Downloaded from http://www.jci.org on October 2, 2015. http://dx.doi.org/10.1172/JCI83194

COMMENTARY The Journal of Clinical Investigation It’s never too late to save a photoreceptor

James B. Hurley1,2 and Jennifer R. Chao1 1Department of Ophthalmology and 2Department of Biochemistry, University of Washington, Seattle, Washington, USA.

threshold could be raised), then “the likeli- hood that a mutant neuron can be rescued Recent therapy progress has raised the possibility that vision loss by treatment is not diminished by age, caused by inherited retinal degeneration can be slowed or prevented. although fewer cells will be available to Unfortunately, patients are not usually diagnosed until enough rescue. Therefore, treatment at any stage degeneration has occurred that the deterioration in vision is noticeable. of the illness is likely to confer benefit” (9). Therefore, effective gene therapy must halt degeneration to stabilize and This prediction was made in the year preserve any remaining vision. Gene therapy methods currently in human 2000, and since then, tremendous prog- clinical trials rely on subretinal or intravitreal injections of adeno-associated ress has been made in many aspects of virus to deliver the therapeutic gene. To date, long-term results in patients gene therapy (10, 11). For example, one treated with subretinal injections for Leber congenital amaurosis have promising study (12) showed that injec- been mixed. Proposed limitations include variability in the gene delivery tion of adeno-associated viruses (AAV) method and a possible point of no return, at which treatment would be carrying a guanylyl cyclase–encoding gene ineffective. In this issue of the JCI, Koch et al. describe a well-controlled halted degeneration of guanylyl cyclase– and precise mouse model for testing the ability of gene therapy to halt deficient rods in mice. This result sug- the progress of degeneration. Instead of viral-mediated therapeutic gene gested that AAV-mediated gene therapy delivery, the authors induced expression of an integrated transgene at could be used to treat human patients with specific times during the course of photoreceptor degeneration. In Pde6b- guanylyl cyclase deficiencies. In fact, such deficient retina, this strategy halted degeneration, even when more than studies are underway (13), but results have 70% of photoreceptors had already degenerated. The results of this study not been reported yet. demonstrate that retinal degeneration can be stopped, even at late stages of disease. What limits success of gene therapy in human retina? In human patients with retinal degenera- tion, gene therapy has so far largely focused on treatment of the inherited retinal degen- Congenital blinding diseases Why and when do mutant eration LCA2, which is caused by an RPE65 Hereditary blindness has been linked to photoreceptors die? deficiency. Clinical trials began more than at least 238 mutations that cause photo- In some animal models of inherited reti- seven years ago, and patient outcomes have receptor neurons in the retina to degen- nal degeneration, the timing of individual been analyzed extensively. Retinal sensi- erate (1). For example, mutations in rod death and the spatial distribution of tivity in the treated regions of the retinae the gene that encodes the β subunit of dying rods appear to be random. In gen- improved significantly (14–16); however, PDE6 cause a type of rod photoreceptor eral, the rate at which rods degenerate in over time, retinae exhibited continued thin- degeneration known as retinitis pigmen- these models is simply proportional to the ning, and it is apparent that degeneration tosa (RP) (2–4). PDE6 hydrolyzes cyclic number of remaining rods, reminiscent of was not effectively stopped (17–19). GMP (cGMP) to 5′GMP in rods, is acti- the spontaneous decay of radioactive iso- The most pessimistic explanation of vated by light, and has a central role in topes. On the basis of these observations, the lack of efficacy is that continued thin- phototransduction (5, 6). Defects in the Clark et al. (9) argued that degeneration ning of the retina reflects an inherent fea- that make cGMP from GTP also of a photoreceptor is not caused by grad- ture of gene therapy that limits the ability result in rod degeneration (7). For exam- ual accumulation of damage and instead to halt the progress of degeneration. While ple, guanylyl cyclase deficiencies cause proposed that degeneration is linked to a this interpretation seems unlikely, it is a the inherited degenerative disease Leber single random event that occurs in the con- challenge to rigorously exclude the hypoth- congenital amaurosis type 1 (LCA1); text of a perturbed metabolic state on the esis that damage accumulates in photo- however, the biochemical pathway that threshold of insufficiency. Moreover, Clark receptors to a point of no return, where links abnormal cGMP metabolism is not and colleagues suggested that if those sin- the trajectory to death is irreversible (18). fully understood (8). gle events could be prevented (or if the Alternatively, the continued retinal thin- ning may instead reflect only a technical Related Article: p. 3704 limitation, such as insufficient delivery of the virus, or damage caused by temporary Conflict of interest: The authors have declared that no conflict of interest exists. detachment of the neurosensory retina, Reference information: J Clin Invest. 2015;125(9):3424–3426. doi:10.1172/JCI83194. especially in the subfoveal region, that pre-

