MELATONIN, the principal endocrine product of the

Transcription

1 /99/$03.00/0 Vol. 140, No. 4 Endocrinology Printed in U.S.A. Copyright 1999 by The Endocrine Society Neither Functional Rod Photoreceptors nor Rod or Cone Outer Segments Are Required for the Photic Inhibition of Pineal Melatonin* ROBERT J. LUCAS AND RUSSELL G. FOSTER Department of Biology, Imperial College of Science Technology and Medicine, London, United Kingdom SW7 2AZ ABSTRACT Pineal melatonin production is rapidly suppressed by light. In mammals, the photoreceptors mediating this response are ocular; however, definitive information regarding their nature and precise location is absent. In an attempt to define these photoreceptors, we examined the sensitivity of pineal melatonin production to inhibition by controlled irradiance monochromatic green light ( max 509 nm) in C3H mice bearing either of two mutations affecting the retina: retinal degeneration (rd), a disruption of rod phototransduction, and retinal degeneration slow (rds), an ablation of photoreceptor outer segments. Diurnal profiles of pineal melatonin content were similar in both MELATONIN, the principal endocrine product of the mammalian pineal gland, regulates circadian and seasonal variations in physiology by acting as an internal representative of night. To perform this function it is essential that its production be strictly confined to the hours of darkness. This pattern of production is ensured by two complementary factors: a strong circadian rhythmicity in stimulation of the pineal gland, originating in the hypothalamic suprachiasmatic nuclei (SCN; site of a circadian clock in mammals), and a rapid inhibition of pineal melatonin production upon exposure to light (1). A multisynaptic neural pathway has been described by which photic information is transmitted from the retina via the retinohypothalamic tract to the SCN (2) and thence to the pineal (3). However, the nature of the photoreceptors providing input to this pathway remains unknown. Previous descriptions of the spectral sensitivity of melatonin suppression in rats (4, 5) and Syrian hamsters (6) have suggested the involvement of rod photoreceptors in mediating these responses. However, definitive evidence associating any specific retinal photoreceptor with the task of regulating the mammalian pineal is absent. Here, we set out to examine the involvement of rod and cone photoreceptors in this process. To this end, we assessed the effect of two retinal degeneration mouse models, the retinal degeneration (rd) and retinal degeneration slow (rds) mutations, on the ability of light to acutely suppress pineal melatonin. rd is a mutation Received August 20, Address all correspondence and requests for reprints to: Dr. Robert Lucas, Department of Biology, Sir Alexander Fleming Building, Imperial College Road, Imperial College of Science Technology and Medicine, London, United Kingdom SW7 2AZ. r.j.lucas@ic.ac.uk. * This work was supported by a Biotechnology and Biological Sciences Research Council research grant (to R.G.F.). mutant genotypes and in wild-type mice; melatonin peaked between 3 5 h before lights on. All three genotypes exhibited irradiance dependent inhibition of pineal melatonin content; microwatts/cm nm light induced complete suppression in all three genotypes, whereas lower irradiances were ineffective in all cases. Bilateral enucleation abolished responses even to 6 microwatts/cm nm light. These results demonstrate that the process of irradiance detection for pineal melatonin inhibition is buffered against considerable loss of photoreceptive capacity and that neither rod photoreceptors nor rod or cone outer segments are required for mediating this response in mice. (Endocrinology 140: , 1999) of the -subunit of the rod-specific phosphodiesterase (7, 8). The absence of a functional phosphodiesterase in homozygous rd/rd mice leads to a constitutive elevation of intracellular cgmp in rod photoreceptors, thus destroying their ability to respond to photic stimulation with appropriate changes in membrane potential. This primary ablation of rod phototransduction is accompanied by an attrition initially of rod and subsequently of cone photoreceptors. By days (the age at which the current experiments were carried out), rod cell bodies are absent, and cone cell bodies are reduced by at least 50% (9). rds is an insertion mutation of the peripherin gene that encodes a key structural component of photoreceptor outer segments (10 12). Homozygous rds/rds mice never develop photoreceptor outer segments