Many diseases that ultimately lead to blindness are caused by the degeneration of photoreceptor cells, the rods and cones

PHOTORECEPTOR DEATH IN MOUSE MODELS OF RETINITIS PIGMENTOSA

Claudio Punzo (Postdoctoral Fellow)

Bo Chen (Postdoctoral Fellow)

Karin Roesch (Postdoctoral Fellow)

Many diseases that ultimately lead to blindness are caused by the degeneration of photoreceptor cells, the rods and cones. Non-autonomous photoreceptor death is caused when a disease gene is expressed in e.g. rods only, but cones also die. The spread of non-autonomous death is the cause of blindness in retinitis pigmentosa (RP), and possibly other, very common diseases in humans, including age related macular degeneration (AMD) (Retnet, http://www.sph.uth.tmc.edu; Clin Genet, 2000. 57(5): p. 313-29). RP can have a variable age of onset, but the fact that a person is first night blind due to rod death and/or malfunction, prior to cone death makes it possible to intervene before all vision is lost. For example, if we knew how to arrest photoreceptor degeneration, there is an excellent possibility for gene therapy through infection of the adjacent retinal pigmental epithelium (RPE), or the photoreceptors themselves, with viral vectors. The RPE has been shown to stably express genes transduced by adeno-associated vectors. The most striking example is a treated Briard dog, Lancelot, who is still showing therapeutic effects for his inherited blindness 7 years following infection (Nat Genet. 2001, 28:92-5). More importantly, humans have also shown no untoward effects from delivery of the same gene via the same vector (Maguire et al. 2008). However, we are missing key pieces of information for this type of therapy: the identity of a gene(s) that could be delivered to keep the photoreceptors alive for many cases of human blindness. This is the goal of our studies.

We have used two approaches to learning what types of genes might be able to prolong photoreceptor survival in diseases of autonomous and non-autonomous photoreceptor death. For the non-autonomous cone death, such as occurs in RP, we began by using retinal microarrays to define the gene expression changes that accompany photoreceptor death in mouse genetic models (Punzo et al. 2009). We focused on those changes that occurred at the time of onset to cone death (Figure 1). Cones are a primary target as these are the photoreceptor cells that allow high acuity and daylight vision, while rods are used only in very dim light. The microarray data showed that metabolic genes are altered in their expression levels at the onset of cone death (Figure 1B, C).

Figure 1: We investigated rod and death kinetics of 4 different mouse models of RP. We then made RNA preparations at 4 different time points corresponding to different stages in the disease process. These are different ages in the different mutants as rod death proceeds at very different rates in the 4 different mutants. The disease stages were: R: approximately halfway through the major phase of rod death; C0: onset of cone death; C1 & C2 first and second time point during cone death respectively). Time is indicated in postnatal days (P) or postnatal weeks (PW). Affymetrix microarray analysis was then performed on the samples from the different ages. Genes (195) that were upregulated at C0 only were analyzed for their functional annotation. (a) Equivalent time points in the 4 different mutants at which the microarray analysis was performed Cartoons depicting the progression of cone death are shown below the corresponding time points. (b) Distribution in percentage of the 195 genes that were annotated. (c) Distribution in percentage of the 68 genes (34.9%) that are part of metabolism in (b). From Punzo et al. 2009.

To follow this up, we have examined a number of metabolic regulators and metabolic status in the degenerating retina. We saw evidence for lack of activation of mTOR, a central regulator of growth (Figure 2). We also saw that red-green opsin synthesis was reduced in the ventral retina. In addition, there was evidence of autophagy, suggesting that the cones are starving. One hypothesis based on these findings is that there is altered glucose availability in cones as a result of rod death. This might be due to a collapse of the normal architecture between photoreceptor outer segments and an adjacent cell type, the retinal-pigmented epithelium (RPE) (Figure 3). The RPE is normally in tight contact with both rods and cones and it is through the RPE that the blood delivers nutrients and oxygen. In addition, the RPE provides other types of support for rods and cones. Rods outnumber cones by more that 20:1 and roughly 30 photoreceptors contact one RPE cell. Thus the loss of rods could alter the intimate interactions between the RPE and the remaining cones, and might result in reduced uptake of nutrients into cones. We are therefore trying to understand to which extent metabolism is altered in cone photoreceptors during disease progression, as well as the normal metabolic requirements of rods and cones. One clue that the hypothesis of nutrient availability is correct is that insulin was able to promote cone survival in a rapid model of rod and cone death (Figure 4). Insulin signaling can lead to activation of mTOR, which might temporarily fool cone photoreceptors into slowing down their autophagy, and thus live a bit longer.

