Intrinsically photosensitive retinal ganglion cells detect light with a vitamin A-based photopigment, melanopsin -- Data Supplement - HTML Page - 01866SuppText.htslp -- Proceedings of the National Academy of Sciences

Supporting Text
Supporting Materials and Methods

The experimental and animal care procedures were in accordance with institutional guidelines and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

X-Gal Labeling.
For X-Gal labeling of the flat-mounted retina and brain sections of rpe65^-/-opn4^+/-and rpe65^-/-opn4^-/- mice, an animal was anesthetized and circulationally perfused at ~1.5 ml/min with 15 ml of cold PBS and 7-8 ml of cold fixative (4% paraformaldehyde in PBS). To stain for b -galactosidase activity, the animal was further perfused successively with 15 ml of each of the following solutions: PBS, buffer B [100 mM phosphate buffer, pH 7.4/2 mM MgCl₂/0.01% Na-desoxycholate/0.02% (octylphenoxy)polyethoxyethanol], and staining solution (buffer B plus 5 mM K-ferricyanide/5 mM K-ferrocyanide/1 mg· ml^-1 X-Gal). The perfusion speed was reduced to 0.2 ml/min for the staining solution. Mouse eyes and brains were then isolated and incubated in the staining solution overnight at room temperature in light-proof containers. An eyecup was prepared by removing the anterior half of the eye, and the retina was isolated and flat-mounted. The brain was embedded in OCT (Tissue-Tek, Sakura Finetek, Torrance, CA) and sectioned at 100 m m. Brain sections were mounted on microscope slides (Fisherbrand, superfrost-plus) and allowed to air dry before coverslipping in glycerol. Images were obtained using a Zeiss Axiophot microscope and an AxioCam HRc digital camera. Whole-Cell Recording from IpRGCs.
The experimental procedure was as described in refs. 1 and 2. Briefly, red fluorescent latex beads (Lumafluor) were stereotaxically injected into the SCN of 2- to 5-month-old mice. In 2 days to 1 month after injection, an animal dark-adapted overnight was euthanized by CO₂ asphyxiation, and the eyes were removed under red light (>630 nm). A flat-mounted retina was prepared with the RGC layer facing up, and the retrogradely labeled RGCs were identified on an upright fluorescence microscope. Some bleaching of the retina was inevitable during the search for labeled cells by fluorescence; this was somewhat alleviated by the use of red beads (excitation at 546 nm) instead of the green beads described by Berson et al. (1) because the ipRGC photopigment apparently absorbs maximally at ~480 nm.

The retina was superfused with bicarbonate-buffered Ames solution (Sigma) containing a mixture of six drugs: 50 m M DL-2-amino-4-phosphonobutyric acid, 10 m M (RS)-3-(2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid, 20 m M 6,7-dinitroquinoxaline-2,3-dione, 50 m M picrotoxin, 0.3 m M strychnine, and 200 m M hexamethonium bromide. Experiments were performed at ~33-35°C. A clearance was first created on the inner limiting membrane of the retina with a glass micropipette before whole-cell recording was carried out with a 5-7 MW patch electrode using infrared DIC optics. Pipette solution contained (in mM) 106 K-gluconate, 10 Hepes, 5 MgCl₂, 4 K₂ATP, 0.4 Na₂GTP, 7 Na₂-phosphocreatine, 2 EGTA, and 0.5 CaCl₂ (pH 7.4 by KOH); overall osmolarity was 262 milliosmolar. Signals were routinely low-pass filtered at 1 kHz (8-pole Bessel) and digitized at 2 kHz. Voltage-clamp signals were further low-pass filtered at 20 Hz by a digital 8-pole Bessel filter (Clampfit 8.2) before analysis. For all voltage-clamp experiments, the membrane potential was held at -70 mV. A junction potential of about -10 mV has not been corrected in the figures. Light of controlled wavelength, intensity, and duration was fed from an optical bench by means of fiberoptics into a switchable epifluorescence port on the microscope. The circular light spot on the retina was controlled by a field diaphragm and adjusted typically to 300 m m in diameter centered at the soma of the recorded cell. This less-than-diffuse illumination was in part to reduce space-clamp problems. In most experiments, 20-msec flashes of 480-nm light were used for stimulation. When much stronger stimulation was necessary, white light was used; the white-light intensity could be converted into an equivalent 480-nm light by comparing weak responses of ipRGC elicited under the two conditions (data not shown).

