- Email: [email protected]
The ERG responses to light stimuli of melanopsin-expressing retinal ganglion cells that are independent of rods and cones
Neuroscience Letters 479 (2010) 282–286
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
The ERG responses to light stimuli of melanopsin-expressing retinal ganglion cells that are independent of rods and cones Yumi Fukuda a,∗ , Sei-ichi Tsujimura b , Shigekazu Higuchi c , Akira Yasukouchi c , Takeshi Morita a a b c
Department of Living Environmental Science, Fukuoka Women’s University, Fukuoka, Japan Department of Bioengineering, Kagoshima University, Kagoshima, Japan Department of Human Science, Kyusyu University, Fukuoka, Japan
a r t i c l e
i n f o
Article history: Received 8 February 2010 Received in revised form 12 April 2010 Accepted 26 May 2010 Keywords: Melanopsin-expressing retinal ganglion cells Circadian rhythms Photoreception Electroretinogram
a b s t r a c t The mechanisms by which melanopsin-expressing retinal ganglion cells (mRGCs) regulate circadian rhythms in humans have not been established. To understand mRGC characteristics and their role independent of effects due to the rods and cones, mRGC responses should be induced or measured independent of cone and rod responses. In the present study, we obtained results from light stimuli which differentially induce only the mRGC response by using a receptor-silent substitution technique. The mRGCs responded linearly to contrast changes of light stimuli, whereas they showed complicated responses to frequency changes with regard to the latency of response time. These results suggest that mRGC behavior is not a simple response to the various frequencies found in solar light but may be related to intrinsic neural circuits with feedback connections in the mRGC pathway. The results in this study also demonstrated that the test stimuli affected only the mRGC response and that this could be successfully detected by using the electroretinogram (ERG). © 2010 Elsevier Ireland Ltd. All rights reserved.
Over the last 20 years, researchers have tried to understand the photoreceptor mechanisms which regulate the circadian system. In the last decade, 450–500 nm has been identified by actionspectrum analysis as the most potent region for neuroendocrine, neurobehaviroral and circadian regulation in rodents, primates and humans. Since these regions in the action spectrum did not conform to the most sensitive regions for visual perception, it was suggested that a novel photosensory system, distinct from the visual rods and cones, is primarily responsible for circadian responses [2,20]. Melanopsin-expressing retinal ganglion cells (mRGCs) were identified in the rodent [1,7,15,17] and the primate and human retinae [4,16]. These cells differ from rods and cones in many respects. For example, they respond to light much more sluggishly and populate sparsely: There are only 0.2% of mRGCs in approximately 1.5 million ganglion cells in the human retina. Furthermore, light depolarizes these cells tonically and elevates spike frequency while the opposite happens with rods and cones [1,4]. The discovery of this novel photoreceptor aroused researchers’ interest in the role of the eye in non-image-forming visual functions. On the other hand, researchers in the fields of medicine, architecture and illumination engineering approached the mechanism of circadian entrainment differently, in order to clarify
∗ Corresponding author at: 1-1-1 Kasumigaoka, Higashi-ku, Fukuoka 813-8529, Japan. Tel.: +81 92 661 2411. E-mail address: [email protected] (Y. Fukuda). 0304-3940/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2010.05.080
standards for environmental light and to promote light therapy for the regulation of human circadian rhythms. In the early studies, some researchers examined the effects of various illuminances [3,10] or wavelengths [14] on circadian rhythms. Following the finding of the mRGC, some researchers investigated circadian phase shifts by using monochromatic and polychromatic lights with short wavelengths in order to improve efficacy at inducing circadian phase shifts. Lockley et al. [11] exposed subjects to 460 nm and 555 nm of monochromatic lights and found that the human circadian melatonin response was more sensitive to the shorter wavelength. Warman et al. [22] demonstrated that a low-intensity short-wavelength light pulse (8 lx, 28 W/cm2 , 6.21 × 1013 photons/cm2 /s) was able to phase advance the human circadian system by a similar magnitude to a bright white light pulse (12,000 lx, 4300 W/cm2 , 1.15 × 1016 photons/cm2 /s). However, it is important to note that the light stimuli used in these studies also affected the other photoreceptors, the rods and cones, as well as the mRGCs. In these cases, therefore, it is not clear whether the circadian responses were caused only through the mRGCs or by the rods and cones also. As it is important to understand mRGC characteristics and their role independent of effects due to the rods and cones, mRGC responses should be caused or measured independent of cone and rod responses. It seems that Kremers et al. [9] also noticed the same necessity and performed an experiment with cone-isolating stimuli, which were based on a silent substitution paradigm [6], and the heterochromatic flicker photometry electroretinogram (ERG) in order to quantify each
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
283
Fig. 1. A diagram illustrating the experimental set-up of the electronic and optical equipment, the integrating sphere exposure and the monitor array.
