| OCR Text |
Show STATE OF THE ART Intrinsically Photosensitive Retinal Ganglion Cells Aki Kawasaki, MD and Randy H. Kardon, MD, PhD Abstract: The recent discovery of melanopsin-expressing retinal ganglion cells that mediate the pupil light reflex has provided new insights into how the pupil responds to different properties of light. These ganglion cells are unique in their ability to transduce light into electrical energy. There are parallels between the electrophysiologic behavior of these cells in primates and the clinical pupil response to chromatic stimuli. Under photopic conditions, a red light stimulus produces a pupil constriction mediated predominantly by cone input via trans-synaptic activation of melanopsin- expressing retinal ganglion cells, whereas a blue light stimulus at high intensity produces a steady- state pupil constriction mediated primarily by direct intrinsic photoactivation of the melanopsin- expressing ganglion cells. Preliminary data in humans suggest that under photopic conditions, cones primarily drive the transient phase of the pupil light reflex, whereas intrinsic activation of the melanopsin- expressing ganglion cells contributes heavily to sustained pupil constriction. The use of chromatic light stimuli to elicit transient and sustained pupil light reflexes may become a clinical pupil test that allows differentiation between disorders affecting photoreceptors and those affecting retinal ganglion cells. (/ Neuro- Ophthalmol 2007; 27: 195- 204) Department of Neuro- Ophthalmology ( AK), Hopital Ophtalmique Jules Gonin and University Eye Clinics of Lausanne, Lausanne, Switzerland; and Department of Ophthalmology and Visual Science ( RHK), University of Iowa Hospitals and Clinics and Veterans Administration Hospitals, Iowa City, Iowa. The authors have no proprietary interests in any of the instruments used in this study This work was supported by an unrestricted grant from Research to Prevent Blindness and a Merit Review Grant from the Veterans Administration. Dr. Kardon is a Lew Wasserman Scholar, Research to Prevent Blindness, New York. Address correspondence to Aki Kawasaki, MD, Hopital Ophtalmique Jules Gonin, Avenue de France 15, Lausanne 1004, Switzerland; E- mail: aki. kawasaki@ ophtal. ud. ch n 1998, Provencio et al ( 1) identified a novel opsin that they termed melanop sin from the dermal melanophore cells of the frog in which it was first isolated. It was not long thereafter that these investigators demonstrated the presence of this opsin in a small subpopulation of retinal ganglion cells in vertebrates, including humans ( 2- 4). The remarkable feature of melanopsin is that it functions as a photopigment and confers intrinsic photosensitivity to cells that express it ( 5,6). In 2002, Berson et al ( 7) unequivocally demonstrated that melanopsin- expressing retinal ganglion cells are capable of depolarization to light stimulation in the absence of any synaptic input from rods and cones. In other words, these ganglion cells can function as independent photoreceptors. This concept represents a breakthrough in our understanding of retinal circuitry and the process of photore-ception and phototransduction that has previously held steadfast for more than 100 years. This select subset of retinal ganglion cells has a dual source of input: trans-synaptic input conveyed from photoreceptor- mediated phototransduction and intrinsic activation via melanopsin-mediated phototransduction. Either input alone or a summed input from both systems can generate cell discharge. It should be no surprise that a parallel nonrod, noncone photoreceptive pathway exists in vertebrate eyes ( Fig. 1) ( 8,9). Ophthalmologists have long recognized that certain clinical observations in patients with profound visual loss from photoreceptor disease could not be adequately explained by the traditional model of photore-ception. Phenomena such as excessive light sensitivity, normal circadian rhythm, relative preservation of the pupil light reflex in outer retinal disorders, and paradoxical pupillary constrictions in darkness were difficult to reconcile with the traditional view that rods and cones were the only photosensitive cells of the eye. Circadian biologists struggled with another seeming contradiction. Mice lacking functional rods and cones maintained a normal day/ night cycle and their diurnal oscillation could be phase- shifted with light entrainment ( 10- 12). Yet, mice lacking eyes did not demonstrate a circadian rhythm, indicating that the eyes were necessary for circadian entrainment, but rods and cones were not ( 13). Thus began the aggressive search for the presence of another ocular photoreceptor with direct connections to the J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 195 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Kawasaki and Kardon OPL INL CB CB CB CB RB A All 10 s On MG NFL FIG. 1. Schematic diagram showing the synaptic circuitry of primate melanopsin- expressing retinal ganglion cells in the retina. Melanopsin- expressing retinal ganglion cells are primarily located in the ganglion cell layer with a minor percentage displaced to the inner nuclear layer ( INL). These ganglion cells have sparse dendrites and extremely large dendritic