The responses at low light intensities often showed a slight supralinearity, reflected by a Hill coefficient of 1.0-1.3 for the intensity-response relation. To compare flash sensitivity across cells, we divided response amplitudes in the range of 20-50 pA by the corresponding flash intensities. An implicit assumption in this calculation is that, on average, each incident photon per unit area elicits a comparable-size response from a cell under normal conditions. Strictly speaking, this is not the case because, among other things, the overall cell morphology under the light spot (therefore the light-capturing surface area) varies from one ipRGC to another. However, this constitutes the best approach at present. Some of the rpe65^-/- ipRGCs were very insensitive, giving responses < 20 pA even at the highest intensity used; in these cases, flash sensitivity was calculated from the largest possible response elicited.

Suction-Pipette Recording.
An animal dark-adapted overnight was euthanized by CO₂ asphyxiation, and the eyes were removed and hemisected under dim red light. The retina was removed under infrared light, cut into several pieces, and stored in L-15 medium (GIBCO) on ice for up to 7 hr. When needed, a retinal piece was chopped with a razor blade under chilled L-15 medium and on a cured-Sylgard surface. The fragments were transferred to the recording chamber and perfused at 37°C with bicarbonate-buffered Locke’s solution (in mM): 149 NaCl, 3.6 KCl, 2.4 MgCl₂, 1.2 CaCl₂, 10 Hepes (pH 7.4), 0.02 EDTA, 3 NaHCO₃, 3 disodium succinate, 0.5 sodium glutamate, 0.1 vitamins and amino acid supplement (GIBCO), and 10 glucose, bubbled with 95% O₂/5% CO₂. Membrane current was recorded from a single rod outer segment projecting from a retinal fragment with the suction-pipette recording method (3). Light stimuli consisted of unpolarized, 10-msec flashes at 500 nm. Signals were low-pass filtered at 20 Hz (8-pole Bessel) and sampled at 500 Hz. ERG Measurements.
Preparation and recording methods were as described in ref. 4. Recordings were made from five adult opn4^-/- and five adult WT (B6/129) mice, as well as seven adult C57BL/6 mice (as another control). The animals, kept in a 12:12-hr light/dark cycle, were dark-adapted overnight and prepared for recording under red illumination (LED, l > 640 nm). An animal was anesthetized with an i.p. injection of ketamine (60 mg/kg) and xylazine (6 mg/kg), and anesthesia was maintained with ketamine (56 mg/kg) and xylazine (5.6 mg/kg) given every 45-60 min by means of a s.c. needle fixed in the flank. The nose was clamped and the two front teeth held in a metal bite bar that also served as the electrical ground. Pupils were fully dilated to 3 mm in diameter with topical atropine (0.5%) and phenylephrine (2.5%). Rectal temperature was maintained at 37-38°C. ERGs were recorded differentially between DTL fiber electrodes (5), moistened with methylcellulose sodium (Celluvisc, Allergan, Irvine, CA), and placed across the center of the cornea of each eye. The cornea of the tested eye was covered with a clear contact lens; the untested eye was covered with a black contact lens and a black aluminum-foil cap that covered also the skull. Recording sessions lasted about 4 hr for dark-adapted ERGs, and up to 8 hr when backgrounds were imposed. The mice were allowed to recover from anesthesia after the session.

