Phototransduction by Retinal Ganglion Cells That Set the Circadian Clock
Tóm tắt
Từ khóa
Tài liệu tham khảo
von Schantz M., Provencio I., Foster R. G., Invest. Ophthalmol. Vis. Sci. 41, 1605 (2000).
Methods conformed to NIH guidelines and were approved by Brown University's Institutional Animal Care and Use Committee. Rats (Sprague-Dawley 56 to 70 days of age 265 to 355 g; n = 256) were anesthetized with ketamine [60 mg/kg intraperitoneal (ip)] and medetomidine (0.4 mg/kg ip). Fluorescent latex microspheres (rhodamine-labeled alone or mixed with fluoroscein-labeled microspheres; Lumafluor; 0.1 to 0.3 μl) were deposited unilaterally into the hypothalamus through glass pipettes tilted 10° from vertical. As expected (1) deposits involving the ventral hypothalamus but sparing the optic chiasm labeled from a few ganglion cells to as many as several hundred scattered over the retina. These cells are the focus of this report. Deposits spreading to the optic chiasm nonspecifically labeled thousands of ganglion cells some of which were targeted for control recordings. Deposits involving overlying structures but sparing the SCN and chiasm labeled no ganglion cells.
Rats were anesthetized (Nembutal 120 mg/kg ip) 2 to 28 days after tracer injection and their eyes were removed and hemisected. Eyecups were rinsed in an enzyme solution [collagenase/dispase (2 mg/ml); DNAase (Sigma A1420; 0.1 mg/mL) in Ames medium 1 min] to remove vitreous. Retinas were isolated affixed to a cover slip with gelatin vitreal surface up and mounted in a chamber (Warner RC-26GLP). Retinas were maintained at room temperature in bicarbonate-buffered Ames medium and fluorescent-labeled cells were identified as described (35). Detachment from pigment epithelium and exposure to bright light during dissection (∼1 × 10 17 photons s −1 cm −2 measured at 500 nm) and epifluorescence examination (∼3 × 10 17 photons s −1 cm −2 at 560 nm) presumably strongly bleached rod and cone photopigments. Ganglion cell somata were exposed by microdissection. Whole-cell current clamp recordings were made with an intracellular amplifier (Cygnus Technologies DR-886 Delaware Water Gap PA) and micropipettes (3 to 7 megaohms) containing 125 mM K-Gluconate 5 mM NaCl 4 mM KCl 10 mM EGTA 10 mM Hepes 4 mM adenosine triphosphate–-Mg 7 mM phosphocreatine 0.3 mM guanosine triphosphate–tris LY (0.1% w/v) and biocytin (0.5% w/v). Resting potentials were not corrected for liquid junction potentials. Light stimuli were introduced from below with the microscope's 100-W tungsten-halogen lamp and transillumination optics. Neutral density and narrow-band interference filters (Oriel Stratford CT) controlled stimulus energy and wavelength. Energies were measured with a calibrated radiometer (UDT Instruments Baltimore MD). Except where noted all stimuli were full-field and spectrally unfiltered.
The probability of obtaining such light responses was very high among back-filled cells in very sparsely labeled retinas but declined substantially when labeled ganglion cells totaled >100 presumably because involvement of the chiasm reduced labeling specificity. Peak depolarizations for saturating stimuli typically ranged from 15 to 30 mV.
Cobalt chloride was superfused at a concentration (2 mM in Ames) that blocks all light-evoked transmitter release from rods and cones in rat retina (36) and eliminates the otherwise robust light responses of ganglion cells in rat and cat eyecup preparations (37). Small chamber volume (<200 μl) and inlet tubing dead space (<2 ml) ensured complete exchange in minutes. Light responses also persisted when 2 mM [Co 2+ ] o replaced 2 mM [Ca 2+ ] o in a bath solution containing 126 mM NaCl 3 mM KCl 1.3 mM NaH 2 PO 4 2 mM MgSO 4 26 mM NaHCO 3 and 10 mM dextrose ( n = 2).
