|
|
|
|
|
|
|
||
ARTICLE |
Correspondence to Edward N. Pugh Jr.: pugh{at}mail.med.upenn.edu
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-subunit of the G-protein transducin (Gt–/–), a loss that renders them functionally rodless. Most cones were found to functionally co-express both S- (
max = 360 nm) and M- (max = 508 nm) cone opsins and to be maximally sensitive at 360 nm ("S-cones"); nonetheless, all cones from the dorsal retina were found to be maximally sensitive at 508 nm ("M-cones"). The dim-flash response kinetics and absolute sensitivity of S- and M-cones were very similar and not dependent on which of the coexpressed cone opsins drove transduction; the time to peak of the dim-flash response was 
70 ms, and 0.2% of the circulating current was suppressed per photoisomerization. Amplification in WT cones (A 4 s–2) was found to be about twofold lower than in rods (A 8 s–2). Mouse M-cones maintained their circulating current at very nearly the dark adapted level even when >90% of their M-opsin was bleached. S-cones were less tolerant to bleached S-opsin than M-cones to bleached M-opsin, but still far more tolerant than mouse rods to bleached rhodopsin, which exhibit persistent suppression of nearly 50% of their circulating current following a 20% bleach. Thus, the three types of mouse opsin appear distinctive in the degree to which their bleached, unregenerated opsins generate "dark light."
 |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Paupoo et al., 2000), an ability involving a number of distinctive molecular and physiological factors that remain only partially understood (Pugh et al., 1999; Rebrik and Korenbrot, 2004). Second, cones generate the signals for color vision by virtue of their diverse spectral sensitivities and their spectrally "opponent" retinal connections (Dacey, 1996, 2000). Third, cones initiate vision in the macula, the highly specialized central region of the retina that maps to a large fraction of human primary visual cortex (Engel et al., 1997). Because of the roles that cone photoreceptors play in normal human vision, cone disease and cell death, as occurs in age-related macular degeneration, the leading cause of blindness in aging humans (Klein et al., 2002), is devastating. To investigate the molecular mechanisms that allow cones to perform their unique functions, and the molecular mechanisms of cone disease, it is critical to have mammalian models which allow (a) genomic analysis and manipulation of genes expressed specifically in cones, (b) molecular and biochemical characterization of the protein products of such genes, and (c) electrophysiological analysis of cones and their neural circuits.
The mouse is the mammal of choice for the investigation of organ function and the molecular mechanisms of disease. There are many reasons for this choice, including the genomic proximity of mice to humans, the large and rapidly growing array of molecular biological tools for targeted gene manipulations in mice, the large knowledge base of molecular, cellular, and behavioral experimentation using mice, and the relatively short generation time and economics of mouse husbandry. Nonetheless for these compelling reasons, the investigation of the functional consequences of molecularly manipulated cone-specific genes in mice has been an elusive goal, having only been achieved in a few studies using electroretinographic methods (Lyubarsky et al., 2000, 2001; Pennesi et al., 2003a,b). In contrast, while recordings from individual mouse rods (most with targeted gene manipulations) have been presented in at least 35 primary publications since the report by Chen et al. (1995), not a single paper has yet been published describing single-cell recordings from WT mouse cones. We believe this defect to arise from a number of factors, including (a) the 30-fold numerical dominance of rods over cones in mouse retina (Carter-Dawson and LaVail, 1979), (b) the lack of morphological features distinguishing cones from rods in mouse retinal slices viewed under the infrared illumination requisite for single-cell recording, and (c) the relative lability of cone vs. rod outer segments removed from their interphotoreceptor matrix sheaths. The latter lability was revealed in experiments with mice lacking the neural retina leucine zipper transcription factor (Nrl–/–) (Nikonov et al., 2005).
The apparent fragility of Nrl–/– outer segments provided the impetus for the development of a novel "loose-patch" method, in which a portion of the photoreceptor circulating current was recorded by drawing the "inner segment" (perinuclear region) of mouse photoreceptors in a retinal slice into a suction pipette (Nikonov et al., 2005). By application of this new method, along with a battery of other analyses, including EM analysis of ultrastructure, quantification of cone-specific proteins, and spectral and kinetic criteria, it was unequivocally established that Nrl–/– photoreceptors are a species of cones (Daniele et al., 2005; Nikonov et al., 2005), and not the "cone–rod" (or "cod") intermediates previously supposed (Mears et al., 2001). While the classification of Nrl–/– photoreceptors as cones has opened the door to the identification and characterization of many cone-specific genes (Yoshida et al., 2004), questions remain as to the ultimate validity of the Nrl–/– retina as model system for the investigation of mouse cone physiology. In part, such questions arise because Nrl–/– outer segments exhibit a degree of disorder not present in their WT counterparts, and undergo a slow degeneration that is evident by 6 wk of age (Mears et al., 2001; Daniele et al., 2005). However, a critical question that inevitably arises and must be answered is whether WT mouse cones have functional properties like those of the cones of 4–6-wk-old Nrl–/– mice.
