Stephanie A. Hagstrom

Defects in RGS9 or its anchor protein R9AP in patients with slow photoreceptor deactivation

The RGS proteins are GTPase activating proteins that accelerate the deactivation of G proteins in a variety of signalling pathways in eukaryotes. RGS9 deactivates the G proteins (transducins) in the rod and cone phototransduction cascades. It is anchored to photoreceptor membranes by the transmembrane protein R9AP (RGS9 anchor protein), which enhances RGS9 activity up to 70-fold. If RGS9 is absent or unable to interact with R9AP, there is a substantial delay in the recovery from light responses in mice. We identified five unrelated patients with recessive mutations in the genes encoding either RGS9 or R9AP who reported difficulty adapting to sudden changes in luminance levels mediated by cones. Standard visual acuity was normal to moderately subnormal, but the ability to see moving objects, especially with low-contrast, was severely reduced despite full visual fields; we have termed this condition bradyopsia. To our knowledge, these patients represent the first identified humans with a phenotype associated with reduced RGS activity in any organ.

Variations in cone populations for red-green color vision examined by analysis of mRNA.

IN the central human retina, there are estimated to be nearly two L cone photoreceptors for each M cone. The extent to which this value varies across individuals is unclear and little is known about how the M:L cone ratio might change with retinal location. To address these questions, the ratio of M:L cone pigment mRNA was examined at different locations. For patches of central retina, the average M:L ratio was about 2:3 which decreased to about 1:3 for patches 40°eccentric. There were also large individual differences among the 23 eyes examined. The extremes differed in central M:L mRNA ratio by a factor of > 3. The measured differences in mRNA ratio are proposed to reflect differences in photo-receptor ratio. Such variations provide unique opportunities for understanding how the neural circuitry for color vision is affected by changes in cone ratio.

Cone pigment gene expression in individual photoreceptors and the chromatic topography of the retina

Human trichromatic vision is based on three classes of cones: L, M, and S (long-, middle-, and short-wavelength sensitive, respectively). Individuals can have more than one M and/or more than one L pigment gene on the X chromosome along with an S pigment gene on chromosome 7. In some people the X-linked pigment gene array can include polymorphic variants that encode multiple, spectrally distinct cone photopigment subtypes. A single-cell, polymerase chain reaction approach was used to examine visual pigment gene expression in individual human cone cells and identify them as L or M. The ratio of L:M pigment gene expression was assayed in homogenized retinal tissues taken from the same eyes. Results indicate that there is a close correspondence between the cone ratio determined from counting single cells and the L:M pigment mRNA ratio estimated from homogenized pieces of retina. The results also show that the different pigment genes in one array are often expressed at very different levels, giving rise to unequal numbers of L and M cones. Expression of only one photopigment gene was detected in each cone cell. However, individual males can have more than the classically described three spectrally distinct cone types in their retinas.

Ratio of M/L pigment gene expression decreases with retinal eccentricity

Evidence has accumulated to indicate that, on average, there are about 1.5-2.0 long-wave cones in the human fovea for each middle-wave cone. Much less is known about how the ratio of these cone populations might change with retinal eccentricity. We have examined how the ratio of middle- to long-wave cone pigment mRNA changes with eccentricity in individual human retinas. Retinas were examined from seven male eye donors. Patches of retina, 6 mm in diameter were removed using a trephine at three different eccentricities. mRNA was reverse transcribed, photopigment cDNAs were amplified, and the relative amounts of middle- and long-wave pigment mRNA were determined for each patch. For 6 mm patches centred on the fovea, the average ratio of middle- to long-wave mRNA was about 2:3. This value is similar to the average ratio of middle- to long-wave cones previously estimated for the human fovea. There were significant changes in the cone ratios with increasing eccentricities (p = 0.0004). For patches of retina taken from the most peripheral locations (centred 12 mm eccentric from the fovea), the average middle- to long-wave mRNA ratio decreased to about 1:3. These results can be explained by a decrease in the ratio of middle- to long-wave cone populations with increasing eccentricity from central retina to periphery

Protein partners of dynamin-1 in the retina.

Dynamin proteins are involved in vesicle generation, providing mechanical force to excise newly formed vesicles from membranes of cellular compartments. In the brain, dynamin-1, dynamin-2, and dynamin-3 have been well studied; however, their function in the retina remains elusive. A retina-specific splice variant of dynamin-1 interacts with the photoreceptor-specific protein Tubby-like protein 1 (Tulp1), which when mutated causes an early onset form of autosomal recessive retinitis pigmentosa. Here, we investigated the role of the dynamins in the retina, using immunohistochemistry to localize dynamin-1, dynamin-2, and dynamin-3 and immunoprecipitation followed by mass spectrometry to explore dynamin-1 interacting proteins in mouse retina. Dynamin-2 is primarily confined to the inner segment compartment of photoreceptors, suggesting a role in outer segment protein transport. Dynamin-3 is present in the terminals of photoreceptors and dendrites of second-order neurons but is most pronounced in the inner plexiform layer where second-order neurons relay signals from photoreceptors. Dynamin-1 appears to be the dominant isoform in the retina and is present throughout the retina and in multiple compartments of the photoreceptor cell. This suggests that it may function in multiple cellular pathways. Surprisingly, dynamin-1 expression and localization did not appear to be disrupted in tulp1−/− mice. Immunoprecipitation experiments reveal that dynamin-1 associates primarily with proteins involved in cytoskeletal-based membrane dynamics. This finding is confirmed by western blot analysis. Results further implicate dynamin-1 in vesicular protein transport processes relevant to synaptic and post-Golgi pathways and indicate a possible role in photoreceptor stability.