fields. The dendrites arborize in the inner plexiform layer ( IPL), forming a major plexus in theoutermost boundary of the IPLanda minor plexus in the innermost boundary of the IPL. In primates, green and red cones provide excitatory inputs through bipolar cells to melanopsin- expressing ganglion cell proximal dendrites, and rods provide excitatory inputs through, presumably, rod bipolar cells, type II amacrine cells, and cone bipolar cells successively. Blue cones provide inhibitory inputs, presumably through cone bipolar cells and inhibitory ( probably GABAergic) amacrine cells. Some amacrine cells of unknown identity also make synaptic contacts with melanopsin- expressing ganglion cell somata. Inset: The top panel shows the response of a primate melanopsin- expressing ganglion cell to a 470 nm light pulse. The cell continued to fire action potentials for 30 seconds after the end of the light stimulus. The white line shows membrane potential values averaged over 0.5 second sliding time windows. The bottom panel shows the first 5 seconds of the response shown on top panel. +, excitatory input; -, inhibitory input; resistor symbol, electrical coupling; A, amacrine cell; All, type II amacrine cell; BC, blue cone; C, cone; CB, cone bipolar cell; GC, green cone; GCL, ganglion cell layer; MG, melanopsin ganglion cell; NFL, nerve fiber layer; ONL, outer nuclear layer; OPL, outer plexiform layer; OS, outer segment; R, red; RB, rod bipolar cell; RC, red cone. ( Reprinted with permission from reference 8.) brain. The discovery of melanopsin and a subset of intrinsically photosensitive retinal ganglion cells containing this photopigment has provided an anatomic basis for explaining these clinical puzzles and has prompted a new look at the connections between the retina and the brain. In addition to their role in circadian entrainment, the melanopsin- expressing retinal ganglion cells mediate the pupillary light reflex. Axonal projections of the melanopsin- expressing cells to the midbrain represent the major retinal input to the olivary pretectal nucleus of the midbrain ( 14,15). Although there is evidence from a primate model that melanopsin- expressing ganglion cells may contribute to conscious visual perception, they do not appear to have the functional properties for direct image formation ( 16). It is safe to conclude at this time that the afferent signal for vision ( image formation) and the afferent signal for the pupil ( as well as other non- image- forming light functions) are processed by different types of retinal ganglion cells and travel within the optic nerve on separate sets of fibers. The growing body of literature on the anatomy and physiology of this novel, non- image- forming ocular photoreceptor pathway has been recently reviewed ( 8,9,17,18). Here we review some of the anatomic and physiologic aspects of the melanopsin- expressing ganglion cells and emphasize the relationship between the functional properties of the melanopsin- expressing ganglion cell in primates and the pupil light reflex in health and disease states of the retina and optic nerve. BIOLOGIC ROLE OF MELANOPSIN- EXPRESSING RETINAL GANGLION CELLS The mammalian eye has two different sensory systems that respond to different qualities of light. The classic photoreceptor system ( rods and cones) serves the familiar function of image formation which, by way of connections and processing by recipient neurons in the retina and visual cortex, provides for conscious visual perception. In contrast, the newly recognized melanopsin photoreceptive system functions in irradiance detection, much like a direct- current ( DC) light meter. Irradiance detection is a measure of environmental brightness and occurs at a subconscious level. Irradiance detection serves 196 © 2007 Lippincott Williams & Wilkins Photosensitive Retinal Ganglion Cells J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 to align biologic rhythms to the solar day, that is, setting the circadian clock. The master circadian pacemaker in mammals is the suprachiasmatic nucleus ( SCN) in the anterior hypothalamus. A direct monosynaptic neuron pathway linking the eye to the SCN, called the retinohypothalamic tract, has been known for decades ( Fig. 2) ( 19,20). Recent investigations have demonstrated that the melanopsin- expressing ganglion cells innervate the SCN and form the retinohypothalmic tract ( 3,4,21). These ganglion cells also project to other central sites that modulate the SCN and regulate the circadian clock, including the intergeniculate division of the lateral geniculate nucleus, the ventral subparaventricular zone of the hypothalamus, and the ventrolateral preoptic nucleus ( 22). The melanopsin- expressing retinal ganglion cells also project to the olivary pretectal nucleus ( OPN) of the dorsal midbrain, forming the afferent limb of the pupillary light reflex ( 4). This direct connection to the main pupil integrating center by intrinsically photosensitive cells explains why persons blind from photoreceptor disease can still have an intact pupil light reflex and appropriate day- night cycles and may provide a clue for the basis of their photosensitivity. Other non- image- forming functions mediated by the melanopsin- expressing