All experiments were conducted in accordance with rules and regulations of the National Institutes of Health guidelines for research animals, as approved by the Institutional Animal Care and Use Committee (IACUC) of the University of California, Los Angeles (Los Angeles, CA). Mice were kept under cyclic 12 h of light/12 h of darkness with ad libitum food and water in approved cages. Mice were dark adapted overnight before being euthanized by cervical dislocation under infrared light. Males and females between the ages of 5 to 15 weeks were used in approximately equal numbers. The E122Q-rhodopsin+/+ line was originally generated by Dr. Yoshinori Shichida (Kyoto University) and was generously provided to us by Dr. King-Wai Yau (Johns Hopkins School of Medicine). These mice were mated with Gnat2−/− mice lacking the gene for cone transducin (Ronning et al. 2018), provided to us by Marie Burns of the University of California Davis School of Medicine. Hereafter, the E122Q-rhodopsin+/+; Gnat2−/− mice will be referred to as E122Q and Gnat2−/− mice will be referred to as WT.

The procedure used to prepare retinal tissue for light microscopy have been previously described in Frederiksen et al (2022). In brief, mice were euthanized and eyes were enucleated. A small hole was made in the cornea with a No. 11 scalpel and the eyes were placed in fixative solution containing 2% formaldehyde and 2.5% glutaraldehyde in 100 mM cacodylate buffer. After 60 min, the cornea and lens were removed, and eye cups were fixed overnight at 4 °C. The following day the eyecups were rinsed 3 times (10 min each) in 100 mM cacodylate buffer. The tissue was post-fixed with 1% osmium tetroxide, dehydrated in a graded series of ethanol (50%, 60%, 70%, 80%, 90%, 95%, and 100%), and embedded in Epon resin at 60 °C overnight. Semi-thin sections (1 µm) were cut and stained with 0.5% Toluidine Blue in 1% sodium borate. The light micrographs were photographed with a Zeiss Axiophot microscope equipped with a 20X objective and a CoolSNAP digital camera.

Mouse retinae in MSP, ERG, and patch-clamp recordings were dissected and superfused with Ames’ medium (Ames and Nesbett 1981) at 2 mL/min, which was buffered with 1.9 g/L sodium bicarbonate and bubbled with 95% O2/5% CO2 to maintain a pH of between 7.3 to 7.4 and an osmolarity of 284 ± 1 mOsm. Slices for patch-clamp recording were cut in HEPES-buffered Ames’ medium containing 2.38 g/L HEPES and 0.875 g/L NaCl to give an osmolarity of 284 ± 1 mOsm at pH 7.35 ± 0.05. The internal solution used in patch-clamp recording contained the following (in mM): 125 K-aspartate, 10 KCl, 10 HEPES, 5 N-methyl-glucamine-HEDTA, 0.5 CaCl2, 0.5 MgCl2, 1 ATP-Mg, 0.2 GTP-Tris, and 2.5 NADPH. The pH was adjusted to 7.3 with N-methyl-glucamine hydroxide, and the osmolarity was adjusted to 278 mOsm. In all experiments the temperature was held at 35–37 °C with a temperature controller (Warner Instruments, Camden, CT).

Optical density from intact mouse rods was measured as previously described (Nymark et al. 2012; Frederiksen et al. 2016). Dark-adapted mouse retinae were peeled from the retinal pigment epithelium and cut in half under infrared light. One retinal half was placed on a quartz coverslip and held down by a slice anchor (Warner Instruments, Hamden, CT, USA) before superfusion with Ames’ medium at 2 mL/min. Measurements were made from 6 to 8 rods en face as previously described (Nymark et al. 2012). Data were recorded in LabVIEW 2018 (National Instruments, Austin, TX, USA) and processed with custom scripts in MATLAB (MathWorks, Natick, MA, USA).

The photobleaching products meta II and meta III were measured in the following way. First, a dark rhodopsin spectrum was recorded. This spectrum was baseline corrected by subtracting the average of the data between 600 and 620 nm. Secondly, the retina was bleached by a 15 s step of 505-nm light (5.9 × 108 photons μm−2 s−1) calculated to bleach > 90% of the visual pigment. Absorbance spectra were recorded immediately after the bleach (t = 0 min), and then at 20-s intervals for 10 min, at 2-min intervals for a further 20 min, and at 5-min intervals for the final 30 min of recording. Optical density was calculated with Beer’s law as described previously (Nymark et al. 2012). Each of these spectra was baseline corrected and fitted with a combination template, S(λ), for meta II and meta III constructed from Gaussian functions:

$$S\left(\lambda \right)=MetaII\left(\lambda \right)+MetaIII\left(\lambda \right),$$

