Gordon Fain

When the rhodopsin visual pigment of vertebrate rods absorbs light, photoisomerization of its 11- cis retinal chromophore to all-trans retinal triggers a conformational change, ultimately leading to the formation of metarhodopsin II (meta II), the active signalling state. Meta II initiates the phototransduction cascade, culminating in membrane hyperpolarization. To investigate how meta II lifetime shapes the rod photoresponse, we studied a knock-in mouse expressing E122Q rhodopsin, a mutant pigment with a shortened meta II lifetime. Microspectrophotometry confirmed that E122Q rhodopsin is blue-shifted in its absorbance spectrum and exhibits significantly faster meta II decay than wild-type (WT). Patch-clamp recordings revealed that this shortened lifetime had little effect on response kinetics or on the dominant time constant of recovery after bright flashes. However, following pigment bleaching, rods expressing E122Q showed markedly greater desensitization than WT, indicating that free E122Q opsin activates the phototransduction cascade more efficiently in darkness. To our knowledge, this is the first report of a rhodopsin mutation that alters free opsin-driven activation of phototransduction. Given that many disease-causing mutations affect dark adaptation, such opsin-based mechanisms may be more common than previously appreciated.

Identifying homologous cell types across species is essential for understanding cell type evolution. The retina is ideal for comparative analysis because its six major cell classes have persisted since the origin of vertebrates more than half a billion years ago. Here, we show that the retina’s conserved cellular architecture is mirrored by deep conservation of the cis-regulatory codes that govern gene expression. Through single-cell chromatin accessibility analysis of lamprey, fish, bird, and mammalian retinas, we demonstrate cross-species conservation of cis-regulatory codes in all retinal cell classes despite extensive turnover of cis-regulatory regions. Conservation manifests as clustering of high-affinity transcription factor binding sites in cell class–specific open chromatin regions. Thus, the retina’s cellular Bauplan is controlled by cis-regulatory codes, which predate the divergence of extant vertebrates.

To characterize the function of the Ca ²⁺ -activated Cl ⁻ current I _Cl(Ca) in mammalian rod photoreceptors, we made patch-clamp recordings from retinal slices of mice ( Mus musculus ) of both sexes that lack Ano2 (TMEM16B). Depolarizing voltage ramps in solutions blocking K ⁺ currents elicited a large outward current inhibited by the Cl ⁻ channel blocker niflumic acid; this current was absent in Ano2 ^−/− rods. The membrane potential of Ano2 ^−/− rods was 10–15 mV more depolarized in darkness than WT or Cx36 ^−/− rods, indicating a substantial resting Cl ⁻ permeability. Rod outer-segment photocurrents were similar in waveform and amplitude in Ano2 ^−/− and Cx36 ^−/− rods, but photovoltages in Ano2 ^−/− rods were nearly doubled. Measurements of light-response reversal potentials in rods with and without Ano2 suggest that the outer-segment conductance is nearly linear with a reversal potential of −9 mV and that Δ g Cl increases during the light response. Using these results, we estimated E _Cl from permeabilized patch recordings of reversal potentials of Cx36 ^−/− rods to have a mean value of −35 mV near the rod resting potential, but other evidence suggests that E _Cl may be more positive by as much as 10–15 mV. Thus activation of I _Cl(Ca) during the light response would be depolarizing. At dim intensities, the photocurrents of downstream rod bipolar cells were larger and about twice as sensitive in Ano2 ^−/− retinas with reduced nonlinearity. These experiments show that Ca ²⁺ -activated Cl ⁻ currents in mammalian rods have more important roles in photoreceptor physiology than previously appreciated.