and show a gradual degeneration of both rod and cone cell bodies. By days of age, the outer nuclear layer of the rds/rds retina is reduced by more than 50% (13). In a previous examination of melatonin suppression in rd/rd mice, Goto and Ebihara (14) reported that although rd/rd mice were capable of exhibiting photic melatonin suppression, their sensitivity to light was significantly reduced when compared to wild-type animals. The ability of rd/rd mice to show melatonin suppression indicates that photoreceptors other than rods are capable of mediating this response. However, the reduction in sensitivity was interpreted as strong evidence that, when present, rod photoreceptors contribute to this pathway. On this basis, Goto and Ebihara (14) concluded that regulation of the mammalian pineal is mediated by both rod and non-rod photoreceptors. However, this interpretation is complicated by strain differences between the rd/rd (C3H/He strain) and wild-type (CBA/Ms) mice. Thus, it is possible that the differences reported were related to strain background rather than retinal phenotype. The present 1520

2 PHOTIC MELATONIN SUPPRESSION IN rd AND rds MICE 1521 study set out to resolve this issue and to independently examine the requirement for rod and cone outer segments by comparing photic sensitivity in C3H mice homozygous for either the rd or rds mutation with that in mice wild-type at both loci. Our results indicate that despite the large reductions in photoreceptive capacity induced by rd and rds mutations, there is no concomitant reduction in the sensitivity of pineal melatonin production to monochromatic (509-nm) light exposure in either genotype. These findings demonstrate that neither rod photoreceptors nor rod/cone outer segments are required for the acute inhibition of pineal melatonin in mice. Animals Materials and Methods All experiments described were performed in accordance with the Animals (Scientific Procedures) Act of Breeding colonies of C3H mice homozygous for either the rd or rds mutations or wild type at both loci were maintained at 22 C and 50% humidity at Imperial College. Commercially available C3H/He mice are homozygous for the rd mutation. C3H mice wild-type at this loci and bearing the rds mutation (from O20/A strain) were generated by S. Sanyal (Erasmus University, Rotterdam, The Netherlands) and donated to form the basis of these breeding colonies. The genotype of the breeding and experimental colonies was periodically verified using techniques based on PCR amplification of genomic DNA. 1) The rd mutation has introduced a novel (DdeI) restriction site to the rd locus; PCR primers specific to this portion of the gene were used to amplify a 298-bp portion of the gene covering this novel restriction site. The amplified DNA was subsequently digested with DdeI to test for the presence of the novel restriction site. 2) Testing for the rds mutation employed PCR primers specific for this allele. All three colonies were also tested for contamination with elements specific to the C57 strain genome by PCR amplification of microsatellite markers previously shown to be polymorphic between C57 and C3H strains (Harlan UK Ltd., Bicester, UK). All experimental animals were stably entrained to a 12-h light, 12-h dark cycle for at least 2 weeks before sampling. At around days of age (mean sem; rd/rd,82 0.5; rds/rds,89 2.1; wild-type, ) animals were sampled according to one of the following protocols. Diurnal profile To describe diurnal profiles of pineal melatonin content for each of the genotypes, pineals were collected from between six and eight animals of each genotype at 2-h intervals through the dark phase of the light-dark cycle and at one time point during the light phase, zeitgeber time (ZT) 8, 13, 15, 17, 19, 21, or 23 (where ZT12 is the time of lights off). Animals were killed by cervical dislocation and bilaterally enucleated under infrared illumination. Subsequently, the pineal was quickly removed under white fluorescent light and snap-frozen on dry ice. Pineals were stored at 80 C until assayed for melatonin content. Exposure to monochromatic 509-nm light Mice of all three genotypes were individually exposed to 15 min of defined irradiance monochromatic 509 nm light timed to start between ZT20 and ZT21 of the light-dark cycle. A remote source light-pulsing apparatus was employed, as previously described (15). Importantly, this apparatus employs a fiber optic cable to ensure that the mice are not exposed to any heat output of the light source during pulsing. The spectral