retinal ganglion cells are melatonin secretion by the pineal gland, light-induced suppression of locomotor activity in rodents ( termed negative masking), and circadian- independent regulation of sleep and heart rate ( 22). Although the biologic functions of the classic and melanopsin photoreceptive pathways are widely different, there is also evidence of some interaction between them ( 7,22). Perhaps the most convincing finding comes from intracellular recordings showing that activation of short-wavelength cones attenuates the responses of the melanopsin-expressing retinal ganglion cells, whereas activation of medium- wavelength and long- wavelength cones and rods excites the melanopsin- expressing retinal ganglion cells ( 16). Additionally, retrograde tracer studies do not rule out the possibility that sites participating in image formation, such as the dorsal lateral geniculate nucleus, may receive FIG. 2. Schematic summary of brain regions and circuits influenced by melanopsin- expressing retinal ganglion cells. The melanopsin- expressing retinal ganglion cells and their axons are shown in dark blue and their principal targets in red. Projections of these ganglion cells to the suprachiasmatic nucleus ( SCN) form the bulk of the retinohypothalamic tract and contribute to photic entrainment of the circadian clock. The orange pathway with green nuclei shows a polysynaptic circuit that originates in the SCN and photically regulates melatonin release by the pineal gland ( P) through its sympathetic innervation. Synaptic links in this pathway include the paraventricular nucleus ( PVN) of the hypothalamus, the intermediolateral nucleus ( IML) of the spinal cord, and the superior cervical ganglion ( SCG). Another direct target of melanopsin- expressing retinal ganglion cells is the olivary pretectal nucleus ( OPN), a crucial link in the circuit underlying the pupillary light reflex, shown in light blue ( fibers) and purple ( nuclei). Synapses in this parasympathetic circuit are found at the Edinger- Westphal nucleus ( EW), the ciliary ganglion ( CG), and the iris muscles ( I). Other central targets include two components of the lateral geniculate nucleus of the thalamus, the ventral division ( LGNv), and the intergeniculate leaflet ( IGL). ( Reprinted with permission from reference 9.) 197 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Kawasaki and Kardon some melanopsin projections. Taken together, these findings suggest that irradiance detection may have a modulating influence on conscious visual perception ( 16,22). ANATOMICAL AND FUNCTIONAL FEATURES OF MELANOPSIN- EXPRESSING RETINAL GANGLION CELLS Melanopsin- expressing retinal ganglion cells represent a fractional subset of the total ganglion cell population of the eye: approximately 0.3% in primates ( 16). These ganglion cells are morphologically distinguished by their giant- sized soma and extremely large dendritic field size ( Fig. 3). The long, sparsely branching dendritic processes extend into the inner and outer sublayers of the inner plexiform layer where they interconnect and form a bilay-ered anatomic syncytium that spirals around the foveal pit. This bilayered dendritic meshwork, called a " photore-ceptive net," represents a distinctive anatomic model that appears suited to a " broad- capture" integration of light over long periods of time. Contrast this model to the " narrow-capture" photoreceptive model of rods and cones, which is specialized for encoding fine spatial resolution and a transient, adaptable response ( 23). Although the biochemical molecular cascade mediating melanopsin- based phototransduction is not yet fully elucidated, light activation of this pathway results in direct depolarization of the ganglion cell with generation of fast action potentials. The intrinsic photoactivation of melanopsin-expressing retinal ganglion cells has a relatively higher Eccentricity ( mm from fovea) i i i i i i i i i r 0 4 8 12 16 Eccentricity ( mm from fovea) FIG. 3. Morphology of melanopsin- immunoreactive cells. A. Human cell ( arrow); propidium iodide red counterstain. Scale bar, 50 mm. B. Macaque cell ( arrow). Scale bar, 50 mm. C. Macaque retina tracing; dots represent melanopsin- expressing cells. T, temporal retina; N, nasal retina; S, superior retina; I, inferior retina. D. Melanopsin- expressing ganglion cells in peripheral retina ( left; scale bar, 100 mm). Tracing of a peripheral horseradish peroxidase ( HRP)- stained giant cell ( right; scale bar, 200 mm). Parasol and midget cells ( far right) are shown for comparison. E. Melanopsin- expressing cells encircling the fovea ( left; scale bar, 200 mm). Tracings of two HRP- stained giant cells, 1- 1.5 mm from the fovea ( right; scale bar, 200 mm). Circles ( far right) indicate size of foveal parasol and midget cells. F. Dendritic field size of melanopsin cells versus eccentricity ( inner cells, filled circles, n = 93; outer cells, open circles, n = 63). Parasol ( filled diamonds, n = 333) and midget cells ( open diamonds, n = 93) are shown for comparison. C. Mean cell density of melanopsin cells versus eccentricity ( total 614 cells in 78 1 mm2 samples). H. Dendritic arbors ( green) of melanopsin cells ( arrows) from stacked confocal images of 5 consecutive vertical sections ( 25 mm thick). The soma of the outer cell is displaced to the inner nuclear layer ( INL). Scale bar, 50 mm. GCL, ganglion cell layer; IPL, inner plexiform layer. ( Reprinted with permission from reference 16.) 