(1)

where:

$$MetaII\left(\lambda \right)=__^\right)}^}$$

(2)

and:

$$MetaIII\left(\lambda \right)=__\left[^\right)}^}+0.1^\right)}^}\right].$$

(3)

AII is the peak amplitude of meta II, normalized to rhodopsin, at 384 nm; εII is the extinction coefficient of meta II relative to rhodopsin; AIII is the peak amplitude of meta III, normalized to rhodopsin, at 472 nm; εIII is the extinction coefficient of meta III relative to rhodopsin; and λ is wavelength. For the extinction coefficients we used the values reported in Kolesnikov et al. (2003) of εII = 1.16, and εIII = 1.0.

Results from the production and decay of meta II and meta III were fitted with a model based on similar scheme as in Kolesnikov et al. (2003), see Fig. 2a. Rhodopsin with photosensitivity P = 5.7 × 10−9 µm2 (Woodruff et al. 2004) is bleached by the 505-nm light ϕ(t). This results in a sequence of photoproducts (bathorhodopsin, lumirhodopsin, meta I), all of which are fast, so we ignore them and assume that meta II is the first state. Meta II in turn is in equilibrium with meta III with the rate constants k1 and k−1. Meta II decays to free opsin and all-trans retinal with the rate constant k2, and meta III decays to free opsin and all-trans retinal with a rate constant k3. Assuming first order reactions, the scheme in Fig. 2 yields four equations describing the kinetics of the photoproducts:

$$\frac=-\left[Rho\right]\phi \left(t\right)P$$

(4a)

$$\frac=\left[Rho\right]\phi \left(t\right)P+\left[MIII\right]_-\left[MII\right]_-\left[MII\right]_$$

(4b)

$$\frac=\left[MII\right]_-\left[MIII\right]_-\left[MIII\right]_$$

(4c)

$$\frac=\left[MII\right]_+\left[MIII\right]_$$

(4d)

where the stimulus ϕ(t) is a step of light of amplitude φ and the duration toff—ton:

$$\phi \left(t\right)=\left\\varphi \\ 0\end\begin \, \, }}} \, _\le t\le _\\ \, \, }}}}}}}}}\end.\right.$$

(5)

Equations 4a–d were solved numerically using the Python package odeint from scipy.integrate with the initial parameters 1 for rhodopsin (fractional bleach) and 0 for meta II, meta III and opsin.

Dissections were performed under near infrared illumination using infrared image converters, either head-mounted (ITT Industries, Roanoke, VA) or dissecting-microscope mounted (B.E. Meyers, Redmond, WA). Following euthanasia, eyes from mice were enucleated, the anterior portion of the eye was cut, and the lens and cornea were removed. Eyecups were stored in a light-tight container in bicarbonate buffered Ames’ medium, which was equilibrated with 95% O2 / 5% CO2 as described in the Solutions section. Eyecups were bisected through the optic nerve head with a number 10 scalpel under an infrared-equipped dissection microscope (Carl Zeiss), and the retina was carefully removed from the retinal pigment epithelium with fine forceps. The isolated piece of retina was embedded in a low-temperature gelling agarose (3%; Sigma-Aldrich) in HEPES-buffered Ames' medium. Vertical retinal slices (200 μm) were cut in chilled, oxygenated Ames-HEPES with a vibrating microtome (VT-1000 S, Leica, Deerfield, IL) and transferred either to a recording chamber or the storage container for use later in the experiment. During recordings, the retinal slice was stabilized with a slice anchor. The slice was superfused with Ames' medium at ∼2 ml min−1. The bath temperature was held at 35 ± 2 °C by a temperature controller with feedback (TC-324B, Warner Instruments, Hamden, CT). Recordings from individual cells were made by whole-cell patch clamp from dark-adapted retinal slices as described previously (Arman and Sampath 2010; Frederiksen et al. 2025). Cells were visualized with illumination from an infrared LED (840 nm), attached to the transmitted light path of the physiology microscope (Slicescope 1000, Scientifica, Uckfield, UK). Rod somata were identified by morphology and location in the outer nuclear layer (ONL).