contribution of the different cone types. However, they have not investigated the mRGC contribution in ERG. We report here the first results from light stimuli which differentially induce only the mRGC response in ERG. They are highly relevant to providing further knowledge for studies in biological rhythms. The illumination system has been purpose-built with a receptorsilent substitution technique [21]. It consisted of an optical diffuser and an integrating sphere which presented an 18.9◦ circular field on the optical diffuser (Fig. 1). Four different kinds of light-emitting diodes (LEDs) (Opto Supply Limited, Hong Kong, China) were used as internal light sources in the integrating sphere. The peak wavelengths of the four LEDs were 633 nm, 593 nm, 507 nm and 468 nm with half-height bandwidths 15 nm, 14 nm, 33 nm and 22 nm, respectively. The light emitted from four types of LEDs, which were embedded in the inner wall of the integrating sphere, projected as the internally synthesized test stimuli. Radiance output of each LED was controlled by analogue pulse-width modulation (PWM) units by adjusting the duty cycle of the pulse train to 1 kHz. The PMW units were controlled by a micro computer (H8/3052 Renesas Technology, Japan). The characteristic of dutyluminance was calibrated to minimize the deviation caused by thermal effects. The excitation expresses stimulus levels to each photoreceptor. The 10◦ cone fundamentals, proposed by Stockman et al. [18,19], and spectral radiance of the light stimuli were used to calculate the excitation of cones sensitive to long (L), middle (M) or short (S) wavelengths. We estimated the spectral sensitivity of melanopsin from a pigment template nomogram [5] with a peak wavelength (max ) of 482 nm, as Dacey et al. [4] had shown that a spectral tuning curve of melanopsin was closely approximated by a pigment template with a peak at 482 nm. We used the fundamental, the spectral sensitivity of mRGC, in a 10◦ field displaying a peak wavelength at 502 nm after careful calibration of prereceptoral filters [21]. An excitation of the mRGC was then calculated from the fundamental (with a unity peak) as relative luminous efficiency and the spectral radiance of the light stimuli. The subject’s left pupil was maximally dilated by a mydriatic agent for electroretinogram (ERG) recordings and topical anesthesia for the relief of pain was provided. Following this, an electrode (EA-102, Meiyo, Japan) was placed in contact with the subject’s left cornea for full-field recording. The corneal surface was protected with a non-irritant and non-allergenic ionic conducting solution. A reference electrode was placed on the subject’s forehead and another electrode was attached to the ear as a ground. Either a vision therapist or an ophthalmologist was present throughout the
experiment in order to provide technical treatment and ensure good ocular contact, proper electrode impedance and ophthalmologic safety. Ten healthy subjects (5 females, 5 males; mean ± SD age, 24.3 ± 3.4 years; range 21–31 years; Japanese) were studied at Fukuoka Women’s University in Japan. The study was approved by the Ethics Committee at Fukuoka Women’s University and subjects gave written informed consent prior to study. All subjects had ocular health and normal color vision according to the Ishihara color blindness test. Fig. 1 illustrates the apparatus used in the experiment. The light from the integrating sphere was modulated to a sinusoidal wave and assessed with a spectroradiometer (CS-1000A, Konica Minolta, Japan). Fig. 2(a) and (b) shows examples of a light stimulus in 50% contrast. Peaks of the sinusoidal stimuli are described as high (H), minima as low (L), and constant stimuli without a sinusoidal wave as base (B). The light exposure consisted of 30
Fig. 2. Examples of a light stimulus in 50% contrast. (a) A light stimulus with sinusoidal wave modulation as an example of 50% contrast. The light varies from base (B) to high (H) and low (L) for 3 s and then maintains the base level for 2 s. (b) Spectral radiance distributions of base (B), high (H) and low (L) light stimuli in 50% contrast.
284
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
Table 1 An example of stimulus excitation to L, M, S cones and mRGC in 50% contrast.
Base (B) High (H) Low (L)
L cone
M cone
S cone
mRGC
100.0 99.9 99.2
100.0 100.1 99.3
100.0 98.1 100.6
100.0 152.3 47.1
The relative excitation of base (B), high (H) and low (L) is shown in percentages.
repetitions of 3-s exposure to the sinusoidal wave and 2-s constant (base) light irradiation (Fig. 2(a)). For each 3-s exposure, we were able to choose the light frequency and contrast with the PWM unit. We conducted the following two experiments: (1) contrast changes of 10, 20, 30, 40 and 50% with 5.0 Hz frequency and (2) frequency changes of 0.5, 1.0, 2.0, 5.0, 8.0, 12.0 and 30.0 Hz with 50% contrast. These light stimuli were designed with the help of Tsujimura et al. [21]. The ERG signals from photoreceptors were continuously digitized during the experiment at a sampling rate of 5 kHz by a data input system. The spectral radiance distributions merged light from four kinds of LEDs as shown in Fig. 2(b). The stimulus excitation of L, M and S cones and mRGCs was calculated by multiplying these spectral radiance distributions and the fundamentals of L, M and S cones and mRGCs. We stress that the stimulus levels to the cones did not vary in the study. As shown in Table 1, in the case of 50% contrast, each level of the base excitation to L, M and S cones and mRGCs was maintained as 100%, and only the mRGC excitation was varied – 50% higher than the base excitation in the high excitation and 50% lower in the low excitation; that is, the integrating sphere emitted light which changed only the mRGC excitation but not that of the cones. Furthermore, because the radiance saturated rod sensitivity to the light stimuli during the 5-min adaptation period before the emission, we presumed that rod responses did not appear on the ERG. The experiment was performed from 9 a.m. to 4 p.m. in order to exclude effects of circadian rhythms on ocular responses. A mydriatic agent was dropped into the subject’s left eye for pupil’s dilation in a waiting room; pupils became completely dilated during the next 30 min while the subjects sat quietly. The subjects then entered an artificial climate chamber. They were asked to sit quietly, rest their head in a chin rest, and gaze at the fixation at the center of stimulus circle (100 mm in