In darkness resting potentials of photosensitive (−46.7 ± 14.8 mV; n = 114) and control ganglion cells (−48.5 ± 11.0 mV; n = 45) were statistically indistinguishable ( P > 0.2; t test).
For most cells ( n = 4; including that shown in Fig. 1G) the drug solution consisted of 2 mM CoCl 2 100 μM l (+)-2-amino-4-phosphonobutyric acid to saturate and thus block signal transfer at Group III metabotropic glutamate receptors essential for photoreceptor-to-ON-bipolar cell transmission 20 μM 6 7-dinitroquinoxaline-2 3-dione and 50 μM dl -2-amino-5-phosphonovaleric acid to block ionotropic glutamate receptors [both N -methyl- d -aspartate (NMDA) and non-NMDA types] essential for communication between photoreceptors and OFF bipolar cells and between all bipolar cells and both amacrine and ganglion cells and 50 μM picrotoxin and 0.3 μM strychnine to block ionotropic gamma-aminobutyric acid (GABA) and glycine receptors mediating most inhibition by amacrine cells. In two other cells this cocktail was supplemented with 200 μM hexamethonium bromide to block nicotinic acetylcholine receptors and either 500 or 2000 μM 1-octanol to block gap junctions; light-evoked depolarizations were not attenuated. Blockade typically enhanced depolarization (e.g. Fig. 1 F and G; mean depolarization = 18.5 mV in control medium and 26.7 mV during blockade; n = 7). This may reflect increased membrane resistance due to loss of synaptic currents or of intrinsic calcium conductances. Resting potential was not significantly altered by synaptic blockade (means = 52.8 mV in Ames and 51.5 mV during blockade; n = 7).
Sensitivity usually deteriorated noticeably within 30 to 120 min of break-in probably because of cell dialysis (“run-down”). Group data in Fig. 2C (green curve) were therefore drawn only from responses obtained shortly after break-in first at 500 nm and then at one other wavelength. Slow recovery kinetics necessitated long interstimulus intervals (3 min). Thus spectra for single cells were typically less complete than that shown for cell 62-4 (Fig. 2C red curve). Nonetheless the general form of the action spectrum was consistent. For example among each of seven cells examined the same rank order of sensitivity applied to the first three wavelengths tested: 500 nm > 400 or 420 nm » 570 or 600 nm.
Differences between these physiological and behavioral curves may be attributable in part to spectral filtering by the lens which presumably influenced the behavioral spectrum but not the data we obtained in isolated retina. Available spectral data do not exclude contributions of rods or green cones to the entrainment mechanism (17 26).
Values are based on a 2-mm pupil and retinal area of 1.13 cm 2 (38). For comparison absolute rod threshold in the rat is ∼2 × 10 −4 cd/m 2 [(39) as cited by (27)] or ∼ 10 7 photons cm −2 s −1 retinal irradiance.
Nelson D. E., Takahashi J. S., Am. J. Physiol. 277, R1351 (1999).
Immunostaining for LY avidin-biotin-peroxidase reaction and diaminobenzidine histochemistry were as described (35).
___, J. Biol. Rhythms 15, 31 (2000).
Receptive fields were probed with spots or bars generated by placing an aperture in the microscope's spectrally unfiltered transilluminating tungsten beam and focusing it on the tissue with a ×20 objective lens. There was substantial spatial summation within the receptive field but no surround antagonism an otherwise ubiquitous feature of ganglion cell receptive fields.
F. A. Dunn B. J. O'Brien unpublished data.
D. M. Berson F. A. Dunn M. Takao data not shown.
We thank B. Wooten for assistance with the spectral analysis; M. Seibert for help with the receptive field study; T. Nguyen for technical assistance; and H. Wässle B. O'Brien J. McIlwain I. Provencio M. Slaughter and L. Peichl for their valuable comments on the manuscript. Supported by NIH grant EY12793 to D.M.B.