Mutation in the intracellular chloride channel CLCC1 associated with autosomal recessive retinitis pigmentosa

Abstract

We identified a homozygous missense alteration (c.75C>A, p.D25E) in CLCC1, encoding a presumptive intracellular chloride channel highly expressed in the retina, associated with autosomal recessive retinitis pigmentosa (arRP) in eight consanguineous families of Pakistani descent. The p.D25E alteration decreased CLCC1 channel function accompanied by accumulation of mutant protein in granules within the ER lumen, while siRNA knockdown of CLCC1 mRNA induced apoptosis in cultured ARPE-19 cells. TALEN KO in zebrafish was lethal 11 days post fertilization. The depressed electroretinogram (ERG) cone response and cone spectral sensitivity of 5 dpf KO zebrafish and reduced eye size, retinal thickness, and expression of rod and cone opsins could be rescued by injection of wild type CLCC1 mRNA. Clcc1+/- KO mice showed decreased ERGs and photoreceptor number. Together these results strongly suggest that intracellular chloride transport by CLCC1 is a critical process in maintaining retinal integrity, and CLCC1 is crucial for survival and function of retinal cells.

Fig 1. CLCC1 region, sequence, protein domains, and function.

(a) Electropherograms: the sequence of an affected person (top, Individual 27, Family 2), the sequence of a heterozygous carrier (middle, Individual 29, Family 2) and unaffected control sequence (bottom) surrounding the CLCC1 c.75C>A alteration, electropherograms of additional families are similar, but are not shown. (b) Amino acid sequence alignment around the D25 amino acid of CLCC1 (red) in 22 species ranging from human to zebrafish. D25 is absolutely conserved in all species and the entire region is relatively well conserved, especially among mammals. (c) Analysis with the online PSORT algorithm (https://wolfpsort.hgc.jp/) predicts the presence of a signal peptide at the N-terminal of CLCC1 and three transmembrane domains. (d) Co-immunoprecipitation demonstrates interaction of CLCC1 with Calreticulin. (e) Representative current traces of microsomes from WT and mutant cells. The recording voltage was marked above each trace. The mutant shows clear open/close channel activities while the WT does not in this tracing. The same buffer (5 mM Tris, 5 mM MOPS, 150 mM KCl, pH 7.0) was used on both sides. (f) The current-voltage (I/V) plot of microsomes from WT and mutant cells demonstrating decreased channel activity of the p.D25E mutant relative to the WT.

Fig 2. Fundus photographs and clinical findings.

Fundus photographs of Individual 21 of Family 2 (a and b: OD [right eye] and OS [left eye], respectively), Individual 17 of Family 5 (c and d: OD and OS, respectively) and Individual 20 of Family 6 (n and o: OD and OS respectively) revealed findings similar to those seen in Family 1, including obvious bone spicule-shaped pigment deposits in the mid-periphery, waxy-pale optic discs, attenuation of the retinal arteries, and a generalized grayish carpet-like retinal degeneration as compared to a normal fundus (e and f: OD and OS, respectively). Maculopathy was detected in Family 1 and Family 6 but was not observed in Families 2 and 5. (g) Clinical findings of Family 2 and 5; the presenting symptom in both families is reported to be night blindness by approximately 10 years of age. All affected family members have moderate loss of visual acuity. The clinical characteristics of Family 1 were previously published [6]. h) rod (green) and cone (red) response: OD & OS, respectively; and i) 30Hz flicker response: OD & OS, respectively of Individual 21; Family 2. j) rod (green) and cone (red) response: OD & OS, respectively; and k) 30Hz flicker response of Individual 17 of Family 5. l) rod (green) and cone (red) response: OD & OS, respectively; and m) 30Hz flicker response of an age- and ethnically-matched control. The affected individuals demonstrate loss of ERG responses in keeping with advanced RP.

Fig 3. Fundus photographs and optical coherence tomography (OCT) for the affected individual in family 8.

Colour fundus photographs (a and b: OD and OS respectively) revealed attenuated retinal vessels, mid-peripheral coarse pigment clumping and white dots at the level of the retinal pigment epithelium. 55 degree fundus autofluorescence imaging (c and d: OD and OS respectively) show widespread loss of autofluorescence more marked over the pigment clumps in the mid-periphery. Optical coherence tomography (OCT) (e and f: OD and OS respectively) demonstrated loss of outer nuclear and photoreceptor layers throughout the macula with occasional small foci of retained photoreceptors. This individual presented with a history of night blindness from 1–2 years of age and an intermittent divergent squint with long-standing photophobia. Presenting visual acuity was 6/9 in each eye and at last review at age 18 years had deteriorated to 6/60 each eye with severely restricted visual fields to less than 15 degrees to confrontation. There was a myopic astigmatic refractive error of right 0.25/-1.25 x 21° and left 0.25/-2.25 x 160°. Electrophysiology performed at age 7 was unrecordable.