Figure 2: p*-mTOR in wild type and degenerating retinae. All panels show immunofluorescence on retinal flat mounts (photoreceptor side up) with the exception of (b, c) which show retinal sections. Blue shows the nuclear DAPI stain. (a–c) p*-mTOR levels in wild type retinae. (a) Dorsal (up) enrichment of p*-mTOR. Higher magnification of dorsal and ventral region is shown to the right showing p*-mTOR in red and cone segments in green as detected by PNA. (b, c) Dorsal retinal sections stained for p*-mTOR (red signal) and PNA (b) or a-b-galactosidase, which marks cones in theis strain (c) (green signal). Higher magnification (insets: b, c) suggests that the p*-mTOR signal is located in the lower part of the outer segment (OS; IS: inner segment). (d–i) Reduced levels of dorsal p*-mTOR during photoreceptor degeneration (red signal). (d) Wild type control. (e, f) PDE-b mutant. Reduction starts during rod death at P15 (e) as the OSs (green signal: PNA) detach from the retinal pigmented epithelium. (f) By P30 only few cones (green signal: a-b-galactosidase) show high levels of p*-mTOR (red signal). (g–i) Similar reduction in dorsal cones of the other three mutants (cones marked in green by PNA). (g) PDE-g^–/– P35. (h) Rho^–/– PW20. (i) P23H PW70. From Punzo et al. 2009.

Figure 3: Schematic representation of a retinal cross-section showing the retinal-pigmented epithelium (RPE) and the rod and cone photoreceptors. (a) Cross-section through a wild-type retina. Roughly 30 photoreceptor outer segments interact with one RPE cell of which only 1-2 are from cones. (b) Cross-section through a diseased retina prior to the onset of major cone death but after major rod death. Due to the death of most of the rod photoreceptors, the outer nuclear layer, which contains mainly rods, collapses. As a result of the collapse the remaining rod outer segments and the cone outer segments have perturbed interactions with the RPE.

Figure 4: Insulin levels affect cone survival. (a–c) Retinal flat mounts of PDE-b mutants at PW7 stained for lacZ ⁴⁸ ^, ⁴⁷ to detect cones. (a) Example of untreated control. (b) Example of mouse injected with streptozotocin. (c) Example of mouse injected daily with insulin. (d) Quantification of cone survival after 4 weeks of treatment. Data represents an average of at least 8 retinae and indicates on the y-axis percentage of cone surface area versus surface area of entire retina. (e) Measurements of blood glucose levels and body weight (f) performed weekly over the time span of the experiment. Error bars in (d–f) show standard deviations. From Punzo et al. 2009.

A second approach to prolonging survival is to deliver genes directly to the rods and cones in an attempt to stave off either autonomous or nonautonomous death. We initially targeted rod photoreceptors which have a mutation that leads to rapid death early in the postnatal period (the rd1 mouse model). We had discovered that histone deacetylase 4 (HDAC4) was required for rod survival even in a normal mouse (Chen and Cepko, 2009). We thus reasoned that overexpression of HDAC4 might promote survival of rods in a pathological case, such as in the rd1 mouse model of RP. We delivered the HDAC4 gene and found that indeed, rods survived much longer than they typically did in this model (Figure 5). We subsequently found that at least part of the activity of HDCA4 was mediated by deactelylation of hypoxia inducible factor 1 alpha, which stabilized HIF1alpha. Interestingly, this activity was mediated by HDAC4 in the cytoplasm. We are continuing to study the mechanisms that lead to increased survival in rods as well as in other cell types.

In addition to the approaches listed above, we have also been investigating the role of glia in the degeneration process. Neuronal cell death in the CNS often involves changes in non-neuronal cells. This holds for rod and cone photoreceptor death in the retina. Resident Müller glial cells, the major type of glia in the mammalian retina, have been shown to respond to neuronal cell loss by undergoing reactive gliosis, that is characterized by increased intermediate filament gene expression. Additional changes in gene expression have not been studied. Moreover, the functional significance of reactive gliosis has not been explored.

We are interested in exploring this response of Müller glial cells to photoreceptor degeneration. We have already performed single cell gene expression profiling of single Müller glia from mouse models of retinitis pigmentosa. Comparison of these profiles to wildtype Müller glial cells revealed many exciting hypothesis on the function of Müller glia on which we are currently following up on. (See Roesch et al. 2008).

Figure 5. The left panel of the figure shows a section of a mouse retina in which the HDAC4 gene was delivered, along with the green fluorescent protein, GFP, shown in green. Cells which are green also received the HDAC4 gene, whereas adjacent tissue without green cells did not receive the HDAC4 gene. The middle panel shows the same tissue, now visualized for a red fluorescent protein, DsRed. Surviving rods express DsRed, directed by promoter elements only expressed in rods (the rhodopsin promoter). As can be seen in the area without the HDAC4 gene, all non-transfected rods which did not receive the HDAC4 gene, have died. In the panel on the right, the remaining cone photoreceptors in this area of the tissue, are shown. There are more cones in the area with surviving rods, likely due to the presence of rods.