Filamented borosilicate-glass capillaries (BF120-69–10; Sutter Instruments, Novato, CA) were pulled on the day of the experiment with a P-97 Flaming/Brown micropipette puller (Sutter) to a tip resistance in the bath medium of 16–18 MΩ. Rods were voltage-clamped at -50 mV with a Multiclamp 700B patch-clamp amplifier (Molecular Devices). The amplifier was also used to record membrane resting potentials and voltage light responses from unclamped rods (“current clamp”). For voltage clamp with whole-cell recording, series resistance of the recording pipette was compensated at 75–80% to prevent error in clamping potentials, and pipette capacitance was neutralized before break-in. The seal and access resistance was monitored continuously throughout the recording. Recordings were terminated if the seal resistance was below ∼1 GΩ, or if there was a sudden change in access resistance. Resting membrane potentials were determined from the mean voltage value taken over 3 repetitions of a 3-s measurement in current clamp. All reported values of membrane potential have been corrected for liquid-junction potentials (Neher 1992), which were measured to be about 10 mV for our recording solutions (Ingram et al. 2019). Recordings were low-pass filtered at 3 kHz by the patch-clamp amplifier and digitized at 10 kHz with a 16-bit A/D converter (NI USB-6363, National Instruments, Austin, TX). Data were collected in MATLAB (R2018b, MathWorks, Natick, MA) with the open-source software package Symphony Data Acquisition System (https://symphony-das.github.io). All offline data visualization and analysis were performed with the Iris DVA framework for MATLAB (https://github.com/sampath-lab-ucla/IrisDVA) and custom scripts (MATLAB and Python).

Light stimuli were delivered with a dual OptoLED light stimulation system (Cairn Research) through a custom-built optical pathway (Thorlabs, Newton, NJ), which fed into the transmitted light path of the physiology microscope. The stimulus LED a had peak wavelength of 405 nm. Stimulus light was attenuated by absorptive neutral-density filters (Thorlabs). At the beginning of each experiment, the microscope field-stop aperture was focused at the level of the slice to provide uniform illumination.

The intensities of the LEDs were measured with a calibrated photodiode (Gamma Scientific, San Diego, CA) through a photodiode amplifier (PDA200C, Thorlabs). Light intensities were calibrated as effective photons μm−2 and adjusted for the absorption spectrum of rhodopsin and E122Q rhodopsin (Govardovskii et al. 2000; Nymark et al. 2012; Yue et al. 2017). Stimulus intensities were then converted to light-activated rhodopsins per rod (R* rod−1) by accounting for the effective collecting area of a rod outer segment of 0.32 μm2.

Response-intensity relations were analyzed by relating the response amplitudes, R, to the flash strengths, I, with a Michaelis–Menten equation as follows:

$$R=_\frac_\right)},$$

(6)

where Rmax is the response amplitude to a saturating flash stimulus and I1/2 is the flash strength required to elicit a half-saturated response.

Recovery of the response to saturating light flashes was analyzed as previously, with Tsat taken as response recovery to 75% of the saturating response amplitude plotted as a function of the natural logarithm of the stimulus intensity and fitted with a linear regression (Pepperberg et al. 1992). The slope of this fit represents the recovery time constant.

With this method (see Vinberg and Kefalov 2015; Frederiksen et al. 2021), a piece of filter paper (Millipore, 0.45 µm pore size) was affixed to the bottom compartment of a perfusion chamber. The recording chamber was filled with Locke’s solution which contained (in mM) 140 NaCl, 4 KCl, 5 MgCl2, and 2 CaCl2. The retina was then gently mounted to the filter paper with the photoreceptor side facing upward and the recording chamber sealed. One Ag/AgCl electrode was mounted to the chamber in contact with the solution on each side of the retina. Both electrodes were connected to a DP-311 differential amplifier (Warner Instruments, Hamden, CT). A gravity-fed perfusion system was used to superfuse the retinal tissue with Ames’ medium at a rate of 2 ml/min and held at 35–37 °C with a temperature controller. Photoreceptor responses were isolated by supplementing the solution with 40 µM DL-2-amino-4-phosphonobutyrate (AP4, Tocris Bioscience, Bristol, UK), and 100 µM BaCl2 (Millipore-Sigma, St Louis, MO, USA). A dual OptoLED light source (Cairn Research, Faversham, UK) was coupled to a custom-built optical system to deliver both stimulus and background light. The stimulus and background lights were delivered from separate 505-nm LEDs and attenuated via separate absorptive neutral density filters. To provide the retina with a uniform illumination, the optical system was fitted with a circular field-stop aperture in focus in the same plane as the preparation. Recordings were low-pass filtered at 100 Hz and digitized at 1 kHz with a NI USB-6365, X Series DAQ Device (National Instruments Corp., Austin, TX). Data were collected with a MATLAB-based acquisition package: Symphony Data Acquisition System (https://open-ephys.org/symphony/). Data analysis and plotting were done with a combination of the Iris DVA (https://github.com/Khlick/IrisDVA) and MATLAB scripts.

All data are reported as mean ± standard error of mean (S.E.M.) unless otherwise stated. Statistical significances were determined using two-tailed Welch t-tests.