Fig 4. Localization of mutant and WT CLCC1 in ARPE19, Chick RGC, and HEK 293 cells by immunofluorescence.

ARPE19 cells were transfected with pOTB7 expressing WT and p.D25E mutant CLCC1. (a) ER (red), (b) Golgi (red), (c) Lysosome (red), and CLCC1 (green). (d) Chick RGCs were transfected with CLCC1-FLAG (green) and KDEL (red), here CLCC1 shows extensive colocalisation with immunolabelling of the KDEL motif found in ER-associated proteins (e) Chick RGC were transfected with Bip-mCherry-KDEL (red), here the colocalization is even greater (f) HEK 293 cells were transfected with YFP-CLCC1 (green) and stained for Calreticulin (red). Cells were stained with DAPI (blue, nucleus) as well. Overlays of images from the first three columns are shown in panels labeled Merged. Scale Bar: 10 μm. CLCC1 WT and p.D25E are both shown colocalised with calreticulin. Both WT and mutant proteins localize with ER (a-f), and with neither Golgi (b) nor lysosomes (c). In the ARPE19 cells p.D25E mutant CLCC1 was concentrated in granular accumulations in the cell periphery (a-c). (d-e). The association of CLCC1 with the ER is throughout the cell, including the neurites and there is no discernible difference in the colocalisation observed with the WT or p.D25E CLCC1.

Fig 5. CLCC1 siRNA interference in ARPE19 cells.

(a) Western Blot of SiRNA treated ARPE19 cell lysates probed with CLCC1 antibodies. Lane 1, untransfected lysate; lane 2, transfected with RNAi Max transfection reagent only; lanes 3–12, Increasing CLCC1 and control siRNA amounts from 10 pmol (lanes 3,4) to 70 pmol (lanes 11, 12). CLCC1 proteins migrate at the predicted MW of 67 kDa. The blot shows a dose dependent reduction of CLCC1 protein expression in CLCC1 but not control siRNA treated ARPE19 cells to about 20% of normal. (b) TUNEL assay after siRNA transfection. Left: TUNEL-positive apoptotic cells (green), Second: CLCC1 (red), Third: DAPI (blue, nucleus). Although there is some variation in intensity of individual cells, probably based on cell size, shape, and orientation, staining for CLCC1 is lower overall in the CLCC1 siRNA treated cells than the control siRNA, control, or DNase 1 cells, consistent with the Western blot in Fig 5a. About 10% of CLCC1 siRNA transfected cells were apoptotic (arrows) but there is minimal apoptosis in control siRNA or untransfected cells. DNase I treated cells were 100% TUNEL-positive. Overlays of images from the first three columns are shown in the right column labeled Merged. Scale Bar: 20 μm. (c) Down regulation of CLCC1 induced apoptosis in nearly 10% of the cells (*** P<0.0001, t = 14.63) as compared to approximately 1% of cells treated with the control siRNA and less than 1% of untreated cells.

Fig 6. Relative expression of Clcc1 in mouse eye tissues at various ages, distribution of clcc1 mRNA in the zebrafish, and CLCC1 protein in the human retina.

(a) Expression of Clcc1 mRNA in the cornea, lens, iris, optic nerve, and retina by qRT-PCR at different ages. Values represent the mean (± SD) on an arbitrary scale (y axis) and were calculated from at least three independent experiments. While Clcc1 is expressed in all tissues tested, ocular expression is greatest in the retina and least in the lens. (b-f) In situ hybridization of clcc1 probes in zebrafish. clcc1 Is expressed widely in zebrafish. Staining with a digoxigenin-labeled cRNA probe shows a strong signal (black arrows) in the hindbrain (HB), swim bladder (SB), and eye at 1 dpf (b, c), and in the ganglion cell layer (GCL), outer nuclear layer (ONL), and retinal pigmented epithelium (RPE) at 3 dpf (d, e and f); OS. Scale Bar: 100 μm. (g–j) IHC of formalin fixed and paraffin embedded human retinal sections demonstrated CLCC1 is expressed extensively in the retina and optical nerves. High magnification (g,i) shows more intense CLCC1 staining (arrow) in the lamina cribrosa (LC), optic nerve (ON), ganglion cell layer (GCL), inner nuclear layer (INL), outer nuclear layer (ONL) and retinal pigmented epithelia (RPE) in the retina (counter stain is methyl green). Scale Bar: g, h, 50 μm; i, j, 20μm.

Fig 7. Zebrafish eye development disturbed by knockdown of clcc1 expression.

