- Email: [email protected]
Circadian biology: the physiology of inner retinal photoreceptors
Current Biology, Vol. 13, R667–R669, September 2, 2003, ©2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/S0960-9822(03)00603-1
Circadian Biology: The Physiology of Inner Retinal Photoreceptors Ron Douglas
Non-traditional photoreceptors detect overall irradiance in the vertebrate retina. Such cells in the mouse inner retina show increased intracellular Ca2+ levels following illumination. Neurons in the outer retina of fish also display characteristics appropriate for an irradiance detector.
Annoyingly, I wake up at the same time every morning no matter what time I went to bed. This, and many other daily routines, is initiated by cells within my suprachiasmatic nucleus (SCN). Such ‘clocks’ underlie many aspects of an animal’s physiology and behaviour, and while it might not be a matter of life and death when I awake, for many animals, being prepared for regularly predictable events is vital for survival. The eyes of fish, for example, take about 20 minutes to change from night-time mode to daylight vision. An animal whose eyes are prepared for the coming of dawn will be able to avoid a predator and catch its prey when the sun rises more efficiently than one who simply reacts to the light. Internal clocks give animals such a huge selective advantage that it is no surprise they are found throughout the animal kingdom and regulate a myriad responses. Many biological clocks have a period of around 24 hours aligned to the daily light–dark cycle. No clock is perfect, however, and in time a biological clock will, if not reset, begin to run out of synchrony with the Earth’s rotation. Light therefore acts as a signal to ensure the clock remains perfectly attuned to the 24 hour period. What photoreceptors are responsible for such ‘photoentrainment’? Two papers [1,2] published recently in Current Biology make significant contributions to our understanding of the nature of these circadian photoreceptors. Many animals have a variety of such photoreceptors both within the eye and the brain. Although a widely publicised study erroneously [3] suggested that illumination behind the knee might entrain the human circadian clock, mammals appear to rely solely on ocular photoreceptors to supply information about environmental lighting. Benjamin Franklin wrote “In this world nothing is certain but death and taxes”. A visual scientist might have added “and that rods and cones are the only retinal photoreceptors”. This view is changing: an avalanche of data generated during the last five years has shown this fundamental tenet of biology to be false. In retrospect, however, the discovery of non-rod, non-cone retinal photoreceptors is not surprising. Applied Vision Research Centre, Dept. Optometry & Visual Science, City University, Northampton Square, London EC1V 0HB, UK.
Dispatch
The eye obviously provides a detailed threedimensional representation of the environment. But it also measures the overall level of irradiance, in order to adjust the bodies endogenous physiological rhythms to the light–dark cycle and regulate the size of the pupil, and this may be a more fundamental function of the eye than ‘image forming vision’. There is no reason to suppose that such ‘non-image forming vision’ uses the same photoreceptors as other visual tasks. In fact, these two forms of vision might require receptors with quite different properties. Traditional photoreceptors, for example, are optimised to respond to change, adapting rapidly to prolonged illumination, while irradiance detectors might be better served by longer integration times. Similarly, while cones have small receptive fields to optimise acuity, a receptor that measures overall light levels should have a larger dendritic spread. It would be logical to have a separate system for detecting irradiance, that does not compromise the high spatial and temporal resolution of the traditional image-forming visual pathway. Nonetheless, it came as a surprise to most people that ‘blind’ transgenic mice without functional rods or cones exhibit near normal photoentrainment of locomotor rhythms [4], suppression of nocturnal pineal melatonin secretion by light [5] and pupil light responses [6]. The suggestion was not that rods and cones do not have a role to play in these behaviours, but that there is an additional ocular photoreceptor capable of initiating tasks requiring overall irradiance detection. The shape of the action spectrum of the pupil light response in rodless, coneless animals indicated that the photopigment within such receptors was, like the traditional visual pigments, rhodopsinlike, based on a protein (opsin) linked to a vitamin A1derived chromophore (retinal) [6]. The most likely candidate to emerge as the circadian retinal photopigment is melanopsin, which was first described in the dermal melanophores of Xenopus laevis [7] and is present in a subset of retinal ganglion cells that form a net across the inner retina of mice [7,8]. These cells, unlike most retinal ganglion cells, project not to the main visual areas that subserve conventional vision, but to those brain centres generating circadian rhythms and the pupil light response [9]. They are also inherently photosensitive, responding to light even when deprived of all input from rods and cones [10]. Sekaran et al. [1] have taken the study of these photosensitive retinal ganglion cells one step further, using a technique that allows the visualisation of intracellular calcium levels by fluorescence (Figure 1). This is a natural evolution of previous physiological work [10,11], as it allows the simultaneous study of hundreds of cells. The use of transgenic mice lacking functional rods and cones avoids the use of extensive pharmacological or surgical manipulation to isolate photosensitive retinal ganglion cells. Light causes an
Dispatch R668
Figure 1. A sequence of false colour calcium images following illumination of an intrinsically light sensitive cell in the inner retina of a rodless-coneless mouse. The integrated calcium fluorescence signal for this cell in response to light is also shown. (Images courtesy Mark Hankins.)