198 © 2007 Lippincott Williams & Wilkins Photosensitive Retinal Ganglion Cells J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 threshold to light compared with rod- mediated and cone-mediated light responses. In addition, the intrinsic activation has a longer latency time and a characteristic pattern of discharge of action potentials. The discharge rate builds relatively slowly and reaches a maximal firing rate that is maintained linearly proportional to light intensity ( Fig. 1, inset) ( 8,9,16). If a bright light is left on, the cell firing rate is remarkably steady and sustained without evidence of fatigue or adaptation to the continuous light stimulation. When the light is turned off, the cell does not immediately stop firing but gradually decreases its firing rate until it ceases. This may occur 10- 15 seconds or more after the light stimulus has been removed, depending on the brightness of the light. In classical phototransduction, the rods and cones hyperpolarize in response to light activation, have very short latency times to onset of ganglion cell firing, and demonstrate early adaptation. The features of the classical photoreceptors are compared with those of the ganglion cell photoreceptors in Table 1. Another characteristic feature of the intrinsic phototransduction pathway of the melanopsin- expressing ganglion cells is a broad spectral sensitivity with a peak in the short-wavelength range ( blue light toward 484 run in rats and 482 nm in macaque monkeys) ( 9,16). Why is there sensitivity to blue light? One investigator ( 24) has speculated that it has to do with setting the circadian clock, noting that 480 nm light is the dominant wavelength at dawn and at dusk. Despite the various anatomic and physiologic differences between these two types of photoreceptors, they are functionally complementary and synergistic systems for driving the pupil light reflex. Hattar et al ( 25) conclusively demonstrated this interaction in mice using three genetically modified strains. Melanopsin- knockout mice ( melanopsin gene ablated, but this class of retinal ganglion cell is still present and functioning) showed a recordable but incomplete pupil response to bright light, proving that rods and cones alone can drive the pupil light reflex but only to a certain extent. On the other hand, rodless and coneless mice were capable of generating a maximal pupil constriction to bright light in the action spectrum of melanopsin but with greatly reduced sensitivity. Finally, melanopsin- knockout mice also lacking rod and cone function failed to show any pupil light reflex regardless of the brightness or wavelength of the stimulus ( 25). Recently, S. Hattar et al ( personal communication, March 2007) showed that genetic ablation of melanopsin- containing retinal ganglion cells severely attenuates light- dependent physiologic functions, including the pupil light reflex, giving further evidence that this class of intrinsically photosensitive retinal ganglion cells mediates most of the afferent portion of the pupil light reflex. Although we assume that a dual photoreceptor system also exists in higher animals and humans, confirmation still awaits a systematic study of the pupil light reflex in patients with known gene mutations affecting phototransduction. ELECTROPHYSIOLOGY OF MELANOPSIN- EXPRESSING GANGLION CELLS IN PRIMATES Electrophysiologic studies have provided some details on the functional properties of melanopsin- expressing retinal ganglion cells in primates. Dacey et al ( 16) recorded the firing activity of individual melanopsin- expressing ganglion cells isolated from intact in vitro retina of macaque monkeys. After pharmacologic blockade of rod and cone function, they found that the ganglion cell activity TABLE 1. Features the mammalian eye Photoreceptor cell Location Photopigment Total number Receptive field Properties \ sensitivity Function of the classic ( visual) photoreceptive pathway and the Rods and cones Outer nuclear layer Rhodopsincone opsin 92,000,000 rods 5,000,000 cones Very small Adaptationfine spatial resolution All visible wavelengths Image formationpupillary light reflex melanopsin ( nonvisual) photoreceptive pathway in Melanopsin- expressing ganglion cells Ganglion cell and inner nuclear layers Melanopsin Several 1000s Very large ( photoreceptive net) Temporal integration of ambient light Broad band, most sensitive to blue wavelength light Circadian clockpupillary light reflex 199 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Kawasaki and Kardon mediated solely by its intrinsic activation ( the melanopsin pathway) was characterized by a long latency time before the first spike, slow build- up to the maximum firing rate, and maintenance of a tonic firing rate proportional to light intensity for the duration of the light stimulus. After light termination, the firing rate did not abruptly cease but rather exhibited a slow, gradual diminution to eventual cessation, features previously noted in rodent ganglion cells ( 