Here we address this question and establish that single cone photoreceptors of WT and Gt–/– mice can be characterized with the suction pipette method previously developed to record stable electrical responses of Nrl–/– cones. While most of the physiological features of WT cones determined with this method, including their response kinetics and amplification and their functional coexpression of both S- and M-cone opsins in most cells, are very similar to those of the Nrl–/– mouse, one notable difference was found. Thus, there appears to exist in the dorsal retina of the WT mouse a subset of cones that express M-opsin at a higher level than S-opsin. These "M-cones" appear more tolerant to high levels of bleached pigment than the predominant cone type in which the S-opsin is expressed at a higher level. (The mouse genome contains the genes for three opsins expressed in retinal photoreceptors: rhodopsin with max = 498 nm, and two cone opsins with max = 360 nm and 508 nm, respectively [Yokoyama and Yokoyama, 2000]. As the cone opsin with max = 360 nm is a member of the SWS1 family, which also contains the human S-cone opsin, and the cone opsin with max = 508 nm is a member of the LWS/MWS family, which contains the human M-cone opsin, throughout this paper we will simply identify the two mouse cone opsins as mouse "S-opsin" and "M-opsin.")
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
–/– mice were generated at the New England Medical Center (Calvert et al., 2000). Animals used for recordings were born and maintained in controlled ambient illumination on a 12 h light/dark cycle, with an illumination level of 2–3 lux, and dark adapted for at least 12 h before experimentation.
Tissue Preparation and Electrophysiological Methods
Mice were killed, the eyes enucleated, and whole retinas removed from eye cups under infra-red illumination. Small pieces of retina were dissected in a drop of chilled Locke's solution (112.5 mM NaCl, 3.6 mM KCl, 2.4 mM MgCl2, 1.2 mM CaCl2, 10 mM HEPES, 0.02 mM EDTA, 20 mM NaHCO3, 3 mM Na2-succinate, 0.5 mM Na-glutamate, 10 mM glucose), and placed into a recording chamber. The chamber was continuously refreshed with Locke's solution, pH 7.4, equilibrated with 95% O2/5% CO2, and maintained at 35–37°C with a heating system designed for microscopy (ALA Scientific). Using silanized suction pipettes, we recorded from photoreceptors embedded in 50–100-µm diameter slices of retina exclusively in the "OS out" configuration (Nikonov et al., 2005); in this effort several nuclei and conjoined "inner segment" tissue were intentionally drawn into the pipette. Once the tissue was drawn into the pipette, responses were evoked with calibrated flashes of light delivered under control of a customized LabView (National Instruments) interface. The optical system in the configuration used for these experiments has two stimulation channels: the light source in one channel is a tungsten-halogen lamp, and in the second a xenon flash lamp that delivers 20-µs pulses. Experiments with WT mouse retinal slices required the use of steady illumination to suppress rod activity, and the tungsten-halogen channel was employed for this purpose.
The "inner segment" limb of the rod and cone circulating current is an outward membrane current, carried primarily by K+ channels; light responses recorded from inner segment membranes are thus recorded by the amplifier as negative-going, resulting from the suppression of the outward membrane current as the cell hyperpolarizes toward the K+ reversal potential. Here we will present all photocurrent responses in the conventional manner as positive-going. However, the actual sign (and direction) of the recorded membrane currents will be referred to as needed.
As the expression of mouse M-cone opsin in mice varies in a dorso-ventral gradient (Applebury et al., 2000), we developed a method that allows the dorsal or ventral region of the retina to be dissected under infrared illumination and used for suction pipette recordings (Nikonov et al., 2005). This method has played a critical role in the complete characterization of cone function in the WT mouse.
Light Stimulation and Calibration; WT Mouse Cone Light Collecting Area
The methods of light stimulation and the calibration of flash and step intensities were as previously reported (Nikonov et al., 2005). The number of photoisomerizations 
per photoreceptor produced by a flash was estimated as the product of the energy density (photons µm–2) and the outer segment light collecting area, ac (µm2), calculated with the following formula:
 | (1) |
max is the extinction coefficient at the max of the pigment in solution, 
the quantum efficiency of photoisomerization, C the concentration (M) of the pigment in the outer segment, and VOS (µm3) the envelope volume of the outer segment, and the factor 10–4 is required for consistency with the dimensions of VOS. We previously summarized these factors in detail and obtained the estimates ac = 0.5 µm2 for WT mouse rods and ac = 0.11 µm2 for Nrl–/– cones for light flashes and steps delivered in our recording chamber at the max's of the three mouse opsins (Nikonov et al., 2005). In a separate investigation (Daniele et al., 2005), we summarized relevant quantitative features of the ultrastructure of the outer segments of Nrl–/– cones and of WT mouse cones and rods (Carter-Dawson and LaVail, 1979) that contribute to ac. In particular, the diameter and length of the WT mouse cone outer segment are 1.2 µm and 13.4 µm, respectively, yielding an envelope volume VOS = 14 µm3, as compared with VOS = 37 µm3 for the WT mouse rod. At a standard concentration of 3 mM relative to the envelope volume, a dark adapted mouse cone with VOS = 14 µm3 comprises a total of Ndark = 2.7 x 107 opsin molecules. Based on these facts, we thus estimate the transversely stimulated, fully dark-adapted WT mouse cone to have a collecting area ac = 0.2 µm2 at the max of its dominant opsin. This likely somewhat overestimates ac of the cones whose results are reported here for three reasons. First, as most cones coexpress both opsins, and as the concentration of total opsin in the outer segment is likely an approximately conserved quantity, the concentration of the principal cone opsin may be reduced somewhat due to coexpression of the second opsin. Second, photoreceptors have evolved to guide light from their inner segments to their outer segments, and cones in particular have inner segments whose tapering and refractive index distribution assist this guiding, and "impedance match" inner and outer segment refractive indices relative to the index of the extracellular space; transverse stimulation, particularly of very thin outer segments, may reduce collection efficiency due to the refractive index mismatch with the extracellular medium. Third, the nearly continuous exposure to the strong 500-nm background light used to suppress rod activity, and the exposure to the still more intense steps and flashes of light required to determine a cone's step response vs. intensity relation, produce substantial bleaching of the M-opsin. To deal with this latter problem, assuming no regeneration occurs, we programmed a (post-hoc) analysis of the data to create a "bleach progress meter" that estimated the fraction of unbleached pigment p(T) remaining at any given time T during an experiment. The rate equation for bleaching of a transversely stimulated photoreceptor can be written
 | (2) |
 | (3) |
Quantitative Analysis of Response Data
The activation phase of families of normalized responses R(t) were fitted with a model of the phototransduction cascade (Lamb and Pugh, 1992; Pugh and Lamb, 1993),
 | (4) |
" signifies a definition, r(t) is the photoresponse, rmax its saturating amplitude, the number of photoisomerizations produced by the flash, and teff a brief (several ms) delay. Traces computed with Eq. 4 were convolved with digital filters to incorporate the effects of the membrane time constants of cones (Smith and Lamb, 1997) (set to 
m = 5 ms), and the measured impulse response function of the 8-pole analogue Bessel filter, whose bandwidth was set to 20 Hz.