diameter) on the diffuser in front of them. The diffuser was at a distance of 300 mm from a subject whose visual angle to the circle was 18.9◦ (Fig. 1). After 5-min adaptation in the chamber, an ophthalmologic technician applied ERG electrodes to subjects. In the first experimental session, effects of contrast changes were observed (the order of contrast changes was randomly arranged). The experiment lasted for approximately 30 min, during which time subjects were instructed to keep their eyes open and try to avoid blinking, especially when the integrating sphere emitted experimental light for the 3-s period. The mRGC response to the photostimuli was monitored by a personal computer during the experiment. In the second experimental session, effects of frequency changes were observed (the order of frequency changes was randomly arranged). The session started 30 min after the end of the first, to minimize stress induced by an electrode in contact with the cornea. After the two experimental sessions, subjects received eye-drops containing an antibacterial agent and were released from the laboratory. Power and phase spectra of mRGC responses were calculated by using Fast Fourier Transformation (FFT). The power indicates degree of the mRGC response and the phase expresses latency in the mRGC response to the sinusoidal light stimuli. We observed the phase response of the mRGC to the light stimuli in order to extract the mRGC response from other artifacts. We show the phase in degree as phase shifts of the mRGC response to the light stimuli with
Fig. 3. The mRGC response to the light stimuli with 10, 20, 30, 40 and 50% contrasts modulated at 5 Hz. (a) Scatter plot of the power in the ERG. The power is an index of the mRGC response to the light stimuli. The identical symbols express the individual power from the same subject and the emphasized and filled circles the medians with the linear approximation (solid line, R2 = 0.999). There was a significant correlation between the mRGC response and the contrasts (rs = 0.66, p < 0.001). The power varied linearly according to the contrasts. (b) The relationship between the contrasts and the phase. The phase in degrees (left axis) represents phase shifts converted into 360◦ of the mRGC response to the sinusoidal light stimuli. The phase in msec (right axis) shows latency of the mRGC response to each light stimulus. The dots express the averages, and the bars the standard deviations. There was no significant difference in these measures with changing contrasts of stimuli.
the sinusoidal wave and converted this measure into the latency to each stimulus in milliseconds. The homogeneity of variances of data was analyzed with SPSS, and appropriate parametric tests or nonparametric tests were chosen for data analysis. Fig. 3(a) shows the power in the ERG, as an index of the mRGC response following light stimuli, with five contrasts modulated at 5 Hz. The response appears to elevate linearly from 10% to 50% contrast. As the test of homogeneity of variances indicated that the data of the mRGC were not homoscedastic, nonparametric tests were applied for these results. Spearman’s rank correlation coefficient between the mRGC response and the contrasts showed high statistical significance (rs = 0.66, p < 0.001). Fig. 3(b) shows the relationship between the contrasts and the phase in degree and in msec. One-way ANOVA showed that the phase did not differ significantly between the contrasts. This means that we could collect data on mRGC responses according to the contrasts. Fig. 4(a) shows the power in the ERG, which expresses the mRGC response to the light stimuli with 50% contrast at seven frequencies. The response varied according to the frequency. One-way ANOVA indicated that the data of the mRGC response were homoscedastically distributed, and there were significant differences in mRGC responses with frequency (p < 0.01). Pearson’s correlation did not show any significant association between the frequencies and the mRGC response. For the phase of seven different frequencies,
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
285
change with increased frequency in a simple manner, as shown in Fig. 4(b). The phase would be predicted to change linearly with frequency if it was only due to the latency in the mRGC response; however, the results do not support this possibility as the variance in the low-frequency range differed from that in the high-frequency range. The phase may be affected not only by the latency but also by other, unexpected factors – feedback from the optic nerve tract, for example. Such an interaction could explain the observed complex latency response. Honma and Honma [8] and Minors et al. [12] reported that phase responses of human circadian rhythms varied depending on when exposure to light took place and the light intensity. The observed phase advances and delays to light can be ascribed to photosensitivity in the eye as well as individual responses and physiological regulation in the hypothalamus. Morita et al. [13] demonstrated that human eye function has a diurnal rhythm, using the 100-hue test in which subjects arranged color chips according to gradation of hue. Thus, it is also possible that mRGC responses vary with times of day. There were two different experimental sessions in this study, in the morning and in the afternoon. There was no significant difference in the results between the two times of testing, although the mRGC response was slightly higher in the afternoon than in the morning. Factors affecting mRGC responses by light of different frequencies are still unknown; clarification of diurnal physiological rhythms and the mechanisms involved in mRGC responses will be the subjects of our next study. In addition, further studies, such as in animal models with the illumination system and ERG, will enable us to verify and interpret ERG measurements and that should contribute directly to further knowledge of mRGC characteristics. Acknowledgement
Fig. 4. The mRGC response to the light stimuli with 50% contrast at seven frequencies. (a) The ERG power of the mRGC response to the light stimuli. The results are expressed in the same way as in Fig. 3(a) but the medians fitted the quadratic approximation (R2 = 0.628). (b) The phase in msec with seven frequencies. The dots express the averages, and the bars the standard deviations. The variance in low-frequency range between 0.5 and 2 Hz was significantly higher than that in high-frequency range between 5 and 30 Hz (p < 0.001).
