Lysosomal Storage and Albinism Due to Effects of a De Novo CLCN7 Variant on Lysosomal Acidification

Abstract

Optimal lysosome function requires maintenance of an acidic pH maintained by proton pumps in combination with a counterion transporter such as the Cl^-/H⁺ exchanger, CLCN7 (ClC-7), encoded by CLCN7. The role of ClC-7 in maintaining lysosomal pH has been controversial. In this paper, we performed clinical and genetic evaluations of two children of different ethnicities. Both children had delayed myelination and development, organomegaly, and hypopigmentation, but neither had osteopetrosis. Whole-exome and -genome sequencing revealed a de novo c.2144A>G variant in CLCN7 in both affected children. This p.Tyr715Cys variant, located in the C-terminal domain of ClC-7, resulted in increased outward currents when it was heterologously expressed in Xenopus oocytes. Fibroblasts from probands displayed a lysosomal pH approximately 0.2 units lower than that of control cells, and treatment with chloroquine normalized the pH. Primary fibroblasts from both probands also exhibited markedly enlarged intracellular vacuoles; this finding was recapitulated by the overexpression of human p.Tyr715Cys CLCN7 in control fibroblasts, reflecting the dominant, gain-of-function nature of the variant. A mouse harboring the knock-in Clcn7 variant exhibited hypopigmentation, hepatomegaly resulting from abnormal storage, and enlarged vacuoles in cultured fibroblasts. Our results show that p.Tyr715Cys is a gain-of-function CLCN7 variant associated with developmental delay, organomegaly, and hypopigmentation resulting from lysosomal hyperacidity, abnormal storage, and enlarged intracellular vacuoles. Our data supports the hypothesis that the ClC-7 antiporter plays a critical role in maintaining lysosomal pH.

Keywords: ClC-7 antiporter; chloroquine; cutaneous albinism; lysosomal hyperacidity; lysosomal membrane counterion; lysosomal pH; lysosomal storage disease; oculocutaneous albinism.

Figure 1

Clinical and Histological Features of Probands (A) Proband 1 has hypopigmented skin and hair but normally pigmented irides. (B) Proband 2, with hypopigmentation of skin and hair compared with those of his African parents. Also note the normally pigmented irides. (C and D) Abdominal ultrasound of proband 1 (C); the left (i) and right (ii) kidneys are enlarged to more than 10 cm in length. Abdominal ultrasound of proband 2 (D); the left (i) and right (ii) kidneys are enlarged to more than 10 cm in length. For comparison, please see normal images in Figure S1B. (E and F) Brain MRIs of proband 1 (E) at 20 months and proband 2 (F) at 12 months. Both MRIs show delayed myelination, a thin posterior corpus callosum, hyperintensity of the subthalamic nuclei, and cerebellar atrophy. Proband 1 had a thin posterior corpus callosum, hyperintensity of the brainstem, particularly in the subthalamic nuclei, and atrophy of the cerebellar vermis and hemispheres. Proband 2 had prominent subarachnoid spaces and abnormal signal intensity in the cerebral peduncles, pons, and brainstem, but no leukodystrophy. Please refer to Figure S1 and to the studies of Sanchez et al. for MRI images in control individuals. (G) Transmission electron micrographs from proband 1 show abnormal cytoplasmic inclusions (arrowhead) sometimes containing amorphous floccular storage material (double arrowhead) in the macrophages of the liver and histiocytes of the duodenum. These findings are characteristic of those seen in lysosomal-storage disorders. Similar cytoplasmic inclusions were found in the skin of proband 1. The scale bar represents 1 μm. (H) Transmission electron micrograph showing storage in the kidney of proband 2. There are numerous interstitial macrophages containing abundant cytoplasmic inclusions (arrowhead), highly suggestive of a lysosomal-storage disorder. The scale bar represents 5 μm. (I) Transmission electron microscopy of control melanocytes isolated from a skin punch biopsy. Melanosomes mature from stage I through stage IV as they acquire homogeneous melanin pigment. The scale bar represents 1,000 nm. (J) Transmission electron microscopy of melanocytes derived from the proband 1. Very few stage IV melanosomes are present; most are stage II. Also note that the unusual non-homogeneous distribution of pigment in the melanosomesgives a “chunky” appearance. The scale bar represents 1,000 nm.