3424 jci.org Volume 125 Number 9 September 2015 Downloaded from http://www.jci.org on October 2, 2015. http://dx.doi.org/10.1172/JCI83194

The Journal of Clinical Investigation COMMENTARY

Figure 1. Strategy that allows evaluation of the point of no return for gene therapy to treat retinal degeneration. (A) Schematic of the genetic strategy developed by Koch et al. (21) to stimulate photoreceptor degeneration and evaluate the effect of rescuing the defect on demand by injecting the mouse with tamoxifen at different points during disease progression. The mouse has two different Pde6b alleles. One allele produces a mostly inactive protein as the result of an H620Q substitution. The other allele is normal, but it is interrupted by a floxed stop cassette. The very low PDE6 activity in the retinae of these mice causes rod photoreceptors to degenerate. (B) The Pde6g promoter is used to express CreERT2 in rods, and when the mouse is injected with tamoxifen, CreERT2 is activated and able to excise the stop cassette, resulting in expression of active PDE6. (C) In the absence of intervention, photo- receptors degenerate over time, resulting in vision loss. Tamoxifen-induced expression of active PDE6 at four or eight weeks of age halts degeneration. Although not shown in this schematic, the retinae continued to be stable out to 52 weeks of age.

vents an optimal response. Subretinal injec- those studies were hampered by the vari- can halt degeneration. Importantly, this tions are inherently invasive and variable, ability of subretinal injections. While gene strategy was able to show that expression and injected viruses do not reach the entire therapy for LCA2 still looks promising at of Pde6b completely halts progression of retina. Moreover, the reduction of trophic this point, the reasons for the partial rather photoreceptor degeneration in Pde6b- support that results from widespread cell than complete success are unclear. deficient retinae, regardless of the sever- loss in untreated areas could contribute to ity of loss. Specifically, continued photoreceptor degeneration in New strategy evaluates the Koch et al. devised an elegant genetic strat- treated areas. Some progress was made in point-of-no-return hypothesis egy that bypasses the need for AAV-medi- animal models to address these issues by In this issue, Koch et al. (21) describe a ated gene delivery (Figure 1). The authors using AAV to rescue LCA2 mouse retinae at new, highly effective strategy to directly engineered a mouse in which a therapeu- late stages of degeneration (20), but even evaluate how well expression of a gene tic gene can be activated on demand in a

jci.org Volume 125 Number 9 September 2015 3425 Downloaded from http://www.jci.org on October 2, 2015. http://dx.doi.org/10.1172/JCI83194