the data were not homoscedastic. Although nonparametric tests showed that there were no significant differences between the frequencies, the phase does appear to vary. Fig. 4(b) shows the latencies of mRGC responses to the light stimuli. Brown–Forsythe test indicated that the variance in low-frequency range between 0.5 and 2 Hz was significantly higher than that in high-frequency range between 5 and 30 Hz (p < 0.001). Absolute deviations from medians were 283.7 in the low-frequency range and 19.6 in the high-frequency range. The results in this study demonstrate that the test stimuli affected only mRGC responses and that the response could be successfully detected by using the ERG. Although the stimuli would have affected cone and rod photoreceptors, the stimulus excitation to these receptors was the same throughout the experiment, and this means that changes in responses measured by the ERG were due to changes in mRGC excitation only. It was clear that the mRGC responded to the light contrasts. As shown in Fig. 3(a), the power of the mRGC response increased according to the contrasts of 10, 20, 30, 40 and 50%. However, the changes in the response with the stimulus frequency illustrated a non-linear curve, as indicated in Fig. 4(a). This result suggests that mRGC behavior is not a simple response to the various frequencies found in solar light. Similarly, the latency of the response time, derived from the phase with respect to the light stimulus, did not
We gratefully appreciate the editorial support and helpful advice and comments on the manuscript of Prof. Jim Waterhouse at Liverpool John Moores University. References [1] D.M. Berson, F.A. Dunn, M. Takao, Phototransduction by retinal ganglion cells that set the circadian clock, Science 295 (2002) 1070–1073. [2] G.C. Brainard, J.P. Hanifin, J.M. Greeson, B. Byrne, G. Glickman, E. Gerner, M.D. Rollag, Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor, J. Neurosci. 21 (2001) 6405–6412. [3] C.A. Czeisler, G.S. Richardson, J.C. Zimmerman, M.C. Moore-Ede, E.D. Weitzman, Entrainment of human circadian rhythms by light–dark cycles: a reassessment, Photochem. Photobiol. 34 (1981) 239–247. [4] D.M. Dacey, H.-W. Liao, B.B. Peterson, F.R. Robinson, V.C. Smith, J. Pokorny, K.-W. Yau, P.D. Gamlin, Melanopsin-expressing ganglion cells in primate retina signal colour and irradiance and project to the LGN, Nature 433 (2005) 749–754. [5] H.J. Dartnall, The interpretation of spectral sensitivity curves, Br. Med. Bull. 9 (1953) 24–30. [6] O. Estévez, H. Spekreijse, The ‘silent substitution’ method in visual research, Vision Res. 22 (1982) 681–691. [7] S. Hattar, H.-W. Liao, M. Takao, D.M. Berson, K.-W. Yau, Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity, Science 295 (2002) 1065–1070. [8] K. Honma, S. Honma, A human phase response curve for bright light pulses, Jpn. J. Psychiatry Neurol. 42 (1988) 167–168. [9] J. Kremers, T. Usui, H.P.N. Scholl, L.T. Sharpe, Cone signal contributions to electrograms in dichromats and trichromats, Invest. Ophthalmol. Vis. Sci. 40 (1999) 920–930. [10] A.J. Lewy, T.A. Wehr, F.K. Goodwin, D.A. Newsome, S.P. Markey, Light suppresses melatonin secretion in humans, Science 210 (1980) 1267–1269. [11] S.W. Lockley, G.C. Brainard, C.A. Czeisler, High sensitivity of the human circadian melatonin rhythm to resetting by short wavelength light, J. Clin. Endocrinol. Metab. 88 (2003) 4502–4505. [12] D.S. Minors, J.M. Waterhouse, A. Wirz-Justice, A human phase-response curve to light, Neurosci. Lett. 133 (1991) 36–40. [13] T. Morita, M. Ohyama, H. Tokura, Diurnal variation of hue discriminatory capability under artificial constant illumination, Experientia 50 (1994) 641–643. [14] T. Morita, H. Tokura, The influence of different wavelengths of light on human biological rhythms, J. Physiol. Anthropol. Appl. Human Sci. 17 (1998) 91–96.
286
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
[15] S. Panda, T.K. Sato, A.M. Castrucci, M.D. Rollag, W.J. DeGrip, J.B. Hogenesch, I. Provencio, S.A. Kay, Melanopsin (Opn4) requirement for normal light-induced circadian phase-shifting, Science 298 (2002) 2213–2216. [16] I. Provencio, I.R. Rodriguez, G. Jiang, W.P. Hayes, E.F. Moreira, M.D. Rollag, A novel human opsin in the inner retina, J. Neurosci. 20 (2000) 600–605. [17] I. Provencio, M.D. Rollag, A.M. Castrucci, Photoreceptive net in the mammalian retina, Nature 415 (2002) 493. [18] A. Stockman, L.T. Sharpe, The spectral sensitivities of the middle- and longwavelength-sensitive cones derived from measurements in observers of known genotype, Vision Res. 40 (2000) 1711–1737. [19] A. Stockman, L.T. Sharpe, C. Fach, The spectral sensitivity of the human shortwavelength sensitive cones derived from thresholds and color matches, Vision Res. 39 (1999) 2901–2927.