Amplitude vs. intensity functions were derived from flash response families and fitted with hyperbolic saturation functions of the form
 | (5) |
 | (6) |
is the steady-state response to a light step of intensity I, and I1/2 is the intensity that suppresses half the light-sensitive current. |
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). We took advantage of this histological feature to enhance the probability of drawing a cone "inner segment" into the suction pipette (Fig. 1, A–C). Once the suction pipette was positioned nearby the outer layer of the ONL, the experimenter drew nuclei into the pipette, one at a time. Background steps of 500-nm light were applied to suppress the rod current, and superimposed flashes delivered to test for the presence of current arising from a cone (Fig. 1 D). Additional nuclei were drawn in until a flash response was recorded in the presence of a strong background. Several nuclei (up to 10) were drawn into the pipette to obtain such a response. To determine the background intensity used to suppress rod currents, we measured the step response amplitude vs. intensity relation for the sensitive component of current (Fig. 1 E, colored symbols), and compared this with the step response functions of individual rods (Fig. 1 E, gray symbols). The data can be described by a hyperbolic saturation function (Eq. 6), with parameter I1/2, the intensity required to suppress 50% of the circulating current; for these experiments I1/2 350 photons µm–2 s–1. Given a rod light collecting area ac = 0.5 µm2, the corresponding half-saturating photoisomerization rate, 175 s–1, is consistent with previous determinations: in rat rods, 161 s–1 (Nakatani et al., 1991); in mouse rods, 250 s–1 (Lyubarsky and Pugh, 1996) and 120 s–1 (Xu et al., 1997). Based on Eq. 6, and the value I1/2 350 photons µm–2s–1, a background of 20,000 photons µm–2s–1 is predicted to suppress 98.3% of the circulating current of individual rods. As the typical current recorded by the suction pipette in our experiments was 40 pA, it follows that the rod current not suppressed by the background should have been <0.7 pA. Nonetheless, in the presence of this background, the suction electrode routinely recorded a residual current of 5–15 pA, which responded only weakly to modest increases in steady background, but which could be suppressed by strong flashes (Fig. 1 D). We concluded that this residual current originated in a single cone cell based on its properties, which we now describe.
|
Six Lines of Evidence Establish that the Responses Are Generated by Mouse Cones
Six distinct lines of evidence can be presented at this point in support the conclusion that the photoresponses recorded in the presence of the standard background were generated by WT mouse cones. First, as mentioned above, recordings from rods and calculations with Eq. 6 both support the view that rod responses are completely suppressed by the background. Second, the time to peak of the dim flash response was typically 70 ms, more than twofold briefer than the time to peak of any reported (even light adapted) mammalian rod response (Fig. 2, A and D; Table I). Third, the so-called dominant recovery time constant ("Pepperberg" constant), estimated from responses in the "just saturating" flash intensity regime, was typically 70 ms, almost threefold shorter than that (200 ms) typically reported for WT rods (Fig. 2, C and F; Table I). Fourth, the absolute sensitivity of the light responses to 361-nm flashes from WT retinal slices recorded in the presence of the background is comparable to that for the cells recorded in the absence of the background for cells from retinal slices of Gt–/– mice, and >40-fold lower than that of rods (Table I). Rods of Gt–/– mice are structurally normal, but do not respond electrically to light (Calvert et al., 2000). (The effect of the background on the cone responses is considered further below.) Fifth, other properties of the responses obtained from retinal slices of Gt–/– mice (Fig. 2, G–L; Table I), including the amplification constant and dominant recovery time constant, are very similar to those of responses obtained from slices of WT retina in the presence of the background. Sixth, the spectral sensitivity of the responses was typically maximal at 360 nm, the max of mouse cone S-opsin (Fig. 3 A). (An exception is that responses recorded from slices taken from the most dorsal retina were maximally sensitive at 510 nm, the max of mouse cone M-opsin, as described below). We conclude that the responses obtained from WT retinal slices in the presence of the standard background indeed arise from cones and proceed to their further characterization.
|
|
|
4 pA was set by the need for adequate dynamic range to measure a dim-flash response, with peak amplitude of 
20% of rmax.) Because the spatial distribution of the "inner segment" limb of the circulating current and the total area of the cone membrane drawn into the pipette are unknown, the only rigorous conclusion that can be drawn is that the total circulating current of a WT mouse cone must be at least 7 pA. However, the total cone circulating current is likely to be at least 15 pA (the largest recorded value), and could be considerably greater. Comparison of the measured rmax of WT cones with that obtained from cones of the Nrl–/– mouse (Table I), whose outer segments are 40% shorter, suggests that drawing several rod nuclei in the pipette may act to lower the current recording efficiency.