transmission of this light source was controlled using a monochromatic filter (Oriel Corp., Stratford, CT; max, 509 nm; half-band width, 10 nm). Irradiance was controlled by the use of a series of neutral density impedance filters (Oriel Corp.) and was measured using an optical power meter (Graseby Optronics, Orlando, FL). Between six and eight animals from each genotype were exposed to monochromatic green light at irradiances of , , and microwatts ( W)/cm 2. In addition, six to eight animals from each genotype were placed in the pulsing apparatus for 15 min without exposure to light to act as experimental controls. At the end of the 15-min pulses, mice were removed from the pulsing apparatus under infrared illumination, their eyes were removed, and pineals were collected as described above. Enucleated mice Young adult mice from each of the three genotypes (five wild-type, three rd/rd, and two rds/rds) were bilaterally enucleated under halothane anesthesia. After recovering from the surgery, these animals were exposed to 12-h light, 12-h dark cycles. To assess circadian phase in these animals they were singly housed with free access to a running wheel from which circadian rhythms of wheel-running activity were monitored using a DataQuest system (Minimitter Co. Inc., Sunriver, OR). Enucleated mice did not entrain to the light-dark cycle, but after some transient arrhythmicity exhibited free running rhythms. When at least 14 days of stable circadian activity rhythm had been observed, mice were exposed to a 15-min 509-nm light pulse of 6 W/cm 2 using the apparatus described above. These pulses were timed with respect to the free running activity rhythm to start around circadian time (CT) 20, where CT12 is the time of activity onset. At the end of the pulse they were killed by cervical dislocation under infrared light, and pineals were collected. Melatonin assay A direct RIA was employed for the detection of melatonin concentrations in pineal homogenates (after Ref. 16). Briefly, single pineal glands were rapidly homogenized in assay buffer using ultrasound. These samples were then incubated in duplicate with a specific antiserum for melatonin raised in sheep (Stockgrand Ltd., Guildford, UK), and [ 3 H]melatonin (Amersham, Aylesbury, UK) was added. Free and antibody-bound fractions of melatonin were separated using dextran (Sigma Chemical Co., Poole, UK)-coated activated charcoal (Sigma Chemical Co.), and the amount of bound [ 3 H]melatonin was estimated using a scintillation counter (Fluoransafe, Fisher Life Science, Loughborough, UK; RackBeta, Wallac AC, Turku, Finland). The concentration of melatonin in the sample was estimated by comparison with standards of known melatonin (Sigma Chemical Co.) concentration. Quality control samples were included at 73 and 159 pg/ml; the intraassay coefficients of variation were 4.6% and 4.2%, and the interassay coefficients of variation were 10.5% and 7.6% for the lower and higher quality controls, respectively. The minimum detectable dose was 10 pg/ml. The assay was validated for use with mouse pineal homogenates by demonstrating parallelism over the range pg/ml. Statistical analysis The effect of increasing irradiances of 509 nm light on pineal melatonin content was tested in each genotype using a one-way ANOVA; post-hoc t tests were made against the unpulsed control group employing Bonferroni s correction. The t test comparisons were made between pineal melatonin content in enucleated mice exposed to 6 W/cm 2 of light and intact animals with either peak (not light pulsed) or completely suppressed (exposed to W/cm 2 light) melatonin. Because of the differences in estimating circadian phase from the external light cycle compared with the running wheel activity rhythm, pineal samples from intact animals at ZT19, -20, and -21 were compared with the enucleated animals exposed to 6 W/cm 2 at CT20. To check for differences between genotypes in peak melatonin production or the response to enucleation, one-way ANOVA tests were employed on the pineal melatonin contents of the nonlight-pulsed and enucleated mice. Statistical significance was defined as P Results Mice from all three genotypes showed a robust diurnal rhythmicity in pineal melatonin content under exposure to a 12-h light, 12-h dark cycle (Fig. 1). In each case melatonin content was basal during the light phase and reached a peak during the second half of the dark phase before falling again preceding lights on. Pineal melatonin content was stable between ZT19 and ZT21 in all three genotypes.