[20] K. Thapan, J. Arendt, D.J. Skene, An action spectrum for melatonin suppression: evidence for a novel non-rod, non-cone photoreceptor system in humans, J. Physiol. 535 (2001) 261–267. [21] S. Tsujimura, K. Ukai, D. Ohama, A. Nuruki, K. Yunokuchi, Contribution of human melanopsin retinal ganglion cells to steady-state pupil responses, Proc. R. Soc. B., doi:10.1098/rspb.2010.0330. [22] V.L. Warman, D.-J. Dijk, G.R. Warman, J. Arendt, D.J. Skene, Phase advancing human circadian rhythms with short wavelength light, Neurosci. Lett. 342 (2003) 37–40.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
The ERG responses to light stimuli of melanopsin-expressing retinal ganglion cells that are independent of rods and cones Yumi Fukuda a,∗ , Sei-ichi Tsujimura b , Shigekazu Higuchi c , Akira Yasukouchi c , Takeshi Morita a a b c
Department of Living Environmental Science, Fukuoka Women’s University, Fukuoka, Japan Department of Bioengineering, Kagoshima University, Kagoshima, Japan Department of Human Science, Kyusyu University, Fukuoka, Japan
a r t i c l e
i n f o
Article history: Received 8 February 2010 Received in revised form 12 April 2010 Accepted 26 May 2010 Keywords: Melanopsin-expressing retinal ganglion cells Circadian rhythms Photoreception Electroretinogram
a b s t r a c t The mechanisms by which melanopsin-expressing retinal ganglion cells (mRGCs) regulate circadian rhythms in humans have not been established. To understand mRGC characteristics and their role independent of effects due to the rods and cones, mRGC responses should be induced or measured independent of cone and rod responses. In the present study, we obtained results from light stimuli which differentially induce only the mRGC response by using a receptor-silent substitution technique. The mRGCs responded linearly to contrast changes of light stimuli, whereas they showed complicated responses to frequency changes with regard to the latency of response time. These results suggest that mRGC behavior is not a simple response to the various frequencies found in solar light but may be related to intrinsic neural circuits with feedback connections in the mRGC pathway. The results in this study also demonstrated that the test stimuli affected only the mRGC response and that this could be successfully detected by using the electroretinogram (ERG). © 2010 Elsevier Ireland Ltd. All rights reserved.
Over the last 20 years, researchers have tried to understand the photoreceptor mechanisms which regulate the circadian system. In the last decade, 450–500 nm has been identified by actionspectrum analysis as the most potent region for neuroendocrine, neurobehaviroral and circadian regulation in rodents, primates and humans. Since these regions in the action spectrum did not conform to the most sensitive regions for visual perception, it was suggested that a novel photosensory system, distinct from the visual rods and cones, is primarily responsible for circadian responses [2,20]. Melanopsin-expressing retinal ganglion cells (mRGCs) were identified in the rodent [1,7,15,17] and the primate and human retinae [4,16]. These cells differ from rods and cones in many respects. For example, they respond to light much more sluggishly and populate sparsely: There are only 0.2% of mRGCs in approximately 1.5 million ganglion cells in the human retina. Furthermore, light depolarizes these cells tonically and elevates spike frequency while the opposite happens with rods and cones [1,4]. The discovery of this novel photoreceptor aroused researchers’ interest in the role of the eye in non-image-forming visual functions. On the other hand, researchers in the fields of medicine, architecture and illumination engineering approached the mechanism of circadian entrainment differently, in order to clarify
∗ Corresponding author at: 1-1-1 Kasumigaoka, Higashi-ku, Fukuoka 813-8529, Japan. Tel.: +81 92 661 2411. E-mail address: [email protected] (Y. Fukuda). 0304-3940/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2010.05.080
standards for environmental light and to promote light therapy for the regulation of human circadian rhythms. In the early studies, some researchers examined the effects of various illuminances [3,10] or wavelengths [14] on circadian rhythms. Following the finding of the mRGC, some researchers investigated circadian phase shifts by using monochromatic and polychromatic lights with short wavelengths in order to improve efficacy at inducing circadian phase shifts. Lockley et al. [11] exposed subjects to 460 nm and 555 nm of monochromatic lights and found that the human circadian melatonin response was more sensitive to the shorter wavelength. Warman et al. [22] demonstrated that a low-intensity short-wavelength light pulse (8 lx, 28 W/cm2 , 6.21 × 1013 photons/cm2 /s) was able to phase advance the human circadian system by a similar magnitude to a bright white light pulse (12,000 lx, 4300 W/cm2 , 1.15 × 1016 photons/cm2 /s). However, it is important to note that the light stimuli used in these studies also affected the other photoreceptors, the rods and cones, as well as the mRGCs. In these cases, therefore, it is not clear whether the circadian responses were caused only through the mRGCs or by the rods and cones also. As it is important to understand mRGC characteristics and their role independent of effects due to the rods and cones, mRGC responses should be caused or measured independent of cone and rod responses. It seems that Kremers et al. [9] also noticed the same necessity and performed an experiment with cone-isolating stimuli, which were based on a silent substitution paradigm [6], and the heterochromatic flicker photometry electroretinogram (ERG) in order to quantify each
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
283
Fig. 1. A diagram illustrating the experimental set-up of the electronic and optical equipment, the integrating sphere exposure and the monitor array.