The Amplification of WT Mouse Cone Photoresponses
Amplification is a critical feature of the activation phase of the vertebrate photoreceptor light response (Lamb and Pugh, 1992; Pugh and Lamb, 1993), and the results show the responses of mouse cones to be highly amplified (Fig. 2, B, E, H, and K; Table I). Nonetheless, the analyses reveal that the cone amplification constants are two- to fourfold lower than that of rods recorded under the same conditions, and that the amplification of M-cones is somewhat lower than that of S-cones.
S- and M-opsins Are Coexpressed and Functional in WT Mouse Cone Photoreceptors
Histochemical evidence has revealed that most cones of C57Bl/6 ("WT") mice coexpress both S- and M-opsins, with the M-opsin expression varying in a dorso-ventral gradient (Applebury et al., 2000), and ERG b-wave evidence consistent with this conclusion has been published (Lyubarsky et al., 1999). Recently, we established such coexpression to occur in the all-cone retina of the Nrl–/– mouse and established with single-cell recordings that the coexpressed opsins are both functional, i.e., both capable of activating phototransduction in the same cone (Nikonov et al., 2005). In the current investigation, we confirmed these basic features in our recordings from single cones of the WT mouse but found a quantitative difference in the degree of expression of the M-cone opsin by the Nrl–/– and WT retinas.
Most WT mouse cones were found to be maximally sensitive at 360 nm, indicating that the majority of their opsin is S-cone opsin (Fig. 3 A). Nonetheless, 19/20 such cells exhibited a secondary mode of sensitivity at 510 nm, establishing that they functionally coexpress M-cone opsin. Such coexpression was found even in cones in slices of the most ventral portion of the retina, with one exception (Fig. 3, purple symbol identified by arrow). By recording from slices from the most dorsal part of the retina (MATERIALS AND METHODS), we found a subset of cones in which M-opsin was expressed to a higher degree than S-opsin (Fig. 3 A). By routinely measuring the sensitivity of all cones at both 361 and 501 nm, we obtained for each cone a spectral sensitivity ratio, S501/S361; cones for which this ratio exceeds unity will be classified as "M-cones," while those for which the ratio is less than unity will be designated "S-cones." The classification ratio varied systematically with the dorso-ventral position (Fig. 3). Because of the so-called ß-band of absorption, which is maximal in the near UV for opsins with max near 500 nm, and whose absorbance is 20% that of the primary -band (Govardovskii et al., 2000), it was not readily determinable whether all M-cones coexpress S-opsin. Thus, a few of the M-cones for which S501/S361 was >5 could be pure "M-opsin cones."
Phototransduction Activated by the S- and M-cone Opsins in Individual WT Cones Is Very Similar
As both S- and M-cone opsins are expressed in individual WT cones and activate phototransduction in the same cell, it is of interest to ascertain whether the kinetics of the light responses driven by the two cone opsins are the same. Because of the wide separation in the max's of the two cone opsins, it is possible to unequivocally stimulate either the S- or the M-pigment in the S-cones (Fig. 3 A). While it was not possible, due to the problem of the absorption ß-band (mentioned above), to unequivocally stimulate either opsin in all M-cones, in many cases it was. The results provide an unequivocal answer: in individual WT cones, the dim-flash responses driven by S- and M-cone opsins are effectively indistinguishable (Fig. 4).
|
–/– cones (tpeak = 90–100 ms) (Table I). (It bears mention that the 8-pole analogue filter used in the experiments produces a measured delay of 25 ms, and the values of tpeak in Table I have not been corrected for this delay.) The question arises, however, whether these differences are intrinsic to the cone types or whether they may arise as a consequence of the use of a rod-suppressing background.
Most WT Cones Are Adapted by the Rod-suppressing Background due to M-opsin (Co)expression
We examined the hypothesis that the standard rod-suppressing 500-nm background, combined with the degree of expression of M-opsin in a given WT cone, might underlie the differences in dim-flash kinetics. Two qualitative predictions of the hypothesis can be readily made: (1) tpeak should be a decreasing function of the degree of M-opsin coexpression by S-cones, and (2) flash sensitivity should be a decreasing function of the degree of M-ospin coexpression. Two additional predictions can be made on the hypothesis that the properties and recordings of Gt–/– cones differ from those of WT only in that no background was needed to isolate their responses: (3) tpeak of the cones of Gt–/– mice should not depend on the degree of expression of M-opsin, and (4) tpeak of WT S-cones should approach tpeak of Gt–/– S-cones for low M-opsin coexpression. These predictions are reasonably well confirmed (Fig. 5 A).