Figure 2

Cellular pH and Chloride Transport Studies (A) ClC-7 p.Tyr715Cys expression causes increased current in Xenopus oocytes. Two-electrode voltage-clamp current traces are shown for uninjected (black), WT ClC-7_PM (blue), and p.Tyr715Cys ClC-7_PM (red) in response to a series of 1 s voltage steps from −100 mV to + 80 mV in increments of 20 mV, followed by a step to −80 mV for 1.4 s before return to a holding potential of −30 mV. (B) Bar graphs of current (+/− SEM, measured at +80 mV) or surface expression of uninjected (black), WT ClC-7_PM, or p.Tyr715Cys ClC-7_PM. For current measurements, n = 3 (uninjected), 8 (WT), or 6 (p.Tyr715Cys) from a single batch of oocytes; the experiment was reproduced 11 times in separate oocyte batches with essentially identical results. For surface expression, five separate surface biotinylation experiments were performed (with 20 each of WT and p.Tyr715Cys oocytes); intensities were normalized by the WT band intensity and averaged. The currents shown in (A) were from one of the same batches of oocytes used for the surface-expression experiments. (C) Lysosomal pH in control and proband fibroblasts was measured with OG488 ratiometric imaging. Raw fluorescence ratios are shown for individual cells (pale circles) and averages (+/− SEM) of n = ∼50 cells (dark circles) from two independent experiments each on control neonatal fibroblasts (1.54 ± 0.1) or primary fibroblasts from either proband (proband 1, 1.30 ± 0.07; proband 2, 1.28 ± 0.08). Ratios from bafilomycin-treated cells from each are also shown as a negative control. The lower the 490/440nm fluorescence ratio, the lower the pH. An image of an OG488-treated proband cell is shown in the inset. (D) Averaged lysosomal pH obtained from the experiments shown in part (C) (control, 4.44 ± 0.03; proband 1, 4.26 ± 0.03; proband 2, 4.19 ± 0.06). Separate calibrations were performed for each experiment and used for converting the ratios to pH values (Figure S3). Error bars indicate SEM; p values as shown. (E) Cultured fibroblasts incubated with LysoTracker Red DND-99. Fluorescent intensity was significantly greater in cells of probands 1 and 2 than in control cells. Nuclei are stained with Hoechst (blue channel). (F) Mean (+/− SEM) integrated intensity of LysoTracker Red DND-99 (red channel) per cell is plotted in relative fluorescence units (RFU); values are 10⁷. Nuclei are stained with Hoechst (blue channel). Fluorescent intensity was significantly greater in cells of probands 1 and 2 than in control cells. Abbreviations are as follows: SEM, standard error the mean; WT, wild-type.

Figure 3

Cellular Phenotype and Lysosomal Staining (A) Representative bright-field images of control, proband 1, and proband 2 primary cultured fibroblasts, with large cytoplasmic inclusions in probands’ fibroblasts. The scale bar represents 100 μm. (B) Lamp1 staining highlights the endolysosomal character of the vacuolar structures in proband 1 and 2 fibroblasts compared to WT control, in which distinct punctate lysosomes can be identified. Lamp-1 staining surrounds some part of the vacuoles of the mutant cells. Phalloidin stains actin; DAPI stains the nucleus. i, magnified image of control Lamp1 staining; ii and iii, magnified Lamp 1 staining of cells from proband 1. In WT cells, distinct lysosomes can be identified. In proband 1, some vacuoles are clearly decorated with Lamp1 (closed arrows), whereas others have less defined membranes (open arrows). The scale bar represents 25 μm. (C) Overexpression of human CLCN7 (ClC-7) p.Tyr715Cys in fibroblasts recapitulates the cellular phenotype of cells from probands. pEGFP1-c1 transfection, which does not contain ClC-7, in control fibroblasts is shown to show efficiency of transfection. GFP protein expression was detected via a 488 channel. Control fibroblasts overexpressing a Myc-DDK-tagged ClC-7–WT allele exhibited a mottled staining pattern. Fibroblasts overexpressing Myc-DDK-tagged ClC-7 p.Tyr715Cys displayed distinct, enlarged vacuoles, like those seen in in ClC-7 p.Tyr715Cys proband fibroblasts. Protein expression in both the WT and p.Tyr715Cys cells were detected with a Myc antibody. Hoechst stains the nucleus. The scale bar represents 25 μm. WT indicates wild-type, referring to cells overexpressing the full-length human CLCN7 (ClC-7).