Validation: Injection of the 5’-modified EGFP (a) or the unmodified EGFP (d) gave a fluorescent signal (arrowheads). Co-injection of the clcc1-MO eliminated the fluorescent signal from morpholino-sensitive 5’-modified EGFP mRNA (b) but not unmodified EGFP (e). Co-injection of the MM-MO had no effect (c, f). Eye Size: Injection of the clcc1-MO (h, k) significantly reduced eye size (black arrows) compared to MM-MO (i, l) and buffer-injected (g, j) embryos. a-f: 24 hpf, (g-l): 36 hpf. Retinal frozen sections: from 4 dpf MM-MO- (m, n) and clcc1-MO-injected (o, p) embryos stained for PKCß1 (bipolar cells, green), Zpr-1 (cone receptors, red, n and p), 1D1 (rod receptors, red, m and o), and DAPI (nuclei, blue). clcc1-MO-injected embryos show decreased thickness of ONL and IPL layers. MM-MO-injected (q, s, u, w) and clcc1-MO-injected (r, t, v, x) embryos were stained with anti-blue opsin (q, r, green), anti-green opsin (s, t, green), anti-red opsin (u, v, green), or anti-UV opsin (w, x, green), and 4D2 (all, Rhodopsin, rods, red). All photoreceptors in clcc1-MO-injected embryos show reduced staining and damaged photoreceptor cell structure, with the greatest decreases in blue and green opsin cones. m-x: 4 dpf. Scale Bar: m-p: 50 μm, q-x: 10 μm. Comparison of eye size and retinal layers: (y). GCL = ganglion cell layer, IPL = inner plexiform layer, INL = inner nuclear layer, and ONL = outer nuclear layer. Proportions of embryos with eye size phenotype: (z) with clcc1-MO injection and rescue by coinjected clcc1 WT mRNA. Lens and eye areas are given in mm2, and retinal thicknesses are given in μm. Forty clcc1-MO-treated embryos and 41 MM-MO-treated embryos were analyzed.

Fig 8. Retinal morphology and function is damaged in TALEN clcc1-KO zebrafish.

(a-d) Merged photographs of frozen retinal sections prepared from the heads of 5 dpf larvae. Merged photos of frozen sections from KO (b, d) and WT (a, c) embryos stained for PKCß1 (bipolar cells, green), 4D2 (rod receptors, red, a and b), Zpr-1 (cone receptors, red, c and d), and DAPI (nuclei, blue). clcc1 KO embryos show destruction of the rod photoreceptor layer (b) compared with WT (a). While somewhat better preserved than in the morpholino clcc1 knockdown embryos, cone photoreceptors and the other retinal layers also appear decreased. (e) No background, 20mM sodium aspartate, mutant cone responses are 50% depressed relative to WT. (f) PIII 6dpf Talenclcc1 vs WT. Raw means and SEMs (all spectra). Dark adapted, no background. Cone spectral sensitivity is depressed about 60% in mutants. (g) No background, 50μM CNQX. For signals from ON bipolar cells, which are 2X more sensitive than cone signals, TALEN clcc1 mutant responses decrease by over 50%. Stimuli are saturating at 490nm. (h) b2 5dpf Talenclcc1 vs WT. Raw means and SEMs (all spectra). Dark adapted, no background. ON bipolar cell spectral sensitivity is depressed over 50% in mutants. Sensitivity axis is in units of nV per quantum as calculated from the amplitude of responses to constant quanta stimulation across the spectrum (Eq 1). The quanta level of 2500 hν·μm⁻²·s⁻¹ at the cornea is below semi-saturation for all cone types. (i) coinjection of zebrafish embryos with WT but not p.D25E mutant clcc1 mRNA can rescue the KO phenotype. *p 0.022 vs. WT injected with buffer control, p = 0.0082 vs KO injected with p.D25E mutant clcc1 mRNA, ** p = 0.00095 vs. WT injected with buffer control *** p > 0.00012 vs. WT injected with buffer control.

Fig 9. Effects of heterozygosity for Clcc1 KO on mouse retinas.

Hematoxylin & eosin staining of (a) WT and (b) Clcc1^-/+ knockout 7-month-old mouse retinas. While the overall structure of the retina is preserved, staining revealed decreased cell density in the outer and inner nuclear layers, as well as the outer and inner plexiform layers as well as structural disarray of the photoreceptor layer in Clcc1 KO heterozygous compared with the WT mice. (c-h) Immunostaining of cone arrestin in WT (c-e) and heterozygous Clcc1 knockout (f-h) mice. The WT mice exhibit normal cone photoreceptors staining pattern while Clcc1 heterozygous KO mice revealed reduced number of cone photoreceptors. (i-l) Electroretinography of WT and Clcc1 knockout heterozygous mice show approximately 20–50% decreases in the amplitude of both the scotopic and photopic a and b wave amplitude responses in Clcc1 KO heterozygous mice compared to WT at all levels of luminance.