|
of expression of M-opsin to S-opsin (the "M-opsin co-expression ratio") corresponds closely to the sensitivity ratio, i.e., = S501/S361. Using the conservation assumption (b), one finds the fraction of M-opsin in a cone satisfies fMOps = /( + 1), while the fraction of S-opsin is fSOps = 1 – fMOps. And so one arrives at the quantitative relations
 | (7A) |
 | (7B) |
measured with a flash of wavelength (=501 or 361 nm) in the presence of the standard background of intensity I500 (=20,000 photons µm–2), aC = 0.2 µm2 is the collecting area of a cone expressing only one opsin, Idark* is the "dark light" expressed in an equivalent isomerization rate, and SF,dark is the dark-adapted sensitivity of such a cone in the absence of the background. The theoretical curve generated by Eq. 7 is plotted as the dark gray line in Fig. 5 B; for < 1, it exhibits a familiar Weber Law dependence on . Again using hypothesis (c) that cones of Gt–/– mice differ only from those of WT in that no background was needed to isolate them, one predicts that the sensitivity of WT S-cones that coexpress low levels of M-opsin should asymptote (
0) to the sensitivity of Gt–/– cones; this is indeed observed. In addition, as expected, all the WT M-cone sensitivities, unlike some of the WT S-cone sensitivities, lie reliably below those of the Gt–/– mice. The principal discrepancy between prediction and observations is that the sensitivities of the WT M-cones lie systematically above the theory line (Fig. 5 B, green symbols). Possible reasons for the discrepancy include a violation of one or more of the assumptions (a–c) underlying the derivation. Future physiological experiments will address the issue of whether the cones obey Weber's Law, and whether the S- and M-opsins are inactivated with the same kinetics by Grk1. Biochemical experiments will be required to assess potential differences in the S- and M-opsins in activating Gnat2.
In summary, then, we conclude that the degree of coexpression of M-opsin in WT mouse cones leads to varying degrees of desensitization and speeding of the dim-flash response kinetics (shortening of tpeak) by the rod-suppressing background, and conclude further that the results are generally consistent with the notion that the properties of fully dark-adapted WT cones can be inferred from those of Gt–/– cones, whose isolation does not require the use of backgrounds.
Step Responses of S- and M-cones
Cones differ from rods in their responsivity to steady light (INTRODUCTION), and so we undertook experiments to determine how mouse cones responded to light steps of varied intensity (Fig. 6). Both S- and M-cones reached steady state in 100 ms and recovered their full circulating current from even the most intense steps used in 1 s when the step was extinguished (The recovery from the steps was determined in experiments with Gt–/– retina, in which there was no rod current suppression [Fig. 6, E and G].) The step response vs. intensity relations were characterized with hyperbolic saturation relations (Eq. 6); population averages of the half-saturating intensities are provided in Table I. Taken at face value, the data suggest that the cones of Gt–/– retinas were more sensitive to light steps, a surprising result given that the cones of Gt–/– mice appear to have somewhat lower amplification. One possible explanation for the discrepancy is that the sensitivity of the WT cones was lowered by bleaching by the rod-suppressing backgrounds, in effect lowering their collecting area. We suspected that substantial fractions of the M-cone pigment were bleached during the course of the step experiments, as the rod currents in records from WT retinal slices were persistently suppressed. We thus developed a rigorous approach to calculating the level of bleached pigment at any time in the experiment, applying Eq. 3 to the sequence of stimulations. These analyses confirm that at least part of the discrepancy in step responsivity between Gt–/– and WT cones was due to cone pigment bleaching; thus, the corrected data and saturation curves (Fig. 6, gray symbols and traces) bring the WT cone results into closer agreement in regards with those of Gt–/– cones, reducing the estimates of I1/2 in WT cones by two- to threefold. Even with a blanket threefold adjustment for bleaching, however, the WT cones appear to be less sensitive to steps than Gt–/– cones, suggesting that WT cones may possess some capacity for light adaptation that is attenuated in the Gt–/– retina.
|
30 min total). Cone circulating current was invariably found to reach steady state at the termination of the bleaching exposure within the few seconds needed to make a reliable measurement. In contrast, for rods it was necessary to wait many minutes after bleaching for achievement of a steady-state recovery of the circulating current; 15 min were required for the lower bleaching levels and up to 30 min for the largest bleaching exposures. For both rods and cones, once reached, the steady state was maintained for 10 min or more (at which time the experiments were usually terminated).
|
max's of these two pigments. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10,000 mm–2) than the peripheral retina of primates (3,000–5,000 mm–2) (Carter- Dawson and LaVail, 1979; Jeon et al., 1998; Ghosh et al., 2004). Moreover, like primates, mice have 10 highly conserved types of bipolar cells, and 9 of these (constituting approximately one half the total bipolar cell population and including four OFF-bipolars and five ON-bipolars) make exclusive contact with cones (Ghosh et al., 2004). The robust mouse cone-driven ERG b-wave, whose amplitude is approximately one third of the total rod- and cone-driven b-wave of a dark adapted mouse, is consistent with the histological evidence (Lyubarsky et al., 1999). The b-wave is generally held to be the field potential arising from light-dependent inactivation of mGluR6 metabotropic receptors in the dendritic terminals of ON-bipolar cells (Robson et al., 2004). Nonetheless for the importance of cone signaling in the mouse retina, perhaps because of the 30-fold higher numerosity of rods, or because of the fragility of cone outer segments separated from their extracellular matrix sheaths (Nikonov et al., 2005), mouse cones have been an elusive target for electrophysiological characterization at the single-cell level.