3 1522 PHOTIC MELATONIN SUPPRESSION IN rd AND rds MICE Endo 1999 Vol 140 No 4 There was no significant effect of genotype on the ability of the pineal to produce melatonin (by one-way ANOVA, P 0.05) in the absence of a light pulse; mean sem at ZT20, for wild-type, for rd/rd, and for rds/rds; Fig. 2A). All three genotypes showed an irradiance-dependent suppression of pineal melatonin content in response to 15 min of exposure to monochromatic 509 nm light (Fig. 2B; by one-way ANOVA, P 0.01). There was no indication that FIG. 1. Diurnal rhythms of pineal melatonin content in rd/rd, rds/ rds, and wild-type C3H mice. The mean SEM melatonin content, normalized against the mean value for each genotype at ZT21, is plotted by ZT (where ZT12 is the time of lights off). A portion of the corresponding light-dark cycle is depicted below; the shaded portion represents darkness. either rd/rd or rds/rds mice showed a decrease in sensitivity compared with wild-type animals, with all three genotypes showing suppression (P 0.01, by post-hoc t tests with Bonferroni s correction against unpulsed animals) in response to W/cm 2. Enucleation effectively abolished pineal melatonin suppression in response to light exposure. Animals of all three genotypes showed a similar response (by one-way ANOVA, P 0.05) to 6 W/cm 2 light after enucleation (Fig. 3). Pineal melatonin content was significantly (by t test, P ) elevated compared with that in intact animals exposed to W/cm 2 light. There was a modest reduction compared with intact animals not exposed to a light pulse, but this was not statistically significant (by t test, P 0.05). Discussion Despite the massive decrease in photoreceptive capacity in both rd/rd and rds/rds mice, neither of these genotypes showed evidence of a reduction in the sensitivity of pineal melatonin production to photic inhibition. These findings indicate that the photic pathway mediating pineal melatonin suppression is buffered against considerable loss of photoreceptive capacity. Moreover, the defined retinal phenotypes of these two mutants allow us to make two more specific conclusions: 1) as rd/rd mice lack functional rod photoreception, our results confirm that rod photoreceptors are not required for mediating this response; and 2) as rds/rds mice never develop rod or cone outer segments, it is clear that these structures are not required for photic suppression of pineal melatonin. The results of bilateral enucleation reported here confirm previous reports in mammals that the photoreceptors regulating the mammalian pineal are located in the eyes (1, 17). There has been a recent report suggesting that humans are FIG. 2. A, Melatonin content (mean SEM) of pineal glands collected between ZT20 and ZT 21 from rd/rd, rds/rds, and wild-type C3H mice without exposure to a light pulse. Pineal melatonin content was not significantly different in the three genotypes (by one-way ANOVA, P 0.05). B, Melatonin content (mean SEM) of pineal glands collected between ZT20 and ZT21 from rd/rd, rds/rds, and wild-type C3H mice following 15 min exposure to controlled irradiance monochromatic 509 nm light. All three genotypes showed an irradiancedependent suppression of pineal melatonin content (by one-way ANOVA, P 0.001), with W/cm 2 sufficient to induce significant suppression with respect to that in the unpulsed controls (by post-hoc t tests with Bonferroni s correction: **, P 0.01). FIG. 3. Melatonin content (mean SEM) of pineal glands from intact C3H mice (all genotypes) collected from ZT19 to ZT21 without exposure to light or from ZT20 to ZT21 after exposure to 15 min of W/cm nm light and from enucleated C3H mice (all genotypes) collected between CT20 and CT21 after exposure to 15 min of 6 W/cm nm light. The t test comparisons indicated that the pineal melatonin content of the enucleated mice was significantly (P 0.001) higher than that of the intact light-pulsed animals, but was not significantly different from that of the intact unpulsed animals (P 0.05). See text for discussion.