7,9). Dacey et al ( 16) also studied the spectral sensitivity of the melanopsin- expressing retinal ganglion cells and their firing behavior as a function of light intensity ( Fig. 4). To a low intensity 610 nm ( red) light of 10 seconds' duration, they noted a rapid- onset, maximal burst of cell firing that attenuated and ceased ( adapted) during the light stimulus, that is, a transient cell response. At higher intensity ( by approximately 3 log units), cell firing was present for the duration of the light stimulus, but it demonstrated considerable attenuation. When the light was turned off, there was a correspondingly rapid OFF response of cell activity. This pattern of cell firing to a red light stimulus is consistent with a cone- driven, transient, adapting response. A low intensity, blue light stimulus ( 470 nm) evoked a rapid- onset cell firing rate that attenuated over the duration of the light stimulus. However, when the same blue stimulus was given at a 3 log unit brighter intensity, the cell maintained a steady rate of firing that persisted for more than 10 seconds after the light was turned off. This pattern of cell firing to bright blue light is consistent with a summed input from cone activation and intrinsic activation. The authors proposed that the melanop-sin- mediated response may serve to compensate for the attenuation, or transience, of cone input to deliver steady-state light information to the brain ( 16). Is there a clinical corollary to the electrophysiologic behavior of the melanopsin- expressing ganglion cell? Because these cells convey the primary light input to the olivary pretectal nucleus of the midbrain, the pupil light reflex might be expected to reflect the observed activity of the melanopsin- expressing retinal ganglion cells. This is supported by a recently published study by Gamlin et al ( 15), who confirmed loss of the photoreceptor- mediated transient pupil light reflex to red light but not to bright blue light after pharmacologic photoreceptor blockade in the monkey using intravitreal injections. THE PUPILLARY LIGHT REFLEX REVISITED Most clinical studies of the pupil light reflex have emphasized its transient response properties including the low intensity high intensity 61 Onm red light L c 10 s On 10 s On 470nm blue light f o 10 10 s On 10 s On FIG. 4. Spike histogram of a primate melanopsin- expressing retinal ganglion cell in response to a long- wavelength and a short- wavelength light stimulus. Top: Pure cone- mediated responses to a 610 nm light pulse at both low ( left; 12 log quanta cm22 s21) and high ( right; 15.2 log quanta cm22 s21) photopic levels. The cone- mediated response is transient at low light intensity ( left) and even at the higher photopic level used declines ( adapts) during 10 seconds of continuous light ( right). Bottom: Summed cone and intrinsic response to a 470 nm pulse at low ( left; 11 log quanta cm22 s21) and high ( right; 14.6 log quanta cm22 s21) photopic levels. When a short- wavelength light is used, the depolarizing intrinsic response is added to the cone response, making the response at low intensity more sustained ( left). At the higher light intensity, the intrinsic response maintains a steady discharge that persists even after termination of the light stimulus ( right). ( Modified and reprinted with permission from reference 16.) 200 © 2007 Lippincott Williams & Wilkins Photosensitive Retinal Ganglion Cells JNeuro- Ophthalmol, Vol. 27, No. 3, 2007 briskness, latency, and amplitude of pupil constriction to a brief, bright white light stimulus ( 26- 30). Other studies have also assessed the pupil response to longer duration light stimuli including both steady- state pupil size and " pupil escape," in which the pupil dilates or " gives way" during adaptation to continuous light ( 31,32). Figure 5 depicts a typical pupil response to a 10 second bright light in a normal human subject. Note that there are two components forming the response waveform during the constriction phase. When the light stimulus is turned ON, there is a rapid- onset, high- velocity pupil constriction until it reaches a minimum pupil size ( maximal constriction amplitude). This early transient response is quickly followed by pupillary redilation, or escape, to a more sustained state of partial pupil constriction that continues for the remainder of the light stimulus. Later, we will provide some preliminary evidence that the transient and sustained components of the pupil light reflex in humans can be explained by the proportional light input from the rod and cone photoreceptors and intrinsic retinal ganglion cell photoactivation. Until recently, the transient and sustained behavior of the pupil to light onset did not have a solid electrophysiologic foundation on which to understand its response to light varying in intensity, duration, and color. The electrophysiologic behavior of the melanopsin- expressing retinal ganglion cells provides a much- needed framework for understanding the pupil light reflex in healthy and disease states to varying conditions of light stimulus. This framework provides a basis for optimal use of the pupil light reflex in clinical practice and confers new significance to the use of the transient and