increase of intracellular Ca2+ in around 2.7% of cells in the retinal ganglion cell layer, indicating that, as in invertebrate phototransduction, an increase in intracellular Ca2+ plays a role in the response of these cells to light. The percentage of inner retinal photosensitive cells seen is almost twice that reported previously, probably because earlier studies concentrated on cells projecting only to the SCN, while the current work highlights the entire network. Morphologically and physiologically, these cells form an extensive, gap junction-linked, syncytium in the inner retina. Intriguingly, there are three types of Ca2+ influx, suggestive of distinct functional classes of light sensitive cell in the inner retina. Whether these project to separate areas of the brain or are involved in different forms of non-imaging forming vision is unclear. While melanopsin undoubtedly has a major role to play in a mammalian retinal irradiance detecting system, precisely what it does is uncertain. It might itself not be the photopigment, but rather a necessary protein for that pigment to function. Like some other opsins, melanopsin may, for example, function as a photoisomerase needed for visual pigment regeneration [12]. There is evidence that other pigments, such as the flavin-based cryptochromes, are also involved in vertebrate irradiance detection [13]. But although circadian behaviours and pupil light responses of mice lacking the melanopsin system are only reduced and not eliminated [12,14,15], in triple-knockout mice which lack functional rod, cone and melanopsin systems, these responses are virtually abolished [16], making the involvement of an additional pigment unlikely.
In addition to melanopsin, a variety of other novel retinal opsins have been identified. One of these, VA opsin, found in fish, actually provided the first evidence that retinal photosensitivity might not be confined to rods and cones [17]. While the mammalian retinal melanopsin-based system is fairly well characterised, little is known about photosensitivity involving other novel opsins. Jenkins et al. [2] describe a population of teleost melanopsin and VA opsin containing retinal horizontal cells, which display a novel delayed depolarising response to the cessation of the light stimulus. This response is characterised by long integration times, persists when the conventional photoreceptor inputs are saturated by light, and has a spectral sensitivity suggestive of visual pigment distinct from rods and cones. While there is no direct proof that these cells are intrinsically photosensitive, the data suggest they form the basis of a system of non-rod non-cone irradiance detection similar to the melanopsin-based network described for inner retinal neurons in rodents. The spectral characteristics of all non-rod non-cone mediated visual responses are remarkably similar. The action spectra for the pupil light response [6] and phase shifting of activity rhythms [16] in rodless coneless mice and the intrinsic retinal ganglion cell photosensitivity of rats [10] are all virtually identical — their action spectra all have maxima at around 480 nm — which suggests that the same rhodopsin-like photoreceptor can trigger all ‘non-visual’ responses to light in rodents. That this conservatism persists throughout vertebrates is suggested by the new study of photosensitive fish horizontal cells [2], whose novel response to light is also dependant on a rhodopsin with an absorption maximum of 477 nm. While several of my colleagues grudgingly accept the existence of this novel photosensitive pathway in rodents, they seem unwilling to do so for primates. But the action spectra of both light-induced melatonin suppression [18,19] and diurnal alterations in retinal electrophysiology [20] in man suggest that they too are mediated by a non-rod non-cone photoreceptor, whose spectral sensitivity is virtually identical to that observed in rodents and fish. Furthermore, several speakers at the recent ARVO meeting reported that the primate retina also has melanopsin-containing retinal ganglion cells with morphology and electrophysiological responses suggesting they too serve as photoreceptors for pathways underlying circadian and pupillary responses. How these novel retinal photoreceptors interact with the traditional neural pathways driven by rods and cones, is perhaps the next big question to answer. References 1. Sekaran, S., Foster, R.G., Lucas, R.J., and Hankins, M.W. (2003). Calcium imaging reveals a network of intrinsically light sensitive inner retinal neurones. Curr. Biol. 13, 1290-1298. 2. Jenkins, A., Muñoz, M., Tarttelin, E.E., Bellingham, J., Foster, R.G., and Hankins, M.W. (2003). VA-opsin, melanopsin, and an inherent light response within retinal interneurones. Curr. Biol. 13, 12691278. 3. Wright, K.P., and Czeisler, C.A. (2002). Absence of circadian phase resetting in response to bright light behind the knee. Science 297, 571.