contribution of the different cone types. However, they have not investigated the mRGC contribution in ERG. We report here the first results from light stimuli which differentially induce only the mRGC response in ERG. They are highly relevant to providing further knowledge for studies in biological rhythms. The illumination system has been purpose-built with a receptorsilent substitution technique [21]. It consisted of an optical diffuser and an integrating sphere which presented an 18.9◦ circular field on the optical diffuser (Fig. 1). Four different kinds of light-emitting diodes (LEDs) (Opto Supply Limited, Hong Kong, China) were used as internal light sources in the integrating sphere. The peak wavelengths of the four LEDs were 633 nm, 593 nm, 507 nm and 468 nm with half-height bandwidths 15 nm, 14 nm, 33 nm and 22 nm, respectively. The light emitted from four types of LEDs, which were embedded in the inner wall of the integrating sphere, projected as the internally synthesized test stimuli. Radiance output of each LED was controlled by analogue pulse-width modulation (PWM) units by adjusting the duty cycle of the pulse train to 1 kHz. The PMW units were controlled by a micro computer (H8/3052 Renesas Technology, Japan). The characteristic of dutyluminance was calibrated to minimize the deviation caused by thermal effects. The excitation expresses stimulus levels to each photoreceptor. The 10◦ cone fundamentals, proposed by Stockman et al. [18,19], and spectral radiance of the light stimuli were used to calculate the excitation of cones sensitive to long (L), middle (M) or short (S) wavelengths. We estimated the spectral sensitivity of melanopsin from a pigment template nomogram [5] with a peak wavelength (max ) of 482 nm, as Dacey et al. [4] had shown that a spectral tuning curve of melanopsin was closely approximated by a pigment template with a peak at 482 nm. We used the fundamental, the spectral sensitivity of mRGC, in a 10◦ field displaying a peak wavelength at 502 nm after careful calibration of prereceptoral filters [21]. An excitation of the mRGC was then calculated from the fundamental (with a unity peak) as relative luminous efficiency and the spectral radiance of the light stimuli. The subject’s left pupil was maximally dilated by a mydriatic agent for electroretinogram (ERG) recordings and topical anesthesia for the relief of pain was provided. Following this, an electrode (EA-102, Meiyo, Japan) was placed in contact with the subject’s left cornea for full-field recording. The corneal surface was protected with a non-irritant and non-allergenic ionic conducting solution. A reference electrode was placed on the subject’s forehead and another electrode was attached to the ear as a ground. Either a vision therapist or an ophthalmologist was present throughout the
experiment in order to provide technical treatment and ensure good ocular contact, proper electrode impedance and ophthalmologic safety. Ten healthy subjects (5 females, 5 males; mean ± SD age, 24.3 ± 3.4 years; range 21–31 years; Japanese) were studied at Fukuoka Women’s University in Japan. The study was approved by the Ethics Committee at Fukuoka Women’s University and subjects gave written informed consent prior to study. All subjects had ocular health and normal color vision according to the Ishihara color blindness test. Fig. 1 illustrates the apparatus used in the experiment. The light from the integrating sphere was modulated to a sinusoidal wave and assessed with a spectroradiometer (CS-1000A, Konica Minolta, Japan). Fig. 2(a) and (b) shows examples of a light stimulus in 50% contrast. Peaks of the sinusoidal stimuli are described as high (H), minima as low (L), and constant stimuli without a sinusoidal wave as base (B). The light exposure consisted of 30
Fig. 2. Examples of a light stimulus in 50% contrast. (a) A light stimulus with sinusoidal wave modulation as an example of 50% contrast. The light varies from base (B) to high (H) and low (L) for 3 s and then maintains the base level for 2 s. (b) Spectral radiance distributions of base (B), high (H) and low (L) light stimuli in 50% contrast.
284
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
Table 1 An example of stimulus excitation to L, M, S cones and mRGC in 50% contrast.
Base (B) High (H) Low (L)
L cone
M cone
S cone
mRGC
100.0 99.9 99.2
100.0 100.1 99.3
100.0 98.1 100.6
100.0 152.3 47.1
The relative excitation of base (B), high (H) and low (L) is shown in percentages.
repetitions of 3-s exposure to the sinusoidal wave and 2-s constant (base) light irradiation (Fig. 2(a)). For each 3-s exposure, we were able to choose the light frequency and contrast with the PWM unit. We conducted the following two experiments: (1) contrast changes of 10, 20, 30, 40 and 50% with 5.0 Hz frequency and (2) frequency changes of 0.5, 1.0, 2.0, 5.0, 8.0, 12.0 and 30.0 Hz with 50% contrast. These light stimuli were designed with the help of Tsujimura et al. [21]. The ERG signals from photoreceptors were continuously digitized during the experiment at a sampling rate of 5 kHz by a data input system. The spectral radiance distributions merged light from four kinds of LEDs as shown in Fig. 2(b). The stimulus excitation of L, M and S cones and mRGCs was calculated by multiplying these spectral radiance distributions and the fundamentals of L, M and S cones and mRGCs. We stress that the stimulus levels to the cones did not vary in the study. As shown in Table 1, in the case of 50% contrast, each level of the base excitation to L, M and S cones and mRGCs was maintained as 100%, and only the mRGC excitation was varied – 50% higher than the base excitation in the high excitation and 50% lower in the low excitation; that is, the integrating sphere emitted light which changed only the mRGC excitation but not that of the cones. Furthermore, because the radiance saturated rod sensitivity to the light stimuli during the 5-min adaptation period before the emission, we presumed that rod responses did not appear on the ERG. The experiment was performed from 9 a.m. to 4 p.m. in order to exclude effects of circadian rhythms on ocular responses. A mydriatic agent was dropped into the subject’s left eye for pupil’s dilation in a waiting room; pupils became completely dilated during the next 30 min while the subjects sat quietly. The subjects then entered an artificial climate chamber. They were asked to sit quietly, rest their head in a chin rest, and gaze at the fixation at the center of stimulus circle (100 mm in