Bipolar cells are vertebrate retinal interneurons conveying signals from rod and cone photoreceptors to amacrine and ganglion cells. Bipolar cells are found in all vertebrates and have many structural and molecular affinities with photoreceptors; they probably appeared very early during vertebrate evolution in conjunction with rod and cone progenitors. There are two types of bipolar cells, responding to central illumination with depolarization (ON) or hyperpolarization (OFF). In most vertebrate species, rod signals are conveyed to specialized rod bipolar cells, which sum signals from many rods and facilitate detection at the visual threshold. Lamprey, which diverged from all other vertebrates in the late Cambrian, have both rod ON and rod OFF bipolar cells, but mammals have only rod ON cells. Rod signals in mammals are conveyed to output neurons indirectly via AII (or A2) amacrine cells, which synapse onto cone ON and cone OFF bipolar‐cells and then to ganglion cells. These findings raise the question of when during retinal evolution rod OFF bipolar cells were lost. Because physiological recordings have been made from rod OFF bipolar cells in both cartilaginous fishes (dogfish) and urodeles (salamanders), rod OFF bipolar cells and their circuits must have been retained in vertebrate progenitors at least until the Devonian. Recent evidence showing that zebrafish retina processes rod signals similar to those in mammals indicates that rod OFF bipolar cells were lost at least twice. The sole utilization of rod ON bipolar cells may have provided a selective advantage from increased signal‐to‐noise discrimination near the visual threshold. image Key points: Rods and cones have many structural and molecular similarities to bipolar cells, which are retinal interneurons conveying signals from photoreceptors to the retinal output. Bipolar cells can be either ON (centre depolarizing) or OFF (centre hyperpolarizing) and either rod or cone dominant. Lamprey, which diverged from all other vertebrates 500 million years ago, have both ON and OFF bipolar cells, which can each be either rod or cone dominant. We argue that this configuration of separate rod/cone bipolar‐cell pathways is representative of early vertebrates. Rod ON and rod OFF bipolars persisted at least until the progenitors of amphibians in the Devonian, but mammals and teleost fishes have only rod ON bipolar cells and convey rod OFF signals via a specialized amacrine cell. We argue that rod OFF bipolar cells were lost in at least two different lineages during vertebrate evolution, probably to increase the signal‐to‐noise of rod vision.

The lamprey, a primitive jawless vertebrate whose ancestors diverged from all other vertebrates over 500 million years ago, offers a unique window into the ancient formation of the retina. Using single-cell RNA-sequencing, we characterize retinal cell types in the lamprey and compare them to those in mouse, chicken, and zebrafish. We find six cell classes and 74 distinct cell types, many shared with other vertebrate species. The conservation of cell types indicates their emergence early in vertebrate evolution, highlighting primordial designs of retinal circuits for the rod pathway, ON-OFF discrimination, and direction selectivity. The diversification of amacrine and some ganglion cell types appears, however, to be distinct in the lamprey. We further infer genetic regulators in specifying retinal cell classes and identify ancestral regulatory elements across species, noting decreased conservation in specifying amacrine cells. Altogether, our characterization of the lamprey retina illuminates the evolutionary origin of visual processing in the retina.

This review article is an excellent introduction to retinal degeneration, which should be just as useful to students as to senior researchers in the field. Highly recommended.

Grabner and colleagues show that different cone OFF bipolar cells in the ground squirrel retina respond to signals from cones quite differently. One bipolar cell called cb2 enters cone invaginations and has GluA4 receptors; these cells respond to changes in cone membrane potential nearly linearly. Two other cell types, called cb1 and cb3, make basal contacts on cone synaptic endings and use kainate receptors; these cells respond nonlinearly. The cb3 cells respond to two or more locally coincident events from cones, but the cb1 cells require a larger number of events and only respond to multivesicular or multiquantal events. This nonlinearity seems not to result from a nonlinearity inherent in the synaptic receptors but rather from differences in the arrival of glutamate at the cone basal synapses due to differences in the distance of diffusion and the distribution of sites for receptor binding and transmitter transport back into the cone terminal by TBOA-sensitive glutamate transporters. These results indicate that different kinds of bipolar cells may sample cone signals in distinct ways, providing parallel pathways for the analysis of visual signals communicated to ganglion cells and thence to the CNS.

In many forms of retinitis pigmentosa, the genetic alteration producing disease is an alteration of a rod gene. The rods first degenerate, but cones then also die leading to complete blindness. Much attention has recently been given to the question of why the cones die after the rods have degenerated, and whether any measures can be taken to keep the cones from dying to preserve at least some visual function. In this paper from the Cepko lab, Xue and colleagues show that mice with a deletion in the gene for CRALBP, a retinal carrier protein; or a mutation in the RPE65 isomerase, both slow the rate of cone degeneration in mouse models of retinitis pigmentosa. These genetic alterations would retard pigment regeneration and slow delivery of 11-cis retinal to cones. Xue and colleagues further show that ectopic expression of RPE65 and LRAT in cones, which would increase the delivery of chromophore to the cones, increases the rate of cone degeneration. The authors hypothesize that high chromophore supply to cones upon the loss of rods may be toxic to cones, and that clinical strategies to reduce retinal supply to cones may enhance photoreceptor survival.