sustained pupil response for diagnosis and differentiation of diseases of the optic nerve and photoreceptors. Pupil responses mediated by intrinsic activation of the melanopsin- expressing retinal ganglion cells in primates have recently been isolated using pharmacologic blockade of the rod and cone input ( 15). The recorded pupil responses matched the electrophysiologic behavior of the melanopsin- expressing ganglion cells in terms of chromatic sensitivity, intensity- dependent response, and response pattern to light onset and termination. In a normal monkey eye stimulated with equiluminant red and blue light, rapid-onset pupil constriction was elicited from both stimuli, but the maximum constriction amplitude was greater and more sustained with blue light under constant illumination. This action is presumably due to intrinsic ganglion cell activation superimposed on cone activation. Furthermore, transient n FIG. 5. Example of a pupillographic recording to a 5 second bright white light in a normal human subject. There are two components forming the response waveform during the constriction phase. When the light is turned ON, the transient phase is characterized by a short- latency, high- velocity maximal change in pupil size. Thereafter, the pupil partly redilates, or escapes, to a state of partial pupil constriction that represents the sustained phase of the pupil light reflex. 201 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Kawasaki and Kardon 1 cd/ m2 0.5 | « I" I 0.2 0.1 - red light stimulus - blue light stimulus 1 \ t light on rs*. * y / f* fv 1 light of . -, 0.5 5 03 ! = D °, Z ! « 0 z 3 1 10 cd/ m2 ed light stimulus :! ue light stimulus W I light of 0.5 I" II t.. 100cd/ m2 ^- red light stimulus - blue light stimulus H1l 1 _ r tf ^ light w- ***• 3n / ^ 7 J i iw • NT « 5 20 40 60 80 100 120 time ( seconds) 20 40 60 80 100 time ( seconds) 120 20 40 60 80 100 time ( seconds) 120 FIG. 6. Example of pupillographic recordings to equiluminant chromatic light stimuli in a normal human subject. The pupil responses shown in red were elicited using a long- wavelength ( 600- 620 nm bandwidth) red light of long duration ( 60 seconds) at 3 different light intensities ( 1 , 10, and 100 cd/ m2). The pupil responses shown in blue were elicited using a short- wavelength ( 4 6 5 ^ 8 5 nm bandwidth) blue light of similar duration and intensities. The y- axis shows the pupil size in relative units of pupil diameter ( not in mm units). The blue light stimulation produces a larger pupil constriction amplitude compared with red light stimulation at all intensities, presumably due to intrinsic activation of the melanopsin photoreceptive pathway that is additive to the cone activation when a blue light is used. Also note that the sustained pupil response to blue light shows no adaptation, particularly at higher intensities ( 10 and 100 cd/ m2), during 60 seconds of continuous light stimulation compared with the sustained pupil response to red light, which does show adaptation ( pupil escape after an initial transient constriction). the constricted state of the pupil persisted after removal of the light stimulus only when the short wavelength ( blue) light stimulus was used. After the monkey eye was treated with an intravitreal injection to pharmacologically inhibit all rod and cone phototransduction, no pupil response could be elicited by red light over a large range of light intensities. However, a bright blue light elicited a delayed, slow, and persistent pupil constriction having a distinct waveform in these primates, which are devoid of functioning rods and cones. In normal humans, Gamlin et al ( 15) also demonstrated a differential pupil response to red and blue light that paralleled the pupil responses obtained from normal monkeys. The data from pupil recordings in primates and humans support the hypothesis that the early transient pupil constriction under photopic conditions represents a 0 - 20 - ~ 40 c . e R„ ntract CO o g 100 f O. ^ r Unaffected Eye 1 - Afl - red light stimulus ^- blue light stimulus r ••"••"" IT cd/ m2 he 3 C07m2 0 cdfm2 10 20 30 40 50 Eye Affected With Photoreceptor Degeneration - 20 r 4o a5 60 CD is 80 c 8 100 I - red light stimulus ^- blue light stimulus r " IT cd/ m2 ho D cd/ m2 0 Ctiim2 60 10 20 30 40 50 60 FIG Time ( seconds) Time ( seconds) 7 Example of pupillographic recordings to equiluminant chromatic light stimuli in a patient with severe unilateral photoreceptor degeneration attributable to X- linked retinitis pigmentosa. The pupil responses shown in red were elicited using a 10 second long- wavelength ( 600- 620 nm bandwidth) red light at 3 different light steps ( 1, 10, and 100 cd/ m2). The pupil responses shown in blue were elicited using a 10 second short- wavelength ( 465^ 85 nm bandwidth) blue light at similar intensity steps. In the normal eye ( left), blue light stimulation produces a larger pupil constriction compared with red light stimulation, and the difference is greatest at the brightest light intensity ( 100 cd/ m2). These pupil responses to red light and blue light in the normal eye of this patient are similar to those of normal subjects tested with the same protocol. In the eye with no light perception due to severe photoreceptor degeneration from an unusual unilateral X- linked retinitis pigmentosa ( right), there is no recordable pupil response to red light even at the brightest intensity. However, a bright blue light stimulus ( 100 cd/ m2) still evokes a large amplitude, sustained pupil constriction. 