diameter) on the diffuser in front of them. The diffuser was at a distance of 300 mm from a subject whose visual angle to the circle was 18.9◦ (Fig. 1). After 5-min adaptation in the chamber, an ophthalmologic technician applied ERG electrodes to subjects. In the first experimental session, effects of contrast changes were observed (the order of contrast changes was randomly arranged). The experiment lasted for approximately 30 min, during which time subjects were instructed to keep their eyes open and try to avoid blinking, especially when the integrating sphere emitted experimental light for the 3-s period. The mRGC response to the photostimuli was monitored by a personal computer during the experiment. In the second experimental session, effects of frequency changes were observed (the order of frequency changes was randomly arranged). The session started 30 min after the end of the first, to minimize stress induced by an electrode in contact with the cornea. After the two experimental sessions, subjects received eye-drops containing an antibacterial agent and were released from the laboratory. Power and phase spectra of mRGC responses were calculated by using Fast Fourier Transformation (FFT). The power indicates degree of the mRGC response and the phase expresses latency in the mRGC response to the sinusoidal light stimuli. We observed the phase response of the mRGC to the light stimuli in order to extract the mRGC response from other artifacts. We show the phase in degree as phase shifts of the mRGC response to the light stimuli with
Fig. 3. The mRGC response to the light stimuli with 10, 20, 30, 40 and 50% contrasts modulated at 5 Hz. (a) Scatter plot of the power in the ERG. The power is an index of the mRGC response to the light stimuli. The identical symbols express the individual power from the same subject and the emphasized and filled circles the medians with the linear approximation (solid line, R2 = 0.999). There was a significant correlation between the mRGC response and the contrasts (rs = 0.66, p < 0.001). The power varied linearly according to the contrasts. (b) The relationship between the contrasts and the phase. The phase in degrees (left axis) represents phase shifts converted into 360◦ of the mRGC response to the sinusoidal light stimuli. The phase in msec (right axis) shows latency of the mRGC response to each light stimulus. The dots express the averages, and the bars the standard deviations. There was no significant difference in these measures with changing contrasts of stimuli.
the sinusoidal wave and converted this measure into the latency to each stimulus in milliseconds. The homogeneity of variances of data was analyzed with SPSS, and appropriate parametric tests or nonparametric tests were chosen for data analysis. Fig. 3(a) shows the power in the ERG, as an index of the mRGC response following light stimuli, with five contrasts modulated at 5 Hz. The response appears to elevate linearly from 10% to 50% contrast. As the test of homogeneity of variances indicated that the data of the mRGC were not homoscedastic, nonparametric tests were applied for these results. Spearman’s rank correlation coefficient between the mRGC response and the contrasts showed high statistical significance (rs = 0.66, p < 0.001). Fig. 3(b) shows the relationship between the contrasts and the phase in degree and in msec. One-way ANOVA showed that the phase did not differ significantly between the contrasts. This means that we could collect data on mRGC responses according to the contrasts. Fig. 4(a) shows the power in the ERG, which expresses the mRGC response to the light stimuli with 50% contrast at seven frequencies. The response varied according to the frequency. One-way ANOVA indicated that the data of the mRGC response were homoscedastically distributed, and there were significant differences in mRGC responses with frequency (p < 0.01). Pearson’s correlation did not show any significant association between the frequencies and the mRGC response. For the phase of seven different frequencies,
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
285
change with increased frequency in a simple manner, as shown in Fig. 4(b). The phase would be predicted to change linearly with frequency if it was only due to the latency in the mRGC response; however, the results do not support this possibility as the variance in the low-frequency range differed from that in the high-frequency range. The phase may be affected not only by the latency but also by other, unexpected factors – feedback from the optic nerve tract, for example. Such an interaction could explain the observed complex latency response. Honma and Honma [8] and Minors et al. [12] reported that phase responses of human circadian rhythms varied depending on when exposure to light took place and the light intensity. The observed phase advances and delays to light can be ascribed to photosensitivity in the eye as well as individual responses and physiological regulation in the hypothalamus. Morita et al. [13] demonstrated that human eye function has a diurnal rhythm, using the 100-hue test in which subjects arranged color chips according to gradation of hue. Thus, it is also possible that mRGC responses vary with times of day. There were two different experimental sessions in this study, in the morning and in the afternoon. There was no significant difference in the results between the two times of testing, although the mRGC response was slightly higher in the afternoon than in the morning. Factors affecting mRGC responses by light of different frequencies are still unknown; clarification of diurnal physiological rhythms and the mechanisms involved in mRGC responses will be the subjects of our next study. In addition, further studies, such as in animal models with the illumination system and ERG, will enable us to verify and interpret ERG measurements and that should contribute directly to further knowledge of mRGC characteristics. Acknowledgement
Fig. 4. The mRGC response to the light stimuli with 50% contrast at seven frequencies. (a) The ERG power of the mRGC response to the light stimuli. The results are expressed in the same way as in Fig. 3(a) but the medians fitted the quadratic approximation (R2 = 0.628). (b) The phase in msec with seven frequencies. The dots express the averages, and the bars the standard deviations. The variance in low-frequency range between 0.5 and 2 Hz was significantly higher than that in high-frequency range between 5 and 30 Hz (p < 0.001).
