Disruption of PIKFYVE causes congenital cataract in human and zebrafish

Abstract

Congenital cataract, an ocular disease predominantly occurring within the first decade of life, is one of the leading causes of blindness in children. However, the molecular mechanisms underlying the pathogenesis of congenital cataract remain incompletely defined. Through whole-exome sequencing of a Chinese family with congenital cataract, we identified a potential pathological variant (p.G1943E) in PIKFYVE, which is located in the PIP kinase domain of the PIKFYVE protein. We demonstrated that heterozygous/homozygous disruption of PIKFYVE kinase domain, instead of overexpression of PIKFYVE^G1943E in zebrafish mimicked the cataract defect in human patients, suggesting that haploinsufficiency, rather than dominant-negative inhibition of PIKFYVE activity caused the disease. Phenotypical analysis of pikfyve zebrafish mutants revealed that loss of Pikfyve caused aberrant vacuolation (accumulation of Rab7⁺Lc3⁺ amphisomes) in lens cells, which was significantly alleviated by treatment with the V-ATPase inhibitor bafilomycin A1 (Baf-A1). Collectively, we identified PIKFYVE as a novel causative gene for congenital cataract and pinpointed the potential application of Baf-A1 for the treatment of congenital cataract caused by PIKFYVE deficiency.

Keywords: Baf-A1; PIKFYVE; congenital cataract; endosome; gene; genetics; genomics; mutation; zebrafish.

Figure 1.. Pedigree structure and ocular manifestations of the cataract family.

(A) Pedigree of the family with congenital cataract. Squares denote males and circles denote females; Symbols crossed by a line indicate deceased individuals. Filled symbols indicate affected individuals, while open symbols indicate unaffected individuals. All affected family members had bilateral congenital cataract. The arrow denotes the proband. The individuals marked with an asterisk (∗) are analyzed by whole-exome sequencing. Genotypes of the PIKFYVE variant (p.G1943E) are indicated below each symbol (+, wild-type allele; −, p.G1943E variant allele). (B) Slit-lamp photographs showing the transparent lens of an unaffected individual (IV-6) and the nuclear pulverulent cataract in the left eye of the proband (III-9).

Figure 2.. The PIKFYVE variant identified from the congenital cataract family.

(A) Sanger sequencing chromatogram showing the cDNA sequences from a healthy control and a cataract patient. The heterozygous c.5828G>A missense variant in the patient is indicated by the red arrow. (B) A schematic diagram showing the human PIKFYVE domains. The p.G1943E variant in the PIPK domain is indicated by the red arrow. (C) Protein sequence alignment of PIKFYVE orthologs in vertebrates. The black triangle denotes the conserved glycine at position 1943. (D) Western blot analysis of PIKFYVE^WT and PIKFYVE^G1943E expression in HEK293T cells that were transiently transfected with either pCS2(+)-CMV-PIKFYVE^WT or pCS2(+)-CMV-PIKFYVE^G1943E. The protein levels were normalized by GAPDH expression. Experiments were repeated three times. (E) Predicted structure model of the p.G1943E variant form of PIKFYVE PIPK domain generated by the PHYPRE2 server (http://www.sbg.bio.ic.ac.uk/~phyre2/html/). N-lobe, C-lobe, and the hinge linker are shown in gold, cyan, and red, respectively. The variant residue E1943 is shown in sticks and labeled with green. The negatively charged residue D1872 close to E1943 side chain is also shown in sticks. (F) A schematic demonstrating the organization of PIKFYVE PIPK domain. N-lobe, C-lobe, and the hinge linker are shown in gold, cyan, and red, respectively. The position of the p.G1943E variant is labeled with a yellow star. (G) Surface electrostatic potential comparison of the PIPK domain of PIKFYVE between wild-type (WT) and p.G1943E variant. The electrostatic potentials are presented as heatmaps from red to blue, and the electrostatic potential scales are shown in the lower panel. See Figure 2—source data 1 for details.

Figure 2—figure supplement 1.. Expression of PIKFYVE in human lens capsule.

(A) RT-PCR results of lens capsules from three individuals. NC represents negative control without cDNA. (B) Quantitative RT-PCR shows the relative expression of PIKFYVE compared with GAPDH.

Figure 2—figure supplement 2.. A schematic diagram showing the distribution of PIKFYVE variants.

(A) The genomic loci of identified variants of PIKFYVE from sporadic cataract patients. (B) The amino acid changes of identified variants and their positions in PIKFYVE protein.

Figure 3.. Disruption of the PIPK domain of Pikfyve in zebrafish caused early-onset cataract.