Using a retinal slice preparation and a novel method of recording in which several outer nuclear layer nuclei and adjacent tissue is drawn into a suction pipette (Fig. 1), we have now recorded and characterized for the first time the electrical responses of single WT mouse cones. At least six distinct lines of evidence and arguments support this contention: the overall thrust is that the photoresponses recorded in the presence of a 500-nm background of 20,000 photons µm–2s–1 have properties that are (a) highly distinct from the corresponding properties of mouse rods, (b) specifically predicted for mouse cones (e.g., peak sensitivity in the near UV), or (c) generally expected of mammalian cones, but not rods.
S-cones, M-cones, and Functional Coexpression of S- and M-cone Pigments in WT Mouse Cones
Mouse cones can be classified as "S-cones" and "M-cones" based on which of the two mouse cone opsins drives phototransduction with a higher sensitivity (Fig. 3 B). Our experiments, as expected from previous immunohistochemical (Rohlich et al., 1994; Applebury et al., 2000; Lukats et al., 2002) and electroretinographic (Calderone and Jacobs, 1995; Jacobs et al., 1991; Lyubarsky et al., 1999) investigations, establish that most WT mouse cones are maximally sensitive at 360 nm, and thus are classified as S-cones. Our experiments also show that WT, unlike Nrl–/– mice (Nikonov et al., 2005), have M-cones in their dorsal retina (Fig. 3 B), as expected from immunohistochemistry (Applebury et al., 2000). While the physiological evidence is still incomplete, we nonetheless suspect that a very high percentage of mouse cones coexpress both S- and the M-cone opsins, in part based on previous electroretinographic results (Lyubarsky et al., 1999). Nearly universal coexpression of M-opsin is clear for the S-cones, for which the action spectrum analysis can detect M-opsin coexpression to as low as 1 part in 10,000 (Fig. 3 A): only one S-cone in >30 recorded to date follows the template for a 360-nm opsin at 500 nm (Fig. 3). For M-cones, universal coexpression of S-opsin is not certain, as the ß-band of the M-opsin prohibits detection in dark-adapted cones of S-opsin coexpressed at a ratio less than approximately one fourth (Fig. 3). Future experiments in which systematic bleaching of M-opsin is employed should allow definitive determination of the sensitivity of the M-opsin ß-band, and detection of S-opsin in "M-cones" coexpressed <1%.
The likely nearly universal coexpression of the two cone opsins and the dorso-ventral gradient of M-opsin coexpression to some extent render a two-category classification of mouse cones moot and misleading, and it may be more useful in many contexts to speak of "S-dominant" and "M-dominant" cones. For example, experimenters not equipped to stimulate the mouse retina with UV light might draw erroneous conclusions about the sensitivity and adaptational properties of cone-driven inner retinal neurons. Indeed, M-opsin coexpression may lead physiologists characterizing the light responses of cone-driven bipolars and other post-receptor cells in the mouse (e.g., Berntson and Taylor, 2000) to misestimate their true sensitivity, which would only be seen with UV stimulation
Physiological Features of WT Mouse Cones
Absolute Sensitivity.
The effects of the coexpression of M-opsin and the requirement for a rod-suppressing background lead to an underestimation of the absolute sensitivity of both WT S- and M-cones, but all the data of WT and Gt–/– cones become mutually consistent when these effects are taken into consideration (Fig. 5); 0.2% of the WT mouse cone circulating current is suppressed per photoisomerization at the peak of the dim-flash response. For rods recorded under the same condition, absolute sensitivity is 5% (these values can be obtained from the sensitivities in Table I by converting the flash intensities into photoisomerizations by multiplying by the collecting areas, aC = 0.2 µm2 [cones] and aC = 0.5 µm2 [rods]). Closely comparable values have been reported for rods and cones of primates (Baylor et al., 1984; Schnapf et al., 1990).
Dim-flash Kinetics.
Dim-flash responses driven by the two cone pigments in individual WT mouse cones are effectively indistinguishable, whether the cone is classified as an S-cone or M-cone (Fig. 4). This result provides support for the hypothesis that the phototransduction cascades activated by the S- and M-cone opsins are (other than the photopigments) identical. Support for this conclusion also comes from genomic evidence that there is only one additional G-protein -subunit, Gnat2, that is highly homologous to rod transducin (Gt) (Wilkie et al., 1993), combined with histochemical evidence that this protein is universally expressed in vertebrate cones (Lerea et al., 1986, 1989; Ying et al., 1998).
Amplification.
For the first time, we are able to compare the amplification of the S- and M-cone pigments: in both WT and Gt–/– M-cones, the amplification constant A appears to be reduced by 20–30% relative to that of S-cones (Table I). But as this apparent reduction is small, in the context of the very similar absolute sensitivities and dim-flash kinetics of the two cone types, it seems likely that fully active S- and M-cone opsins activate Gnat2 (cone transducin) at approximately the same rate. The amplification constants of the cones are, however, very reliably below that of rods, by a factor of two- to threefold (Table I). One of the factors (ßsub, the rate constant of a single phosphodiesterase (PDE) catalytic subunit in the outer segment) that multiply to produce the amplification constant is inversely proportional to the outer segment volume (Lamb and Pugh, 1992; Pugh and Lamb, 1993), and the cone outer segment volume is only 40% that of the rods (Table I). On the assumption that the catalytic efficiency kcat/Km of the cone PDE is the same as that of the rods (Gillespie and Beavo, 1988), it follows that PDE catalytic subunits in mouse cones are activated per fully active cone "R*" at a rate less than one fifth that at which rod PDE catalytic subunits are activated per fully active rhodopsin, R*. Alternatively, kcat/Km for mouse cone PDE may be lower than that of rod PDE, and if so, this would contribute to the lower overall amplification.