The ganzfeld stimulus consisted of brief (0.8-4.1 msec) full-field flashes from LEDs (l _max at 462 nm for dark-adapted experiments and 513 nm for most light-adapted experiments), with intensities ranging from being too weak to produce measurable responses [-6.1 log scotopic troland seconds (sc td s)] to being strong enough (3.6 log sc td s) to elicit a saturated a-wave under dark-adapted conditions. Stimuli stronger than 1.5 log sc td s were produced with a xenon flash tube in the ganzfeld stimulator. Full-field steady backgrounds (LEDs, l _max = 462 nm or 513 nm) ranged from -3.2 to 2.5 log sc td. The interval between flashes was adjusted so that the ERG had returned to baseline before another stimulus was presented. The ganzfeld stimuli were produced by illumination of a concave white diffuser (1-1.5 mm in diameter) positioned very close to the eye. Time 0 was taken as halfway through the flash. Responses were averaged over many trials for weak stimuli and over fewer trials for strong stimuli. Signals were amplified and low-pass filtered at 300 Hz, digitized at 1 kHz, and digitally filtered off-line to remove 60 Hz. The luminance (sc cd· m^-2) and luminous energy (sc cd· s· m^-2) were measured with a scotopically corrected photometer (model IL1700, International Light). Conversion from troland values to photoisomerizations per rod (R*rod^-1) assumes that 1 sc td s gives 122 R*/rod (4). Scotopic calibrations were used for light-adapted experiments as well because the maximum sensitivity of the M-cone of the mouse is very near that of rods.

Supporting Results
Comparison of Dark-adapted ERG between Opn4^-/- and WT Mice.
As shown in Fig. 7A, with the two strongest stimuli, the dark-adapted a-wave had comparable amplitudes in both lines of animals (see the traces for +3.6 and +1.2 log sc td s and Inset). Measured at 6 msec after the flash, on the leading edge near the peak of the saturated response, the average V_max derived from a single-exponential fit was 658 ± 65 m V for opn4^-/- and 687 ± 116 m V for WT. The absolute sensitivity and the saturated amplitude of the dark-adapted b-wave were also similar for both groups of animals (Fig. 7B), with the intensity-response relation saturating at about 1 photoisomerized rhodopsin (R*) rod^-1, as previously reported (4, 6). A small difference was nonetheless observed in the amplitude of the negative scotopic threshold response (nSTR) (4, 7). Although the nSTR (as well as the positive nSTR, which constitutes the b-wave at corresponding intensities) showed similar sensitivities at low intensities (below -5.0 log sc td s) for the two groups of animals, the maximum amplitude of the nSTR was, on average, smaller for the opn4^-/- (-20 m V) than WT animals (-42 m V) (see -4.8, -4.2, and -3.6 traces in Fig. 7A and also the plot in Fig. 7C for averaged data). The nSTR in mice is believed to arise primarily from the syncitium of AII amacrine cells and perhaps also ON-ganglion cells (4). The difference in the nSTR may reflect some influence of ipRGCs on information processing in the inner retina.

1. Berson, D. M., Dunn, F. A. & Takao, M. (2002) Science 295, 1070-1073.

2. Hattar, S., Liao, H. W., Takao, M., Berson, D. M. & Yau, K. W. (2002) Science 295, 1065-1070.

3. Yang, R. B., Robinson, S. W., Xiong, W. H., Yau, K. W., Birch, D. G. & Garbers, D. L. (1999) J. Neurosci. 19, 5889-5897.

4. Saszik, S. M., Robson, J. G. & Frishman, L. J. (2002) J. Physiol. 543, 899-916.

5. Dawson, W. W., Trick, G. L. & Litzkow, C. A. (1979) Invest. Ophthalmol. Visual Sci. 18, 988-991.

6. Zhu, L., Jang, G. F., Jastrzebska, B., Filipek, S., Pearce-Kelling, S. E., Aguirre, G. D., Stenkamp, R. E., Acland, G. M. & Palczewski, K. (2004) J. Biol. Chem. 279, 53828-53839.

7. Sieving, P. A., Frishman, L. J. & Steinberg, R. H. (1986) J. Neurophysiol. 56, 1049-1061.