Background

In the realm of vertebrate visual systems, the phototransduction cascade plays a crucial role in enabling effective vision across a vast spectrum of light intensities, from the dimmest of night conditions to the brightness of midday. This intricate biochemical pathway operates through a multi-stage amplification system, characterized by several feedback loops, allowing photoreceptors to adapt to varying illumination levels. The secondary messengers involved in this process, cyclic guanosine monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP), each serve distinct yet interrelated functions within the cascade. While the role of cGMP in regulating the permeability of photoreceptor plasma membrane channels has been extensively documented [1,2], the nuanced roles of cAMP have garnered less attention, despite its significant involvement in various photoreceptor processes, including modulation of circadian rhythms and the dynamics of retinomotor movements [3–8].

Previous studies have demonstrated that several key proteins within the phototransduction pathway—such as phosphodiesterase type 6 (PDE6) and rhodopsin kinase—are susceptible to phosphorylation by protein kinase A (PKA), suggesting a regulatory interplay between cAMP levels and the efficiency of the phototransduction cascade [8,9]. However, the dynamics of cAMP concentration within photoreceptors remain inadequately characterized, particularly concerning the rapid fluctuations necessary for real-time regulation of the cascade's activity.

Existing protocols for measuring intracellular signaling, including cAMP dynamics, have predominantly relied on fluorescent methods, which are often unsuitable for the unique conditions pertinent to photoreceptor physiology. The potent light emission employed in the fluorescence technique itself constitutes an adequate stimulus for the phototransduction cascade. The rapid cryofixation technique utilized in this method presents significant advantages over these conventional approaches, allowing for precise measurement of cAMP levels immediately following light stimulation. By employing highly sensitive metabolomics techniques paired with cryofixed samples, we provide a robust protocol that significantly enhances our ability to capture real-time biochemical changes during phototransduction.

The general idea of rapid freezing of retinal fragments was previously proposed and realized by Govardovskii [10] and Blazynski and Cohen [11, 12], who used this approach to study the effect of light stimulation on intracellular cGMP levels in amphibian outer segments. In both cases, the cryofixation surface was a mirror-polished surface of a copper cylinder cooled to the temperature of liquid nitrogen [10] or liquid helium [11]. The Govardovskii setup allowed simultaneous fixation of four retinal fragments with a minimum sample cryofixation time of 500 ms. We refer to the minimum time here as the time interval from command to cryofixation of the specimen. The Blazinski setup was a single-channel unit with a minimum cryofixation time of 47 ms.

In designing our setup, we considered that it should be multi-channel and allow us to fix retinal fragments from two eyes of an animal in one experiment. Experimentally, we have found that the optimal size of a retinal fragment for obtaining a tissue sample using a cryotome is 1/2 to 1/3 of the retina of one eye. Thus, in one experiment, it is possible to simultaneously fixate and then measure cAMP levels in 4–6 retinal fragments from two eyes of an animal. To perform such an experiment, we created a setup consisting of six identical sections in which a fast-moving stepper motor (maximum speed 150,000 rpm) drives a lever with a retinal sample pad. The retinal sample, placed on a small circle of filter paper, is placed horizontally on the pad and is illuminated with LED light (λ_max= 525 nm) on command from the computer. The intensity of the light used to stimulate the retina can be varied within a range of 3.6 × 10³–3.6 × 10⁵ quanta/(s·μm²). If necessary, a grey light filter can be used to further attenuate the light flux. Synchronous illumination of all six retinal specimens is achieved by a special design of the light guide that carries the light from the LED to the retinal specimens. The light guide has six identical fiber bundles at one end, each of which illuminates one location with a retinal fragment. At the other end, the fibers from all six terminations are combined into a single bundle by uniform and random mixing. The end of this bundle is illuminated through a condenser by an LED.

After a programmable delay, the sample arm arcs 180° and presses the retinal sample against the polished surface of a copper cylinder cooled to liquid nitrogen temperature. The time taken for the lever to move from the command stage to making contact with the surface of the copper cylinder (the minimum sample cryofixation time) was measured using a slow-motion video of the lever's rotation. The results of multiple measurements showed a small variation in the individual minimum cryofixation times, which for channels 1–6 are 153 ± 12, 94 ± 9, 92 ± 3, 97 ± 6, 90 ± 1, and 110 ± 2 (shown as average ± SD, ms, n = 3–5). The average time for all channels is 106 ms. We used the measured minimum cryofixation times to calculate the actual cryofixation delay time for each individual channel. To prevent the retinal specimens from flying off the platform during the rapid movement of the lever, we glued filter paper circles on which the retinal specimens were placed to the platform on the lever using Tissue-Tek gel. To enable the Tissue-Tek frozen after cryofixation to be easily removed from the lever platform, we made a composite platform with a fluoroplastic core.