Current Biology R669
4. Freedman, M.S., Lucas, R.J., Soni, B., von Schantz, M., Muñoz, M., David-Gray, Z., and Foster, R.G. (1999). Regulation of mammalian circadian behaviour by non-rod, non-cone, ocular photoreceptors. Science 284, 502-504. 5. Lucas, R.J., Freedman, M.S., Muñoz, M., Fernández, J.-M., and Foster, R.G. (1999). Regulation of the mammalian pineal by non-rod, non-cone, ocular photoreceptors. Science 284, 505-507. 6. Lucas, R.J., Douglas, R.H., and Foster, R.G. (2001). Characterization of an ocular photopigment capable of driving pupillary constriction in mice. Nature Neurosci. 4, 621-626. 7. Provencio, I., Cooper, H.M., and Foster, R.G. (1998). Retinal projections in mice with inherited retinal degeneration: implications for circadian photoentrainment. J. Comp. Neurol. 395, 417-439. 8. Provencio, I., Rodruguez, I.R., Jiang, G., Hayes, W.P., Moreira, E.F., and Rollag, M.D. (2000). A novel human opsin in the inner retina. J. Neurosci. 20, 600-605. 9. Hattar, S., Liao, H.-W., Takao, M., Berson, D.M., and Yau, K.-W. (2002). Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 295, 1065-1070. 10. Berson, D.M., Dunn, F.A., and Takao, M. (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295, 1070-1073. 11. Warren, E.J. Allen, C.N., Brown, R.L., and Robinson, D.W. (2003). Intrinsic light responses of retinal ganglion cells projecting to the circadian system. Eur. J. Neurosci. 17, 1727-1735. 12. Lucas, R.J., Hattar, S., Takao, M., Berson, D.M., Foster, R.G., and Yau, K.-W. (2003). Diminished pupillary light reflex at high irradiances in melanopsin-knockout mice. Science 299, 245-247. 13. van Gelder, R.N., Wee, R., Lee, J.A., and Tu, D.C. (2003). Reduced pupillary light responses in mice lacking cryptochromes. Science 299, 222. 14. Ruby, N.F., Brennan, T.J., Xie, X., Cao, V., Franken, P., Heller, H.C., and O'Hara, B.F. (2002). Role of melanopsin in circadian responses to light. Science 298, 2211-2213. 15. Panda, S., Sato, T.K., Castrucci, A.M., Rollar, M.D., DeGrip, W.J., Hogenesche, J.B., Provencio, I., and Kay, S.A. (2002). Melanopsin (Opn4) requirement for normal light-induced circadian phase shifting. Science 298, 2213-2216. 16. Hattar, S., Lucas, R.J., Mrosovsky, N., Thompson, S., Douglas, R.H., Hankins, M.W., Lem, J., Biel, M., Hofmann, F., Foster, R.G., and Yau, K.-W. (2003). Melanopsin and rod-cone photoreceptive systems account for all major accessory visual functions in mice. Nature 424, 76-81. 17. Soni, B.G., Philp, A.R., Foster, R.G., and Knox, B.E. (1998). Novel retinal photoreceptors. Nature 394, 27-28. 18. Thapan, K., Arendt, J., and Skene, D.J. (2001). An action spectrum for melatonin suppression; evidence for a novel non-rod, non-cone photoreceptor system in humans. J. Physiol. Lond. 535, 261-267. 19. Brainard, G.C., Hanifin, J.P., Greeson, J.M., Byrne, B., Glickman, G., Gerner, E., and Rollag, M.D. (2001). Action spectrum for melatonin regulation in humans; evidence for a novel circadian photoreceptor. J. Neurosci. 21, 6405-6412. 20. Hankins, M.W., and Lucas, R.J. (2002). The primary visual pathway in humans is regulated according to long-term light exposure through the action of a nonclassical photopigment. Curr. Biol. 12, 191-198.