4 PHOTIC MELATONIN SUPPRESSION IN rd AND rds MICE 1523 capable of showing circadian phase shifts through an as yet undefined extraocular photoreceptor (18; for review, see Ref. 19). However, that study did not extend to an examination of pineal melatonin suppression, and here we report that after bilateral enucleation, melatonin was not significantly suppressed even by an irradiance more than 2 orders of magnitude higher than that capable of complete suppression in intact animals. Although not statistically significant, pineal melatonin content did appear moderately lower in enucleated animals exposed to bright light compared with that in intact, unpulsed mice. However, we do not consider that this supports the hypothesis of extraocular photoreception in mammals. It is most likely that this effect is accounted for by three difficulties associated with ensuring that pineals were collected at the same phase of the melatonin rhythm in free running (enucleated) and entrained (intact) animals. Firstly, different phase reference points (ZT12, lights off, and CT12, activity onset) were employed for intact and enucleated animals which may not be wholly comparable in terms of the circadian phase they represent or the accuracy with which they do this. Secondly, free running (enucleated) mice typically exhibited a period significantly different from 24 h. Finally, the phasing of melatonin production in mammals varies according to night length (for review, see Ref. 20) and may well have been altered during the several weeks following blinding. Although it seems clear that the photoreceptors regulating the mammalian pineal are exclusively ocular, their precise nature remains unknown. The murine retina contains three known photoreceptor cell types: rods with a maximal absorbance ( max) around 498 nm (21) and two populations of cone photoreceptors absorbing green ( max 508 nm) (22) and UV ( max 359 nm) (23) light. The rd mutation blocks functional phototransduction in rods (7, 8), and by 80 days of age, rd/rd mice lack rod photoreceptors (9). Consequently, the demonstration of melatonin suppression in rd/rd mice indicates that functional rod photoreceptors are not required for mediating this response. This finding implicates the green cones in regulation of the pineal, as these are the sole remaining photoreceptors in the rd/rd retina sensitive to 509 nm light. This hypothesis is currently under examination in our laboratory using mice lacking green cone photoreceptors. In addition, the involvement of some as yet unidentified nonrod-, non-cone-based photopigment cannot be excluded. The evidence suggests that rod photoreceptors are not required for the acute suppression of pineal melatonin. However, it remains possible that, where present in the wild-type mouse retina, they do contribute to this response. Although several investigators have reported significant suppression in rats or Syrian hamsters by light outside the normal sensitivity of rodopsin (24 26), the overall spectral sensitivity of pineal melatonin inhibition in these species is consistent with the involvement of a rod-like opsin photopigment (4 6). Similarly, action spectra for light-induced changes in electrical activity within the mammalian pineal suggest input from both rod and cone photoreceptors (27). These findings suggest that multiple photoreceptor types mediate pineal responses to light. Previously, Goto and Ebihara (14) have presented data from rd/rd mice that seem to support this hypothesis. They reported that although rd/rd mice are capable of showing pineal responses to white light, their sensitivity was greatly reduced compared with that of wild-type animals. However, that study compared C3H/He rd/rd mice with CBA/Ms wild types. Here we have demonstrated that when wild-type and rd/rd mice from the same (C3H) strain are compared, no decrease in sensitivity is evident. This finding indicates that a complete loss of rod phototransduction has no demonstrable effect on the sensitivity of pineal suppression. Consequently, our data suggest either that rod photoreceptors are not involved in regulating this response or that in their absence some other photoreceptor can completely compensate for their loss. Pineal melatonin production is under photic control via two independent mechanisms. In this report we have examined the acute effects of light on melatonin production. However, light also effects melatonin production by entraining the circadian clock in the SCN that drives the activity of the pineal. To date, both