Transgenic mice carrying the H258N mutation in the gene encoding the beta-subunit of phosphodiesterase-6 (PDE6B) provide a model for human congenital stationary night blindness

Abstract

Mutations in the beta-subunit of cGMP-phosphodiesterase (PDE6beta) can lead to either progressive retinal disease, such as human retinitis pigmentosa (RP), or stationary disease, such as congenital stationary night blindness (CSNB). Individuals with CSNB in the Rambusch pedigree were found to carry the H258N allele of PDE6B (MIM# 180072); a similar mutation was not found in RP patients. This report describes an individual carrying the H258N allele, who presented with generalized retinal dysfunction affecting the rod system and a locus of dysfunction at the rod-bipolar interface. Also described are preclinical studies in which transgenic mice with the H258N allele were generated to study the pathophysiological mechanisms of CSNB. While Pde6b(rd1)/Pde6b(rd1) mice have severe photoreceptor degeneration, as in human RP, the H258N transgene rescued these cells. The cGMP-PDE6 activity of dark-adapted H258N mice showed an approximate three-fold increase in the rate of retinal cGMP hydrolysis: from 130.1 nmol x min(-1) x nmol(-1) rhodopsin in wild-type controls to 319.2 nmol x min(-1) x nmol(-1) rhodopsin in mutants, consistent with the hypothesis that inhibition of the PDE6beta activity by the regulatory PDE6gamma subunit is blocked by this mutation. In the albino (B6CBA x FVB) F2 hybrid background, electroretinograms (ERG) from H258N mice were similar to those obtained from affected Rambusch family members, as well as humans with the most common form of CSNB (X-linked), demonstrating a selective loss of the b-wave with relatively normal a-waves. When the H258N allele was introduced into the DBA background, there was no evidence of selective reduction in b-wave amplitudes; rather a- and b-wave amplitudes were both reduced. Thus, factors other than the PDE6B mutation itself could contribute to the variance of an electrophysiological response. Therefore, caution is advisable when interpreting physiological phenotypes associated with the same allele on different genetic backgrounds. Nevertheless, such animals should be of considerable value in further studies of the molecular pathology of CSNB.

FIGURE 1

Representative ERG recordings from a patient (top row) and a normal control (bottom row). Shown from left to right are a rod-isolated response (A), the dark-adapted maximal response (B), the light-adapted responses to a single flash (1 Hz) (C) and flickering (30 Hz) stimuli (D). a- and b-waves are identified using conventional techniques. Note the barely detectable rod-isolated response (A) and the selective loss of the b-wave for the dark-adapted maximal response in the patient’s data (B) compared to the normal control.

FIGURE 2

Immunoblot analysis of PDE6 subunit expression in control and H258N ROSs. Left panel shows an immunoblot incubated with a polyclonal antibody recognizing both rod PDE6α, PDE6β and cone PDE6α′. Right panel is an immunoblot incubated with a polyclonal antibody specific for rod PDE6β. Protein in all lanes was normalized to 150 pmol rhodopsin. Lane 1, H258N; Lane 2, Pde6b^rd1/+; Lane 3, +/+ control.

FIGURE 3

The H258N transgene completely rescues photoreceptors in Pde6b^rd1/Pde6b^rd1 mice. Light micrographs of retina from a 6-month-old H258N mouse (A) and a 3-month-old homozygote Pde6b^rd1/Pde6b^rd1 (B) in DBA background. OS, outer segments; ONL, outer nuclear layer; INL, inner nuclear layer; GC, ganglion cell layer. OS and ONL comprise the photoreceptor layer. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.)

FIGURE 4

Electron micrograph of retinal section from a 3-month-old H258N mouse. The photoreceptor OSs are normal. Pde6b^rd1/Pde6b^rd1 mice do not have photoreceptors at this age.

FIGURE 5

A: Representative panretinal ERG photoresponses from H258N, (upper traces) and Pde6b^rd1/+ control mice (lower traces) on the (B6CBA × FVB) F2 albino background. The dashed lines through each dataset represent the fit of a rod model used to derive Rm_p3, the saturated photoreceptor response, and S, photoreceptor sensitivity. B: Saturated b-wave amplitudes (V_max) across age for H258N (open circles) and wildtype (+/+) control mice (filled circles). C: Rm_p3 vs. V_max amplitudes for H258N (open circles), Pde6b^rd1/+ (filled squares) and +/+ control mice (filled circles). The vertical dashed lines are the standard error bars forV_max for the +/+ control mice. Note the shift in V_max amplitudes to the left, indicating reduced amplitudes for the H258N mice. D–F: Representative ERGs from Pde6b^rd1/+ control mice (D), and two H258N mutant mice from independent founders in (B6CBA × FVB) F2 albino background(E,F).

FIGURE 6

Representative panretinal ERGs from DBA +/+ control mice (left traces) and H258N mice (right traces) in the DBA background at 6 months of age.

FIGURE 7

Responses of H258N and control rods to light. A: C57BL/6J control rod, 20msec flashes of 500-nm light at flash intensities of 17, 43, 160, 450, 1,120, and 4,230 photons μm ⁻². The traces are averages of two to four flashes at each intensity. B: Typical responses from a H258N rod to the same flash intensities. Each trace was averaged from three to eight flashes. C: Response amplitude vs. flash intensity averaged from 34 wild-type (WT; ○) and H258N rods (H258N; X). (The lowest intensity response [smallest response] was obtained by averaging 60 flashes, the others were averaged using five flashes.) D: Averaged single-photon responses from H258N (thin line; n = 18) rods superimposed on averaged single-photon response from control rods (thick line; n = 48). Single-photon responses were calculated for each rod individually from 15 to 60 dim flashes with the squared mean-variance method.