Opsin Inactivation.
Grk1 is the only GPCR kinase in the mouse kinome that is expressed in mouse photoreceptors (Weiss et al., 2001; Chen et al., 2001; Caenepeel et al., 2004), and it is now clear that Grk1 is necessary for normal inactivation of both mouse cone opsins (Lyubarsky et al., 2000; Nikonov et al., 2005). Given the kinetics of the responses driven by S- and M-opsins (whether these occur in S- or M-dominant cones) (Fig. 4), it follows that Grk1 largely inactivates both cone opsins in well less than 50 ms, the inflection point in the activation phase of the dim-flash response, at which the response begins to "peel downward" from the pure activation theory. The almost twofold faster inactivation of mouse cone opsins than rhodopsin by the same kinase, Grk1, argues that this kinase is more effective in mouse cones than in mouse rods, either due to its specific affinity for the S- and M-opsin, or due to its level of expression.
Dominant Recovery Time Constant.
Compelling evidence has recently been presented that the dominant recovery time constant of mouse rods arises from the time constant of the GTPase activity set by the interacting complex of the transducin -subunit (Gt), the phosphodiesterase -subunit (PDE), and regulator of G-protein signaling, Rgs9-1, coupled to its anchor protein R9ap (Krispel, C.M., C.K. Chen, D. Chen, Y.J. Chen, N. Calero, and M.E. Burns. 2005. Invest. Ophthalmol. Vis. Sci. 46:4628). It is generally accepted that the dominant recovery time constant, D, originates in the inactivation of one or the other of the two principal amplifiers of the phototransduction cascade, the photoactivated pigment R* (in which case, D = R) or the activated PDE complex, Gt-PDE (in which case, D = E) (compare Nikonov et al., 1998, 2000). From the observation that D 
70 ms in mouse cones (Fig. 2; Table I), it can be concluded that E is considerably shorter in mouse cones than in mouse rods, likely sped up by a higher level of expression of RGS9-1 in cones as opposed to rods (Lyubarsky et al., 2001; Zhang et al., 2003). Our results, however, do not speak to the issue of which of the two cascade amplifier inactivation steps is dominant in mouse cones, but only imply that both R and E are shorter than the slower step in mouse rods.
The "Nose" on the Photocurrent.
The saturating photoresponses of WT and Gt–/– cones exhibit a "nose," a current that has the same sign as the photocurrent, and which undergoes a rapid decay to a plateau with a time constant <50 ms (Figs. 2 and 6). The photocurrent, recorded from the inner segment plasma membrane in the configuration employed in the experiments reported here, is largely due to a decline in a K+ current (IKx) that tracks the light-activated decline of the cGMP-activated current of the outer segment (Frings et al., 1998). A likely explanation of the "nose" is that it is due to the inwardly rectified, hyperpolarization-activated current, Ih (Hestrin, 1987). Strong flashes that completely close the cGMP-activated channels should hyperpolarize the cone below the reversal potential of Ih and trigger a depolarizing current, as observed. (For simplicity in presentation, we have shown all light responses as positive-going deflections. However, in the "inner segment in" recording configuration used here, the responses are the suppression of an outward membrane current.) This same current is clearly present and of comparable magnitude in responses of Nrl–/– cones as in WT cones, but less prevalent in rods (Nikonov et al., 2005). A practical consequence of this current is that it makes it more difficult to define the saturating level of the photocurrent (Figs. 2 and 6), impacting on the precision of estimating various parameters characterizing the cells.
WT Mouse Cones Can Function "Normally" with Large Fractions of their Opsin Bleached
Differences amongst the three types of mouse opsin were revealed in the apparent degree to which bleached rhodopsin, S-opsin, and M-opsin causes persistent suppression of the circulating current (Fig. 7). A "dark light" generated by bleached rhodopsin has been well documented in recordings from amphibian (Lamb, 1981; Cornwall and Fain, 1994; Matthews et al., 1996), and primate (Baylor et al., 1984) rods, and also in salamander (Cornwall et al., 1995) and primate (Schnapf et al., 1990) cones. As revealed in recent experiments with the human cone ERG a-wave (Kenkre et al., 2005), however, mammalian cone transduction is vastly less activated by cone opsin dark light than is rod transduction by bleached, unregenerated rhodopsin. Our results suggest that different classes of cone opsin may differ in regards to their generation of dark light, with bleached S-opsin acting somewhat more like bleached rhodopsin, and bleached M-opsin having almost negligible effect, as expected from its close homology with human M- and L-cone pigments (Kenkre et al., 2005). The coexpression of the two opsins in mouse cones may prove useful in the investigation of these differences, and we are actively pursuing the problem.