anatomical and experimental evidence supports the hypothesis that the same photoreceptors mediate these parallel irradiance-dependent processes. Both processes are intimately associated with the SCN (2, 28), which receive a direct retinal projection via the retinohypothalamic tract (29, 30) and, through variations in stimulatory input, regulate the activity of the pineal gland according both to circadian phase and environmental illumination (3). In Syrian hamsters, the spectral sensitivity of both pineal melatonin suppression (6, 24) and circadian phase shifting (31) has been examined. In each case, maximal responses were observed to light of around 500 nm, suggesting that these two processes are mediated by photoreceptors with a similar absorbance spectrum. However, although the spectral sensitivity of phase shifting and melatonin suppression responses in this species are similar, the absolute sensitivities of these two tasks are significantly different, with melatonin suppression sensitive to irradiances 1.4 log units lower than those required for phase shifts (32). Whether these differences in sensitivity are caused by differential processing of output from the same photoreceptors or the use of different photoreceptors is unknown. Here, we have shown that both rd/rd and rds/rds mice show unattenuated photic inhibition of pineal melatonin. Previous reports confirm that these genotypes also exhibit unattenuated phase shifts in response to appropriate light pulses (15, 33). Together these data suggest that whatever the ocular elements mediating these two irradiance-dependent responses, both are spared by the massive photoreceptor degenerations caused by the rd and rds mutations. Although it seems likely that different irradiance detection tasks employ the same ocular photoreceptors, the hypothesis that these are the same photoreceptors known to mediate vision (i.e. rods and cones) is currently unproven. The presence of a dedicated retinohypothalamic tract in mammals indicates that at some structural level, visual and irradiance detection functions are separated. Whether this separation extends to the use of different photoreceptor cells and pigments remains unknown. A previous study using a shuttlebox classical conditioning paradigm has suggested that aged rd/rd mice are visually blind (34). Consequently, the demonstration shown here and reported previously (15, 34) that irradiance detection is not significantly impaired in rd/rd

5 1524 PHOTIC MELATONIN SUPPRESSION IN rd AND rds MICE Endo 1999 Vol 140 No 4 mice suggests that visual blindness is not necessarily associated with impaired irradiance detection. This conclusion is supported by reports of visually blind humans showing responses consistent with the presence of functioning irradiance detection (35). It seems clear from these various reports that irradiance detection functions are buffered against a loss of photoreceptive capacity that is sufficient to induce visual blindness. Whether the basis of this buffering is the use of a dedicated irradiance detection photoreceptor or an up-regulation of input from the remaining conventional photoreceptors remains to be determined. In summary, neither rd nor rds mutations induced a significant decrease in the sensitivity of pineal melatonin production to photic inhibition. By contrast, enucleation abolished this response. Thus, our results confirm that the photoreceptors mediating the acute suppression of pineal melatonin are located in the eye. In addition, they demonstrate that at least in mice, neither rod photoreception nor rod or cone outer segments are required for photic regulation of the pineal. References 1. Klein DC, Weller JL 1972 Rapid light-induced decrease in pineal serotonin N-acetyltransferase activity. Science 177: Klein DC, Moore RY 1979 Pineal N-acetyltransferase and hydroxyindole-omethyl-transferase: control by the retino-hypothalamic tract and the suprachiasmatic nuclei. Brain Res 174: Romero JA, Zatz M, Axelrod J Adrenergic stimulation of pineal N- acetyltranserase: adenosine 3 :5 -cyclic monophosphate stimulates both RNA and protein synthesis. Proc Natl Acad Sci USA 72: Cardinali DP, Larin F, Wurtman RJ 1972 Control of the rat pineal gland by light spectra. Proc Natl Acad Sci USA 69: Bronstein DM, Jacobs GH, Haak KA, Neitz J, Lytle LD 1987 Action spectrum of the retinal mechanism mediating nocturnal light-induced suppression of rat pineal gland N-acetyltransferase. Brain Res 406: Brainard GC, Richardson BA, King TS, Reiter RJ 1984 The influence of different