influence the visual system, as development of the retinohypothalamic tract is delayed relative to other visual projections 3"9'16. While the physiological relevance of prenatal light perception remains to be determined, the potential for direct, retina-mediated light perception by the fetus should be considered when studying development of visual systems. This work was supported by N I H Grant HD-14427 to S . M . R . D . R . W . is the recipient of N R S A award HD-09676. S.M.R. is an Established Investigator of the American Heart Association. Portions of this work have been reported in abstract form 8. 7 Jacques, S.L., Weaver, D,R. and Reppert, S.M., Penetration of light into the uterus of pregnant mammals, Photochem. Photobiol., 45 (1987) 637-641. 8 Jacques, S.L., Weaver, D.R. and Reppert, S.M., Precocious spiny mice as a model to assess the potential for retina-mediated light perception in utero, Soc. Neurosci. Abstr., 13 (1987) 864. 9 Lenn, N.J., Beebe, G.M. and Moore, R.Y., Postnatal development of the suprachiasmatic nucleus in the rat, Cell Tissue Res., 178 (1977) 463-475. 10 McGuire, R.A., Rand, W.M. and Wurtman, R.J., Entrainment of the temperature rhythm in rats: effect of color and intensity of environmental light, Science, 181 (19731 956--957. 11 Mesulam, M.M., Tetramethyl benzidine for horseradish peroxidase neurohistochemistry. A noncarcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents, J. Histochem. Cytochem, 26 (1978) 106-117. 12 Moore, R.Y., Organization and function of a central nervous system oscillator: the suprachiasmatic hypotha-
155 lamic nucleus, Fed. Proc., 42 (1983) 2783-2789. 13 Moore, R.Y. and Card, J.P., Visual pathways and the entrainment of circadian rhythms, Ann. N.Y. Acad. Sci., 453 (1985) 123-133. 14 Reppert, S.M. and Weaver, D.R., Maternal transduction of light-dark information for the fetus. In W.P. Smotherman and S.R] Robinson (Eds.), Behavior of the Fetus, Telford, Caldwell, NJ, 1989, pp. 119-139. 15 Schwartz, W.J. and Gainer, H., Suprachiasmatic nucleus: use of 14C-labeled deoxyglucose uptake as a functional marker, Science, 197 (1977) 1089-1091.
16 Stanfield, B. and Cowan, W.M., Evidence for a change in the retinohypothalamic projection in the rat following early removal of one eye, Brain Res., 104 (1976) 129-136. 17 Takahashi, J.S., DeCoursey, P.J., Bauman, L. and Menaker, M., Spectral sensitivity of a novel photoreceptive system mediating entrainment of mammalian circadian rhythms, Nature (Lond.), 308 (1984) 186-188. 18 Weaver, D.R. and Reppert, S.M., Maternal-fetal communication of circadian phase in a precocious rodent, the spiny mouse, Am. J. Physiol., 253 (1987) E401-407.