WT, Gt–/–, and Nrl–/– cones: Preparations for the Investigation of Mammalian Cone Function
In two recent reports, we presented a body of results, including electron microscopy, immunohistochemistry, quantification of key cone transduction proteins, ERG a-wave analysis, and single-cell recordings, that together establishes that the photoreceptors of mice lacking the neural leucine zipper transcription factor (Nrl–/–) are cones (Daniele et al., 2005; Nikonov et al., 2005). The results presented here, which characterize the physiological features of WT mouse cones, show that Nrl–/– photoreceptors are in fact practically indistinguishable from WT cones in their physiological features (Table I). These results thus close that case on the identification of the Nrl–/– retina as an "all-cone" retina and further strengthen the case for its use in the investigation of cone-specific genes and their function. Nonetheless, several exceptional features call for caution.
The first exception is that Nrl–/– cone outer segments are on average only 60% the length of WT cone outer segments, and that they exhibit a degree of disorder of their disc stacking and overall orientation (Daniele et al., 2005). The second exception is that Nrl–/– cone outer segments undergo a slow degeneration that results in the halving of the circulating current by 6 wk of age (Daniele et al., 2005). The third exception is that Nrl–/– cones do not express M-cone opsin to the same degree as WT cones. This latter point is supported by the observations presented here (Fig. 3 B), and also by comparison of the action spectrum of the cone-driven b-wave of the WT mouse (Lyubarsky et al., 1999) with that of the a-wave of the Nrl–/– mouse (Daniele et al., 2005); the secondary mode (at 510 nm) of the action spectrum of the WT mouse cone system is approximately one fourth the sensitivity of the primary mode (at 360 nm), but in the Nrl–/– mouse, the secondary mode falls to one eighth or less. Nonetheless, for these exceptions, the definitive conclusion that photoreceptors in the Nrl–/– mouse in the first 6 wk of life are healthy cones opens the door to many investigations of importance to the understanding of function of genes expressed specifically in cones.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by NIH-EY-02660 (E.N. Pugh Jr.) and EY-12008 (J. Lem). E. N. Pugh Jr. is supported by a Jules & Doris Stein Research to Prevent Blindness Professorship.
Olaf S. Andersen served as editor.
Submitted: 10 January 2006
Accepted: 10 March 2006
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-subunit. Proc. Natl. Acad. Sci. USA. 97:13913–13918 (published erratum appears in Proc. Natl. Acad. Sci. USA. 2000. 98:10515).-Transducin is present in blue-, green-, and red-sensitive cone photoreceptors in the human retina. Neuron. 3:367–376.[CrossRef][Medline] subunits in retinal rod and cone photoreceptors. Science. 234:77–80.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
Related Article
This article has been cited by other articles:
|
|
 |
|
  M. M. Abd-El-Barr, M. E. Pennesi, S. M. Saszik, A. J. Barrow, J. Lem, D. E. Bramblett, D. L. Paul, L. J. Frishman, and S. M. Wu Genetic Dissection of Rod and Cone Pathways in the Dark-Adapted Mouse Retina J Neurophysiol, September 1, 2009; 102(3): 1945 - 1955. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Yin, R. G. Smith, P. Sterling, and D. H. Brainard Physiology and Morphology of Color-Opponent Ganglion Cells in a Retina Expressing a Dual Gradient of S and M Opsins J. Neurosci., March 4, 2009; 29(9): 2706 - 2724. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Umino, E. Solessio, and R. B. Barlow Speed, Spatial, and Temporal Tuning of Rod and Cone Vision in Mouse J. Neurosci., January 2, 2008; 28(1): 189 - 198. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Shi, K.-W. Yau, J. Chen, and V. J. Kefalov Signaling Properties of a Short-Wave Cone Visual Pigment and Its Role in Phototransduction J. Neurosci., September 19, 2007; 27(38): 10084 - 10093. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Cornwall and P. Ala-Laurila A Perfect Marriage: Molecular Genetics Ties the Knot with Electrophysiology in Studies of Visual Transduction J. Gen. Physiol., July 1, 2007; 130(1): 7 - 10. [Full Text] [PDF]
|
|
|
|
|
|
K. Sakurai, A. Onishi, H. Imai, O. Chisaka, Y. Ueda, J. Usukura, K. Nakatani, and Y. Shichida Physiological Properties of Rod Photoreceptor Cells in Green-sensitive Cone Pigment Knock-in Mice J. Gen. Physiol., July 1, 2007; 130(1): 21 - 40. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chan, W. W. Rubin, A. Mendez, X. Liu, X. Song, S. M. Hanson, C. M. Craft, V. V. Gurevich, M. E. Burns, and J. Chen Functional Comparisons of Visual Arrestins in Rod Photoreceptors of Transgenic Mice Invest. Ophthalmol. Vis. Sci., May 1, 2007; 48(5): 1968 - 1975. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D.-Q. Zhang, T.-R. Zhou, and D. G. McMahon Functional Heterogeneity of Retinal Dopaminergic Neurons Underlying Their Multiple Roles in Vision J. Neurosci., January 17, 2007; 27(3): 692 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Yin, R. G Smith, P. Sterling, and D. H. Brainard Chromatic Properties of Horizontal and Ganglion Cell Responses Follow a Dual Gradient in Cone Opsin Expression. J. Neurosci., November 22, 2006; 26(47): 12351 - 12361. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. P. Sakmar Timing Is Everything: Direct Measurement of Retinol Production in Cones and Rods J. Gen. Physiol., July 31, 2006; 128(2): 147 - 148. [Full Text] [PDF]
|
|
|
|
 |
|
 Highlights From The Literature Physiology, June 1, 2006; 21(3): 157 - 161. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
tpeak[
], with tpeak,dark = 96 ms, 