COMMENTARY The Journal of Clinical Investigation

degenerating retina. First, mice were gen- perturb guanine nucleotide metabolism, of the rd mouse. Proc Natl Acad Sci U S A. 1991;88(19):8322–8326. erated with two different alleles of Pde6b, whereas LCA2-causing mutations disrupt 5. Hurley JB. Molecular properties of the cGMP which encodes the PDE6 β subunit. One the visual cycle and cause massive accu- cascade of vertebrate photoreceptors. Annu Rev allele carried a mutation that partially mulation of retinyl esters in the RPE (22). Physiol. 1987;49:793–812. inactivates the and causes rods The diverse etiologies of these and 6. Zhang X, Cote RH. cGMP signaling in verte- to degenerate. The other Pde6b allele was other different types of degeneration could brate retinal photoreceptor cells. Front Biosci. 2005;10:1191–1204. normal, except that it was interrupted by a influence how well different inherited reti- 7. Boye SE. Insights gained from gene therapy in floxed insertion that blocks Pde6b expres- nal diseases respond to gene therapy. The animal models of retGC1 deficiency. Front Mol sion. This floxed Pde6b allele is designed strategy that Koch et al. devised (21) is well Neurosci. 2014;7:43. to be inactive in the absence of Cre-driven controlled and precise and bypasses the 8. Sancho-Pelluz J, et al. Photoreceptor cell death recombination. These mice also harbor uncertainty and variability associated with mechanisms in inherited retinal degeneration. Mol Neurobiol. 2008;38(3):253–269. a rod-specific, tamoxifen-inducible Cre; AAV injections. Gene therapists should 9. Clarke G, et al. A one-hit model of cell death therefore, expression of a normal PDE6 have a tool that helps them predict how in inherited neuronal degenerations. Nature. β subunit within all remaining rod cells well each type of degenerative disease will 2000;406(6792):195–199. can be activated at any time by tamoxifen respond to gene therapy and perhaps how 10. Boye SE, Boye SL, Lewin AS, Hauswirth WW. A injection. Once Koch et al. confirmed that to optimize the response. When inducible comprehensive review of retinal gene therapy. Mol Ther. 2013;21(3):509–519. Pde6b and Cre were expressed in rods as Cre becomes available in other cell types 11. Thompson DA, et al. Advancing therapeutic predicted, they used the engineered mice such as RPE, the genetic strategy devised by strategies for inherited retinal degeneration: to test whether progression of degenera- Koch et al. could provide an effective way to recommendations from the Monaciano tion could be halted by inducing normal make those types of predictions for LCA2 Symposium. Invest Ophthalmol Vis Sci. 2015;56(2):918–931. Pde6b expression at several different time and other types of retinal degeneration. 12. Boye SL, et al. AAV-mediated gene therapy points during the course of degeneration. in the (RetGC1/RetGC2) Remarkably, expression of Pde6b com- Acknowledgments double knockout mouse model of Leber pletely impeded the progression of degen- The authors acknowledge support from congenital amaurosis. Hum Gene Ther. eration, even when less than 30% of pho- NIH grants EY06641 (to J.B. Hurley), 2013;24(2):189–202. 13. Boye SE. Leber congenital amaurosis caused by toreceptors remained. EY017863 (to J.B. Hurley), and EY019714 mutations in GUCY2D. Cold Spring Harb Perspect Together, these findings show that (to J.R. Chao). We thank John Flannery for Med. 2015;5(1):a017350. restoration of a key protein can halt pro- helpful discussions. 14. Bainbridge JW, et al. Effect of gene therapy on gression of a degenerative disease. This visual function in Leber’s congenital amaurosis. demonstration has significant implica- Address correspondence to: James B. N Engl J Med. 2008;358(21):2231–2239. 15. Hauswirth WW, et al. Treatment of leber con- tions for the treatment of retinal degener- Hurley, Department of Biochemistry, UW genital amaurosis due to RPE65 mutations by ation in humans, because patients gener- Box 357350, University of Washington, ocular subretinal injection of adeno-associated ally do not seek care until after substantial 1705 NE Pacific St., Seattle, Washing- virus gene vector: short-term results of a phase I photoreceptor loss has caused a noticeable ton 98195, USA. Phone: 206.543.2871; trial. Hum Gene Ther. 2008;19(10):979–990. deterioration of vision. These results are E-mail: [email protected]. 16. Maguire AM, et al. Safety and efficacy of gene transfer for Leber’s congenital amaurosis. N Engl also consistent with the idea that wide- J Med. 2008;358(21):2240–2248. 1. Daiger SP. RetNET: Summaries of and spread restoration of 17. Bainbridge JW, et al. Long-term effect of gene Loci Causing Retinal Diseases. University of and rod photoreceptor rescue throughout therapy on Leber’s congenital amaurosis. N Engl Texas Health Science Center Web site. the retina prevents further cone and rod J Med. 2015;372(20):1887–1897. https://sph.uth.edu/RetNet/sum-dis.htm. degeneration that otherwise would occur 18. Cepko CL, Vandenberghe LH. Retinal gene Updated June 19, 2015. Accessed July 27, therapy coming of age. Hum Gene Ther. by non–cell-autonomous mechanisms. 2015, 2015. 2013;24(3):242–244. 2. Bowes C, Li T, Danciger M, Baxter LC, Apple- 19. Jacobson SG, et al. Improvement and decline in Etiology matters bury ML, Farber DB. Retinal degeneration in the vision with gene therapy in childhood blindness. The ability of induced gene expression to rd mouse is caused by a defect in the beta sub- N Engl J Med. 2015;372(20):1920–1926. unit of rod cGMP-phosphodiesterase. Nature. halt degeneration is important; however, 20. Li X, et al. Gene therapy rescues cone struc- 1990;347(6294):677–680. ture and function in the 3-month-old rd12 this strategy may or may not apply to all 3. McLaughlin ME, Ehrhart TL, Berson EL, Dryja mouse: a model for midcourse RPE65 leber types of retinal degeneration. For example, TP. Mutation spectrum of the gene encoding congenital amaurosis. Invest Ophthalmol Vis Sci. a key difference between degeneration in the beta subunit of rod phosphodiesterase 2011;52(1):7–15. RP, LCA1, and LCA2 is that the defect orig- among patients with autosomal recessive 21. Koch SF, et al. Halting progressive neurodegen- . Proc Natl Acad Sci U S A. inates in photoreceptors in RP and LCA1, eration in advanced retinitis pigmentosa. J Clin 1995;92(8):3249–3253. Invest. 2015;125(9):3704–3713. whereas it originates in the retinal pig- 4. Pittler SJ, Baehr W. Identification of a non- 22. Redmond TM, et al. Rpe65 is necessary for pro- ment epithelium (RPE) in LCA2. Addition- sense mutation in the rod photoreceptor duction of 11-cis-vitamin A in the retinal visual ally, RP- and LCA1-associated mutations cGMP phosphodiesterase beta-subunit gene cycle. Nat Genet. 1998;20(4):344–351.

3426 jci.org Volume 125 Number 9 September 2015