202 © 2007 Lippincott Williams & Wilkins Photosensitive Retinal Ganglion Cells J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 predominantly cone- driven response and that the sustained pupil constriction represents a summation of the adapted cone response and the steady- state intrinsic retinal ganglion cell activation. From these results, we propose that changing certain characteristics of the light stimulus such as wavelength, intensity, and duration may be a novel way of using the pupil light reflex to independently assess rod and cone function and the intrinsic activation of melanop-sin- expressing ganglion cells. Specifically, we predict that analysis of the transient and sustained pupil response to red and blue light of differing intensities will help to better differentiate normal eyes from diseased eyes and better differentiate pathology of the rods and cones from that of the retinal ganglion cells. TRANSIENT AND SUSTAINED PUPIL RESPONSE TO CHROMATIC STIMULI: PRELIMINARY RESULTS IN HUMANS We have been exploring the pupil responses to chromatic stimuli in human subjects with normal healthy eyes and in patients with various types of neuroretinal visual loss. By using a Ganzfeld bowl as the means to provide equiluminant red and blue light, the pupil responses are tracked and recorded from a pair of lightweight eyeglasses that are integrated with miniature infrared- sensitive video cameras and worn comfortably by the patient. Figure 6 is an example of pupil recordings from a healthy adult with normal eyes who was presented a long duration ( 60 seconds) light stimulus, first using a long - wavelength ( red bandwidth 600- 620 nm) hght at low, medium, and high intensities ( 1, 10, and 100 cd/ m2) and then a short- wavelength hght ( blue bandwidth 465- 485 nm) at similar intensities. The blue light consistently produced a larger pupil constriction amplitude compared with the red light, presumably due to the added input from intrinsic activation that occurs when a blue light is used. The difference in constriction amplitude to red light and blue light increased with increasing light intensity. At intermediate and brighter light intensities ( 10 cd/ m2 and higher), the blue light stimulus produced a sustained pupil constriction that was not seen with red light of the same photopically matched intensity. This finding is consistent with a greater intrinsic activation of melanopsin- expressing retinal ganglion cells at brighter blue light intensities, with sustained cell firing and pupil constriction dominating the response, as demonstrated in rodents and primates ( 15,16,33). Given these findings in normal human eyes, we predicted that there would be a loss of pupil constriction amplitude to red light that is proportionate to the amount of cone loss in patients with isolated photoreceptor disease and a preservation of the sustained pupil constriction to bright blue light. These changes would effectively result in a greater difference of the pupil responses elicited from red light compared with blue light. Figure 7 shows the pupil recordings from such a patient with severe unilateral photoreceptor degeneration due to X- linked retinitis pigmentosa. The unilaterality is thought to be attributable to early X chromosome inactivation or a form of mosaicism. The electroretinogram ( ERG) was unrecordable in the affected eye that had no light perception, and the ERG was normal in the contralateral eye with 20/ 20 visual acuity. The appearance of the optic nerves was normal, with no evidence of damage other than in the outer layers of the retina. Clinically, there was a > 3.0 log unit relative afferent pupil defect in the affected eye. In the normal eye, there was a differential pupil response to red light and blue light similar to that shown for the normal subject in Figure 6. In the photoreceptor- degenerate eye, there was no reliable recordable pupil response to red light, even at the brightest intensity, but a blue light evoked a strong sustained pupil constriction, even though the patient perceived no light in the damaged eye. These preliminary recordings in human subjects demonstrate that changes in the pupil responses to chromatic stimuli are readily detectable and easily quantifiable with standard instruments of clinical testing. The changes in the pupil response follow patterns that predict the underlying the ocular pathologic condition. Such results are encouraging for the development of new methods of pupil testing that may allow earlier distinction between rod and cone photoreceptor disease and retinal ganglion cell disease. We hypothesize that changes in the transient pupil response to red light and low intensity blue light may be more sensitive to cone and rod disease, whereas changes in the sustained pupil response to bright blue light may be more sensitive to optic nerve disease. Ongoing studies in our pupil laboratory are aimed to optimize stimulus conditions that elicit pupil responses that can better localize the site of damage to rods, cones, or retinal ganglion cells, quantify the extent of disease, and provide an objective reflex for monitoring the course of disease and its response to treatment. REFERENCES 1. Provencio I, Jiang G, De Grip WJ, et al. Melanopsin: an opsin in melanophores, brain and eye. Proc Natl Acad Sci USA 1998; 95: 340- 5. 