light spectra on the suppression of pineal melatonin content in the Syrian hamster. Brain Res 294: Bowes C, Li T, Danciger M, Baxter LC, Applebury ML, Farber DB 1990 Retinal degeneration in the rd mouse is caused by a defect in the subunit of rod cgmp-phosphodiesterase. Nature 347: Pittler SJ, Baehr W 1991 Identification of a nonsense mutation in the rod photoreceptor cgmp phosphodiesterase -subunit gene of the rd mouse. Proc Natl Acad Sci USA 88: Carter-Dawson LD, LaVail MM, Sidman RL 1978 Differential effect of the rd mutation on rods and cones in the mouse retina. Invest Ophthalmol Vis Sci 17: Travis GH, Brennan MB, Danielson PE, Kozak CS, Sutcliffe JG 1989 Identification of a photoreceptor-specific mrna encoded by the gene responsible for retinal degeneration slow (rds). Nature 338: Travis GH, Sutcliffe JG, Bok D 1991 The retinal degeneration slow (rds) gene product is a photoreceptor disk membrane-associated glycoprotein. Neuron 6: Connell G, Bascom R, Molday L, Reid D, McInnes RR, Molday RS 1991 Photoreceptor peripherin is the normal product of the gene responsible for retinal degeneration in the rds mouse. Proc Natl Acad Sci USA 88: Sanyal S, de Ruiter A, Hawkins RK 1980 Development and degeneration of retina in rds mutant mice: light microscopy. J Comp Neurol 194: Goto M, Ebihara S 1990 The influence of different light intensities on pineal melatonin content in the retinal degenerate C3H mouse and the normal CBA mouse. Neurosci Lett 108: Foster RG, Provencio I, Hudson D, Fiske S, De Grip W, Menaker M 1991 Circadian photoreception in the retinally degenerate mouse (rd/rd). J Comp Physiol 169: Fraser S, Cowen P, Franklin M, Franey C, Arendt J 1983 Direct radioimmunoassay for melatonin in plasma. Clin Chem 29: Lockley S, Skene D, Thapan K, English J, Ribeiro D, Haimov I, Hampton S, Middleton B, von Schantz M, Arendt J 1998 Extraocular light exposure does not suppress plasma melatonin in humans. J Clin Endocrinol Metab 83: Campbell S, Murphy P 1998 Extraocular phototransduction in humans. Science 179: Foster RG 1998 Shedding light on the biological clock. Neuron 20: Arendt J 1995 Melatonin and the Mammalian Pineal Gland. Chapman and Hall, London 21. Bridges CDB 1959 The visual pigments of some common laboratory animals. Nature 184: Sun H, Macke JP, Nathans J 1997 Mechanisms of spectral tuning in the mouse green cone pigment. Proc Natl Acad Sci USA 94: Jacobs GH, Neitz J, Deegan JD 1991 Retinal receptors in rodents maximally sensitive to ultraviolet light. Nature 353: Podolin PL, Rollag MD, Brainard GC 1987 The suppression of nocturnal pineal melatonin in the Syrian hamster: dose-response curves at 500 and 360 nm. Endocrinology 121: Sun J-H, Yaga K, Reiter RJ, Garza M, Manchester LC, Tan D-X, Poeggeler B 1993 Reduction in pineal N-acetyltransferase activity and pineal and serum melatonin levels in rats after exposure to red light at night. Neurosci Lett 149: Poeggeler BH, Barlow-Walden LR, Reiter RJ, Saarela S, Menedez-Pelaez A, Yaga K, Manchester LC, Chen L-D, Tan D-X 1995 Red-light-induced suppression of melatonin synthesis is mediated by N-methyl-d-aspartate receptor activation in retinally normal and retinally degenerate rats. J Neurobiol 28: Thiele G, Meissl H 1987 Action spectra of the lateral eyes recorded from mammalian pineal glands. Brain Res 424: Kornhauser JM, Nelson DE, Mayo KE, Takahashi JS 1990 Photic and circadian regulation of c-fos gene expression in the hamster suprachiasmatic nucleus. Neuron 5: Moore RY, Lenn NJ 1972 A retinohypothalamic projection in the rat. J Comp Neurol 146: Provencio I, Cooper HM, Foster RG 1998 Retinal projections in mice with inherited retinal degeneration: implications for circadian photoentrainment. J Comp Neurol 395: Takahashi J, DeCoursey P, Bauman L, Menaker M 1984 Spectral sensitivity of a novel photoreceptive system mediating entrainment of mammalian circadian rhythms. Nature 308: Nelson DE, Takahashi JS 1991 Sensitivity and integration in a visual pathway for circadian entrainment in the hamster (Mesocricetus auratus). J Physiol 439: Argamaso SM, Froehlich AC, McCall MA, Nevo E, Provencio I, Foster RG 1995 Photopigments and circadian systems of vertebrates. Biophys Chem 56: Provencio I, Wong S, Lederman AB, Argamaso SM, Foster RG 1994 Visual and circadian responses to light in aged retinally degenerate mice. Vision Res 34: Czeisler CA, Shanahan TL, Klerman EB, Martens H, Brotman DJ, Emens JS, Klein T, Rizzo JF 1995 Suppression of melatonin secretion in some blind patients by exposure to bright light. N Engl J Med 332:6 11