Figure 4

Mouse Model of Clcn7 p.Tyr713Cys (A) Injection of a single-guide RNA targeting exon 23 of Clcn7 in mice generated the variant of interest. p.Tyr715Cys in humans corresponds to p.Tyr713Cys in mice (amino acid Tyr715 is highly conserved in several species up to C. elegans). F1 mice resulting from the injection, containing the heterozygous knock-in variant in C57/B6J background, reveal marked pigmentation defects, including white fur. Also shown is a WT mouse that is from the same C57/B6J background and has normal black fur. (B) Periodic-acid Schiff (PAS) staining of paraffin-embedded slides of liver, spleen, and kidney tissue from WT and Clcn7^Y713C/+ mice, showing increased vacuolar structures in the Clcn7^Y713C/+ mouse, indicating lysosomal storage. The scale bar represents 50μm. (C) Characterization of dermal fibroblasts. Bright-field pictures of cultured mouse fibroblasts from WT (left, +/+) and mutant (right, Y713C/+) mice. The knock-in mouse fibroblasts contain enlarged vacuoles, resembling the p.Tyr715Cys human fibroblasts. (D) X-rays of WT and knock-in Clcn7^Y713C/+ mice at 5 weeks of age, showing normal skeletal and craniofacial bones in both mice. Long bones of Clcn7^Y713C/+ mouse show no evidence of osteopetrosis. (E) Tibiae from WT and Clcn7^Y713C/+ mice stained with von Kossa. These sections reveal no osteopetrosis, an intact marrow cavity, and normally organized trabecular and cortical structure in both WT and mutant mice. (F and G) Luxol fast blue-hematoxylin-eosin stain of the brains from control (+/+) (F) and Clcn7^Y713C/+ (Y713C/+) (G) mice. The blue color represents myelin formation in the corpus callosum. In Clcn7^Y713C/+mice, note the presence of vacuoles in the substance of the midbrain, disrupting the organization of the corpus callosum and parts of brain white matter. The scale bar represents 50μm. Y713C/+ refers to Clcn7^Y713C/+, a mouse heterozygous for the knock-in allele; +/+ refers to a WT mouse with the same background.

Figure 5

Dermal Fibroblasts from Clcn7^Y713C/+ Mice Exhibit Enlarged Vacuoles Partially Stained with a Lysosomal Marker (A) Immunostaining of dermal fibroblasts isolated from WT (left, +/+) and Clcn7^Y713C/+ (right, Y713C/+) mice. Lamp1 (green) decorates the outline of the vacuoles, similarly to what is seen in probands’ cells. The scale bar represents 5 μm. Nuclei are stained with DAPI (blue). (B) Magnified image of mutant (Y713C/+) cells showing that the vacuoles are surrounded by LAMP1 staining. Y713C/+ refers to Clcn7^Y713C/+, a mouse heterozygous for the knock-in allele; +/+ refers to a WT mouse with the same background.

Figure 6

Effects of Chloroquine Treatment on pH and Vacuole Size in CLCN7 p.Tyr715Cys Fibroblasts (A) Dose-dependent changes in pH of proband lysosomes treated with varying doses of chloroquine, as shown. For each drug concentration, pale symbols reflect average lysosomal pH for individual cells, and dark symbols show average pH over the entire cell population (error bars indicate ± SEM). p values: for untreated versus 50 nM, 0.0000145; for untreated versus 100 nM, 0.00000000123; for untreated versus 300 nM, 0.00000000173. (B) Representative images of proband 1 and proband 2 fibroblasts treated for 24 hours with media supplemented (at indicated concentrations) with chloroquine, then fixed and treated with CellMask Red HCA so that the cytoplasm would be stained and changes in the vacuolar composition of these cells would be highlighted. Scale bars represent 30 μM.