2. Provencio I, Rodriguez IR, Jiang G, et al. A novel human opsin in the inner retina. JNeurosci 2000; 20: 600- 5. 3. Hannibal J, Hindersson P, Knudson SM, et al. The photopigment melanopsin is exclusively present in pituitary adenylate cyclase-activating polypeptide- containing retinal ganglion cells of the retinohypothalamic tract. J Neurosci 2002; 22: RC191. 4. Hattar S, Liao HW, Takao M, et al. Melanopsin- containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 2002; 295: 1065- 70. 5. Melyan Z, Tarttelin EE, Bellingham J, et al. Addition of human melanopsin renders mammalian cells photoresponsive. Nature 2005; 433: 741- 5. 203 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Kawasaki and Kardon 6. Qiu X, Kumbalasiri T, Carlson SM, et al. Induction of photosensitivity by heterologous expression of melanopsin. Nature 2005; 433: 745- 9. 7. r DM, Dunn FA, Takao M. Phototransduction by retinal ganglion cells that set the circadian clock. Science 2002; 295: 1070- 3. 8. Fu Y, Liao HW, Do MT, et al. Non- image- forming ocular photoreception in vertebrates. Curr Opin Neurobiol 2005; 15: 415- 22. 9. Berson DM. Strange vision: ganglion cells as circadian photoreceptors. Trends Neurosci 2003; 26: 314- 20. 10. Foster RG, Provencio I, Hudson D, et al. Circadian photoreception in the retinally degenerate mouse ( rd/ rd). J Comp Physiol 1991; 169: 39- 50. 11. Freedman MS, Lucas RJ, Soni B, et al. Regulation of mammalian circadian behavior by non- rod, non- cone ocular photoreceptors. Science 1999; 284: 502^. 12. Lucas RJ, Freedman MS, Lupi D, et al. Identifying the photore-ceptive inputs to the mammalian circadian system using transgenic and retinally degenerate mice. Behav Brain Res 2001; 125: 97- 102. 13. Yamakazi S, Goto M, Menaker M. No evidence for extraocular photoreceptors in the circadian system of the Syrian hamster. J Biol Rhythms 1999; 14: 197- 201. 14. Hattar S, Liao HW, Takao M, et al. Melanopsin- containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 2002; 295: 1065- 70. 15. Gamlin PD, McDougal DH, Pokorny J, et al. Human and macaque pupil responses driven by melanopsin- containing retinal ganglion cells. Vision Res 2007; 47: 946- 54. 16. Dacey DM, Liao HW, Peterson BB, et al. Melanopsin- expressing ganglion cells in primate retina signal colour and irradiance and project to the LGN. Nature 2005; 433: 749- 54. 17. Van Gelder RN Non- visual ocular photoreception. Ophthalmic Genet 2001; 22: 195- 20. 18. Peirson S, Foster RG. Melanopsin: another way of signaling light. Neuron 2006; 49: 331- 9. 19. Moore RY, Lenn NJ. A retinohypothalamic projection in the rat. J Comp Neurosci 1972; 146: 1- 14. 20. Pickard GE. Morphological characteristics of retinal ganglion cells projecting to the suprachiasmatic nucleus: a horseradish peroxidase study. Brain Res 1980; 183: 458- 65. 21. Gooley JJ, Lu J, Chou TC, et al. Melanopsin in cells of origin of the retinohypothalamic tract. Nat. Neurosci 2001 ; 4: 1165. 22. Gooley JJ, Lu J, Fischer D, et al. A broad role for melanopsin in nonvisual photoreception. J Neurosci 2003; 23: 7093- 106. 23. Provencio I, Rollag MD, Castrucci AM. Photoreceptive net in the mammalian retina: this mesh of cells may explain how some blind mice can still tell day from night. Nature 2002; 415: 493. 24. Foster RG. Neurobiology: bright blue times. Nature 2005; 433: 698- 9. 25. Hattar S, Lucas RJ, Mrosovsky N, et al. Melanopsin and rod- cone photoreceptive systems account for all major accessory visual functions in mice. Nature 2003; 424: 76- 81. 26. Kardon R, Kawasaki A, Miller NR. Origin of the relative afferent pupillary defect in optic tract lesions. Ophthlamology 2006; 113: 1345- 53. 27. Volpe NJ, Plotkin ES, Maguire MG, et al. Portable pupillography of the swinging flashlight test to detect afferent pupillary defects. Ophthalmology 2000; 107: 1913- 21. 28. Wilhelm B, Ludtke H, Peters T, et al. Automated swinging flashlight test in patients with optic nerve diseases ( in German). Klin Monatsbl Augenheilkd 2001; 218: 21- 5. 29. Bremner FD, Shallo- Hoffman J, Riordan- Eva P, et al. Comparing pupil function with visual function in patients with Leber hereditary optic neuropathy. Invest Ophthalmol Vis Sci 1999; 40: 2528- 34. 30. Bergamin O, Zimmerman MB, Kardon RH. Pupil light reflex in normal and diseased eyes: diagnosis of visual dysfunction using waveform partitioning. Ophthalmology 2003; 110: 106- 14. 31. Enyedi LB, Dev S, Cox TA. A comparison of the Marcus Gunn and alternating light tests for afferent pupillary defects. Ophthalmology 1998; 105: 871- 3. 32. Bergamin O, Kardon RH. Greater pupillary escape differentiates central from peripheral visual field loss. Ophthalmology 2002; 109: 771- 80. 33. Lucas RJ, Hattar S, Takao M, et al. Diminished pupillary light reflex at high irradiances in melanopsin knockout mice. Science 2003 ; 299: 245- 7. 204 © 2007 Lippincott Williams & Wilkins |