Circadian Biology: The Physiology of Inner Retinal Photoreceptors Ron Douglas
Non-traditional photoreceptors detect overall irradiance in the vertebrate retina. Such cells in the mouse inner retina show increased intracellular Ca2+ levels following illumination. Neurons in the outer retina of fish also display characteristics appropriate for an irradiance detector.
Annoyingly, I wake up at the same time every morning no matter what time I went to bed. This, and many other daily routines, is initiated by cells within my suprachiasmatic nucleus (SCN). Such ‘clocks’ underlie many aspects of an animal’s physiology and behaviour, and while it might not be a matter of life and death when I awake, for many animals, being prepared for regularly predictable events is vital for survival. The eyes of fish, for example, take about 20 minutes to change from night-time mode to daylight vision. An animal whose eyes are prepared for the coming of dawn will be able to avoid a predator and catch its prey when the sun rises more efficiently than one who simply reacts to the light. Internal clocks give animals such a huge selective advantage that it is no surprise they are found throughout the animal kingdom and regulate a myriad responses. Many biological clocks have a period of around 24 hours aligned to the daily light–dark cycle. No clock is perfect, however, and in time a biological clock will, if not reset, begin to run out of synchrony with the Earth’s rotation. Light therefore acts as a signal to ensure the clock remains perfectly attuned to the 24 hour period. What photoreceptors are responsible for such ‘photoentrainment’? Two papers [1,2] published recently in Current Biology make significant contributions to our understanding of the nature of these circadian photoreceptors. Many animals have a variety of such photoreceptors both within the eye and the brain. Although a widely publicised study erroneously [3] suggested that illumination behind the knee might entrain the human circadian clock, mammals appear to rely solely on ocular photoreceptors to supply information about environmental lighting. Benjamin Franklin wrote “In this world nothing is certain but death and taxes”. A visual scientist might have added “and that rods and cones are the only retinal photoreceptors”. This view is changing: an avalanche of data generated during the last five years has shown this fundamental tenet of biology to be false. In retrospect, however, the discovery of non-rod, non-cone retinal photoreceptors is not surprising. Applied Vision Research Centre, Dept. Optometry & Visual Science, City University, Northampton Square, London EC1V 0HB, UK.
Dispatch
The eye obviously provides a detailed threedimensional representation of the environment. But it also measures the overall level of irradiance, in order to adjust the bodies endogenous physiological rhythms to the light–dark cycle and regulate the size of the pupil, and this may be a more fundamental function of the eye than ‘image forming vision’. There is no reason to suppose that such ‘non-image forming vision’ uses the same photoreceptors as other visual tasks. In fact, these two forms of vision might require receptors with quite different properties. Traditional photoreceptors, for example, are optimised to respond to change, adapting rapidly to prolonged illumination, while irradiance detectors might be better served by longer integration times. Similarly, while cones have small receptive fields to optimise acuity, a receptor that measures overall light levels should have a larger dendritic spread. It would be logical to have a separate system for detecting irradiance, that does not compromise the high spatial and temporal resolution of the traditional image-forming visual pathway. Nonetheless, it came as a surprise to most people that ‘blind’ transgenic mice without functional rods or cones exhibit near normal photoentrainment of locomotor rhythms [4], suppression of nocturnal pineal melatonin secretion by light [5] and pupil light responses [6]. The suggestion was not that rods and cones do not have a role to play in these behaviours, but that there is an additional ocular photoreceptor capable of initiating tasks requiring overall irradiance detection. The shape of the action spectrum of the pupil light response in rodless, coneless animals indicated that the photopigment within such receptors was, like the traditional visual pigments, rhodopsinlike, based on a protein (opsin) linked to a vitamin A1derived chromophore (retinal) [6]. The most likely candidate to emerge as the circadian retinal photopigment is melanopsin, which was first described in the dermal melanophores of Xenopus laevis [7] and is present in a subset of retinal ganglion cells that form a net across the inner retina of mice [7,8]. These cells, unlike most retinal ganglion cells, project not to the main visual areas that subserve conventional vision, but to those brain centres generating circadian rhythms and the pupil light response [9]. They are also inherently photosensitive, responding to light even when deprived of all input from rods and cones [10]. Sekaran et al. [1] have taken the study of these photosensitive retinal ganglion cells one step further, using a technique that allows the visualisation of intracellular calcium levels by fluorescence (Figure 1). This is a natural evolution of previous physiological work [10,11], as it allows the simultaneous study of hundreds of cells. The use of transgenic mice lacking functional rods and cones avoids the use of extensive pharmacological or surgical manipulation to isolate photosensitive retinal ganglion cells. Light causes an
Dispatch R668
Figure 1. A sequence of false colour calcium images following illumination of an intrinsically light sensitive cell in the inner retina of a rodless-coneless mouse. The integrated calcium fluorescence signal for this cell in response to light is also shown. (Images courtesy Mark Hankins.)