the data were not homoscedastic. Although nonparametric tests showed that there were no significant differences between the frequencies, the phase does appear to vary. Fig. 4(b) shows the latencies of mRGC responses to the light stimuli. Brown–Forsythe test indicated that the variance in low-frequency range between 0.5 and 2 Hz was significantly higher than that in high-frequency range between 5 and 30 Hz (p < 0.001). Absolute deviations from medians were 283.7 in the low-frequency range and 19.6 in the high-frequency range. The results in this study demonstrate that the test stimuli affected only mRGC responses and that the response could be successfully detected by using the ERG. Although the stimuli would have affected cone and rod photoreceptors, the stimulus excitation to these receptors was the same throughout the experiment, and this means that changes in responses measured by the ERG were due to changes in mRGC excitation only. It was clear that the mRGC responded to the light contrasts. As shown in Fig. 3(a), the power of the mRGC response increased according to the contrasts of 10, 20, 30, 40 and 50%. However, the changes in the response with the stimulus frequency illustrated a non-linear curve, as indicated in Fig. 4(a). This result suggests that mRGC behavior is not a simple response to the various frequencies found in solar light. Similarly, the latency of the response time, derived from the phase with respect to the light stimulus, did not
We gratefully appreciate the editorial support and helpful advice and comments on the manuscript of Prof. Jim Waterhouse at Liverpool John Moores University. References [1] D.M. Berson, F.A. Dunn, M. Takao, Phototransduction by retinal ganglion cells that set the circadian clock, Science 295 (2002) 1070–1073. [2] G.C. Brainard, J.P. Hanifin, J.M. Greeson, B. Byrne, G. Glickman, E. Gerner, M.D. Rollag, Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor, J. Neurosci. 21 (2001) 6405–6412. [3] C.A. Czeisler, G.S. Richardson, J.C. Zimmerman, M.C. Moore-Ede, E.D. Weitzman, Entrainment of human circadian rhythms by light–dark cycles: a reassessment, Photochem. Photobiol. 34 (1981) 239–247. [4] D.M. Dacey, H.-W. Liao, B.B. Peterson, F.R. Robinson, V.C. Smith, J. Pokorny, K.-W. Yau, P.D. Gamlin, Melanopsin-expressing ganglion cells in primate retina signal colour and irradiance and project to the LGN, Nature 433 (2005) 749–754. [5] H.J. Dartnall, The interpretation of spectral sensitivity curves, Br. Med. Bull. 9 (1953) 24–30. [6] O. Estévez, H. Spekreijse, The ‘silent substitution’ method in visual research, Vision Res. 22 (1982) 681–691. [7] S. Hattar, H.-W. Liao, M. Takao, D.M. Berson, K.-W. Yau, Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity, Science 295 (2002) 1065–1070. [8] K. Honma, S. Honma, A human phase response curve for bright light pulses, Jpn. J. Psychiatry Neurol. 42 (1988) 167–168. [9] J. Kremers, T. Usui, H.P.N. Scholl, L.T. Sharpe, Cone signal contributions to electrograms in dichromats and trichromats, Invest. Ophthalmol. Vis. Sci. 40 (1999) 920–930. [10] A.J. Lewy, T.A. Wehr, F.K. Goodwin, D.A. Newsome, S.P. Markey, Light suppresses melatonin secretion in humans, Science 210 (1980) 1267–1269. [11] S.W. Lockley, G.C. Brainard, C.A. Czeisler, High sensitivity of the human circadian melatonin rhythm to resetting by short wavelength light, J. Clin. Endocrinol. Metab. 88 (2003) 4502–4505. [12] D.S. Minors, J.M. Waterhouse, A. Wirz-Justice, A human phase-response curve to light, Neurosci. Lett. 133 (1991) 36–40. [13] T. Morita, M. Ohyama, H. Tokura, Diurnal variation of hue discriminatory capability under artificial constant illumination, Experientia 50 (1994) 641–643. [14] T. Morita, H. Tokura, The influence of different wavelengths of light on human biological rhythms, J. Physiol. Anthropol. Appl. Human Sci. 17 (1998) 91–96.
286
Y. Fukuda et al. / Neuroscience Letters 479 (2010) 282–286
[15] S. Panda, T.K. Sato, A.M. Castrucci, M.D. Rollag, W.J. DeGrip, J.B. Hogenesch, I. Provencio, S.A. Kay, Melanopsin (Opn4) requirement for normal light-induced circadian phase-shifting, Science 298 (2002) 2213–2216. [16] I. Provencio, I.R. Rodriguez, G. Jiang, W.P. Hayes, E.F. Moreira, M.D. Rollag, A novel human opsin in the inner retina, J. Neurosci. 20 (2000) 600–605. [17] I. Provencio, M.D. Rollag, A.M. Castrucci, Photoreceptive net in the mammalian retina, Nature 415 (2002) 493. [18] A. Stockman, L.T. Sharpe, The spectral sensitivities of the middle- and longwavelength-sensitive cones derived from measurements in observers of known genotype, Vision Res. 40 (2000) 1711–1737. [19] A. Stockman, L.T. Sharpe, C. Fach, The spectral sensitivity of the human shortwavelength sensitive cones derived from thresholds and color matches, Vision Res. 39 (1999) 2901–2927.
[20] K. Thapan, J. Arendt, D.J. Skene, An action spectrum for melatonin suppression: evidence for a novel non-rod, non-cone photoreceptor system in humans, J. Physiol. 535 (2001) 261–267. [21] S. Tsujimura, K. Ukai, D. Ohama, A. Nuruki, K. Yunokuchi, Contribution of human melanopsin retinal ganglion cells to steady-state pupil responses, Proc. R. Soc. B., doi:10.1098/rspb.2010.0330. [22] V.L. Warman, D.-J. Dijk, G.R. Warman, J. Arendt, D.J. Skene, Phase advancing human circadian rhythms with short wavelength light, Neurosci. Lett. 342 (2003) 37–40.