Retinal function of the voltage-gated calcium channel subunit α2δ-3 / Light-dependent effects in α2δ-3 mutant and in wild type retina

Abstract

The retina employs a large number of cell types that fulfill a broad spectrum of computations. It comes as no surprise that this complex network would make use of an equal diversity of molecular tools, such as voltage-gated calcium channels (VGCC). In fact, all pore-forming α1 subunits of VGCC and modulatory β and auxiliary α2δ subunits were found in the retina. Yet, little detail is known about the functional implementation of individual VGCC subunits in the retinal circuitry.

My work described in part 1 focused on the retinal expression and function of one VGCC subunit, called α2δ-3, employing an α2δ-3 knockout mouse. I found transcription of all α2δ subunit genes throughout postnatal retinal development and strong expression of α2δ-3 in horizontal cells. Yet, in my patch-clamp recordings from isolated horizontal cells I did not find an impact on their somatic calcium currents, leaving a possible involvement of α2δ-3 in the horizontal cell axon-to-rod connection. Outer retina function, determined by electroretinogram, and optokinetic reflex behavior was normal in α2δ-3 knockout animals. However, I discovered changes to the retinal output in micro-electrode array recordings of ganglion cell responses. I applied a paradigm of light stimulation at different ambient luminance levels that revealed effects of the α2δ-3 knockout only in scotopic and mesopic light levels. In summary, α2δ-3 is a candidate for horizontal cell axon-specific calcium signal modulation and exerts its function in non-photopic regimes.

The retina constantly adapts to features of the current visual environment, most prominently, the ambient light intensity or luminance. These adaptations are based on mechanisms throughout the retinal network. Adaption is commonly considered to keep signal processing within the dynamic range of the system as well as keep the retinal output stable across changing conditions, such as the light intensity. The results of part 1 show that different building blocks of retinal circuits - here the α2δ-3 subunit - can contribute to retinal function at distinct light level regimes.

In part 2, we looked more generally at the output of the retina (responses of ganglion cells) across different levels of ambient luminance. We found that ganglion cell responses were not stable across luminance levels, neither in single ganglion cell types nor in the ganglion cell population, but that they changed their responses qualitatively. These response changes were also reflected downstream in the activity of the lateral geniculate nucleus. Furthermore, we observed that rod photoreceptors could drive visual responses of ganglion cells in photopic luminance levels, where they were commonly thought to be saturated. While experiencing initial incremental saturation upon stepping to photopic luminance, rods recovered responsiveness at all light levels tested, but the rate of recovery was faster with brighter ambient light intensity. Computational modeling suggested adaptive translocation of elements of the signal transduction cascade as potential explanations for rod signaling at high light intensities. The photopic rod activity dynamics have important implications for the interpretation of experimental data and for the question of rod photoreceptor contributions to daylight vision.

In summary, while some circuitry elements associated with luminance regimes are known (e.g. rod and cone pathways), details on the underlying molecular mechanisms are scarce. My data suggests α2δ-3 as a promising candidate for a molecular determinant of light adaptation that could exert its function within horizontal cells in an axonal compartment-specific way. It will be interesting to pinpoint the exact role of α2δ- 3 in retinal light adaptation and to determine what (sub-)cellular function this protein serves in horizontal cells.