increase of intracellular Ca2+ in around 2.7% of cells in the retinal ganglion cell layer, indicating that, as in invertebrate phototransduction, an increase in intracellular Ca2+ plays a role in the response of these cells to light. The percentage of inner retinal photosensitive cells seen is almost twice that reported previously, probably because earlier studies concentrated on cells projecting only to the SCN, while the current work highlights the entire network. Morphologically and physiologically, these cells form an extensive, gap junction-linked, syncytium in the inner retina. Intriguingly, there are three types of Ca2+ influx, suggestive of distinct functional classes of light sensitive cell in the inner retina. Whether these project to separate areas of the brain or are involved in different forms of non-imaging forming vision is unclear. While melanopsin undoubtedly has a major role to play in a mammalian retinal irradiance detecting system, precisely what it does is uncertain. It might itself not be the photopigment, but rather a necessary protein for that pigment to function. Like some other opsins, melanopsin may, for example, function as a photoisomerase needed for visual pigment regeneration [12]. There is evidence that other pigments, such as the flavin-based cryptochromes, are also involved in vertebrate irradiance detection [13]. But although circadian behaviours and pupil light responses of mice lacking the melanopsin system are only reduced and not eliminated [12,14,15], in triple-knockout mice which lack functional rod, cone and melanopsin systems, these responses are virtually abolished [16], making the involvement of an additional pigment unlikely.
In addition to melanopsin, a variety of other novel retinal opsins have been identified. One of these, VA opsin, found in fish, actually provided the first evidence that retinal photosensitivity might not be confined to rods and cones [17]. While the mammalian retinal melanopsin-based system is fairly well characterised, little is known about photosensitivity involving other novel opsins. Jenkins et al. [2] describe a population of teleost melanopsin and VA opsin containing retinal horizontal cells, which display a novel delayed depolarising response to the cessation of the light stimulus. This response is characterised by long integration times, persists when the conventional photoreceptor inputs are saturated by light, and has a spectral sensitivity suggestive of visual pigment distinct from rods and cones. While there is no direct proof that these cells are intrinsically photosensitive, the data suggest they form the basis of a system of non-rod non-cone irradiance detection similar to the melanopsin-based network described for inner retinal neurons in rodents. The spectral characteristics of all non-rod non-cone mediated visual responses are remarkably similar. The action spectra for the pupil light response [6] and phase shifting of activity rhythms [16] in rodless coneless mice and the intrinsic retinal ganglion cell photosensitivity of rats [10] are all virtually identical — their action spectra all have maxima at around 480 nm — which suggests that the same rhodopsin-like photoreceptor can trigger all ‘non-visual’ responses to light in rodents. That this conservatism persists throughout vertebrates is suggested by the new study of photosensitive fish horizontal cells [2], whose novel response to light is also dependant on a rhodopsin with an absorption maximum of 477 nm. While several of my colleagues grudgingly accept the existence of this novel photosensitive pathway in rodents, they seem unwilling to do so for primates. But the action spectra of both light-induced melatonin suppression [18,19] and diurnal alterations in retinal electrophysiology [20] in man suggest that they too are mediated by a non-rod non-cone photoreceptor, whose spectral sensitivity is virtually identical to that observed in rodents and fish. Furthermore, several speakers at the recent ARVO meeting reported that the primate retina also has melanopsin-containing retinal ganglion cells with morphology and electrophysiological responses suggesting they too serve as photoreceptors for pathways underlying circadian and pupillary responses. How these novel retinal photoreceptors interact with the traditional neural pathways driven by rods and cones, is perhaps the next big question to answer. References 1. Sekaran, S., Foster, R.G., Lucas, R.J., and Hankins, M.W. (2003). Calcium imaging reveals a network of intrinsically light sensitive inner retinal neurones. Curr. Biol. 13, 1290-1298. 2. Jenkins, A., Muñoz, M., Tarttelin, E.E., Bellingham, J., Foster, R.G., and Hankins, M.W. (2003). VA-opsin, melanopsin, and an inherent light response within retinal interneurones. Curr. Biol. 13, 12691278. 3. Wright, K.P., and Czeisler, C.A. (2002). Absence of circadian phase resetting in response to bright light behind the knee. Science 297, 571.
Current Biology R669
4. Freedman, M.S., Lucas, R.J., Soni, B., von Schantz, M., Muñoz, M., David-Gray, Z., and Foster, R.G. (1999). Regulation of mammalian circadian behaviour by non-rod, non-cone, ocular photoreceptors. Science 284, 502-504. 5. Lucas, R.J., Freedman, M.S., Muñoz, M., Fernández, J.-M., and Foster, R.G. (1999). Regulation of the mammalian pineal by non-rod, non-cone, ocular photoreceptors. Science 284, 505-507. 6. Lucas, R.J., Douglas, R.H., and Foster, R.G. (2001). Characterization of an ocular photopigment capable of driving pupillary constriction in mice. Nature Neurosci. 4, 621-626. 7. Provencio, I., Cooper, H.M., and Foster, R.G. (1998). Retinal projections in mice with inherited retinal degeneration: implications for circadian photoentrainment. J. Comp. Neurol. 395, 417-439. 8. Provencio, I., Rodruguez, I.R., Jiang, G., Hayes, W.P., Moreira, E.F., and Rollag, M.D. (2000). A novel human opsin in the inner retina. J. Neurosci. 20, 600-605. 9. Hattar, S., Liao, H.-W., Takao, M., Berson, D.M., and Yau, K.-W. (2002). Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 295, 1065-1070. 10. Berson, D.M., Dunn, F.A., and Takao, M. (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295, 1070-1073. 11. Warren, E.J. Allen, C.N., Brown, R.L., and Robinson, D.W. (2003). Intrinsic light responses of retinal ganglion cells projecting to the circadian system. Eur. J. Neurosci. 17, 1727-1735. 12. Lucas, R.J., Hattar, S., Takao, M., Berson, D.M., Foster, R.G., and Yau, K.-W. (2003). Diminished pupillary light reflex at high irradiances in melanopsin-knockout mice. Science 299, 245-247. 13. van Gelder, R.N., Wee, R., Lee, J.A., and Tu, D.C. (2003). Reduced pupillary light responses in mice lacking cryptochromes. Science 299, 222. 14. Ruby, N.F., Brennan, T.J., Xie, X., Cao, V., Franken, P., Heller, H.C., and O'Hara, B.F. (2002). Role of melanopsin in circadian responses to light. Science 298, 2211-2213. 15. Panda, S., Sato, T.K., Castrucci, A.M., Rollar, M.D., DeGrip, W.J., Hogenesche, J.B., Provencio, I., and Kay, S.A. (2002). Melanopsin (Opn4) requirement for normal light-induced circadian phase shifting. Science 298, 2213-2216. 16. Hattar, S., Lucas, R.J., Mrosovsky, N., Thompson, S., Douglas, R.H., Hankins, M.W., Lem, J., Biel, M., Hofmann, F., Foster, R.G., and Yau, K.-W. (2003). Melanopsin and rod-cone photoreceptive systems account for all major accessory visual functions in mice. Nature 424, 76-81. 17. Soni, B.G., Philp, A.R., Foster, R.G., and Knox, B.E. (1998). Novel retinal photoreceptors. Nature 394, 27-28. 18. Thapan, K., Arendt, J., and Skene, D.J. (2001). An action spectrum for melatonin suppression; evidence for a novel non-rod, non-cone photoreceptor system in humans. J. Physiol. Lond. 535, 261-267. 19. Brainard, G.C., Hanifin, J.P., Greeson, J.M., Byrne, B., Glickman, G., Gerner, E., and Rollag, M.D. (2001). Action spectrum for melatonin regulation in humans; evidence for a novel circadian photoreceptor. J. Neurosci. 21, 6405-6412. 20. Hankins, M.W., and Lucas, R.J. (2002). The primary visual pathway in humans is regulated according to long-term light exposure through the action of a nonclassical photopigment. Curr. Biol. 12, 191-198.