(A) A schematic diagram showing the generated pikfyve^Δ8 mutant allele. The underlined base pairs are the sgRNA target. The deleted base pairs are shown in dark blue while inserted ones are shown in red. The stop codon introduced in the mutant form is shown in the grey box. (B) Representative images showing the lens of sibling and pikfyve^Δ8 mutants at 5 dpf. (C) Representative differential interference contrast (DIC) images showing the lens of pikfyve^+/+, pikfyve^+/Δ8, and pikfyve^Δ8/Δ8 embryos at 3 dpf and 5 dpf. The scale bars represent 10 μm in (B) and (C). (D) Quantification of vacuole number in the lens of pikfyve^+/+, pikfyve^+/Δ8, and pikfyve^Δ8/Δ8 embryos at 3 dpf (n=7 for pikfyve^+/+; n=10 for pikfyve^+/Δ8; n=7 for pikfyve^Δ8/Δ8) and 5 dpf (n=7 for pikfyve^+/+; n=11 for pikfyve^+/Δ8; n=6 for pikfyve^Δ8/Δ8). (E) Representative images of HLEB3 cells treated with DMSO or PIKFYVE inhibitor YM201636 for 4 hr. The scale bars represent 25 μm. (F) Quantification of the vacuole numbers in (E). ****, p<0.0001, Student’s t-test. All experiments were repeated three times. See Figure 3—source data 1 for details.

Figure 3—figure supplement 1.. Characterization of pikfyve-deficient zebrafish mutants.

(A) Gross morphology of 5-dpf and 7-dpf siblings and pikfyve^Δ8 mutants. The scale bars represent 500 μm. (B) Survival rate of pikfyve^Δ8 mutants (n=48) and siblings (n=50).

Figure 3—figure supplement 2.. Ectopic overexpression of PIKFYVE^G1943E failed to induce cataract defect in zebrafish.

(A) Schematic view of constructs in which the WT and G1943E variant form of PIKFYVE were expressed under the control of the ubiquitously expressed ubiquitin promoter. (B) Representative images showing the lens of 5-dpf WT, Tg(ubi:PIKFYVEE^WT) and Tg(ubi:PIKFYVE^G1943E) zebrafish embryos. The scale bars represent 20 μm. All experiments were repeated three times.

Figure 3—figure supplement 3.. The G1943E variant form of PIKFYVE is less efficient to rescue the vacuole defect in pikfyve-deficient zebrafish mutants.

(A) Representative images showing the lens of 5-dpf siblings, pikfyve^Δ8, pikfyve^Δ8;Tg(ubi:PIKFYVE^WT), and pikfyve^Δ8;Tg(ubi:PIKFYVE^G1943E) zebrafish. The scale bars represent 20 μm. (B) Whole-mount in situ hybridization (WISH) analysis of human PIKFYVE transcripts in 2-dpf Tg(ubi:PIKFYVE^WT) and Tg(ubi:PIKFYVE^G1943E) zebrafish. The scale bars represent 200 μm. See Figure 3—figure supplement 3—source data 1 for details.

Figure 4.. Detailed characterization of cataract phenotypes in pikfyve^Δ8 mutants.

(A) Confocal imaging of the lens of 5-dpf siblings and pikfyve^Δ8 mutants in Tg(cryaa:DsRed) transgenic background. (B) Hematoxylin-eosin (HE) staining of 5-dpf siblings and pikfyve^Δ8 mutant zebrafish lens after cryostat section. (C) ZL-1 antibody and DAPI staining of 5-dpf siblings and pikfyve^Δ8 mutant zebrafish lens. (D) Transmission electron microscope (TEM) images of the lens of siblings and pikfyve^Δ8 mutants at 3 dpf and 5 dpf. ASS, autophagy lysosome; LD, lipid droplet; N, nucleus. All results were confirmed in three different individuals. All the scale bars represent 10 μm.

Figure 5.. Characterization of vacuoles in pikfyve^Δ8 mutants.

(A) Time-lapse imaging indicating the dynamic changes of vacuole formation in the lens of 4-dpf pikfyve^Δ8 mutants. White arrows indicate the fusion process of two small vacuoles. (B) Representative images showing the lens of 3.5-dpf siblings and pikfyve^Δ8 mutants injected with gfp-rab5c mRNA. (C) Representative images showing lens of 3.5-dpf siblings and pikfyve^Δ8 mutants injected with gfp-rab7 mRNA. (D) Representative images showing lens of 3.5-dpf siblings and pikfyve^Δ8 mutants injected with gfp-rab11a mRNA. (E) Representative images showing lens of 3.5-dpf siblings and pikfyve^Δ8 mutants injected with mcherry-lc3b mRNA. All experiments were repeated three times. All the scale bars represent 10 μm.

Figure 5—figure supplement 1.. Characterization of lysosomes in microglia and lens of pikfyve^Δ8 mutant zebrafish.

Confocal images of the lens and microglia of 4-dpf pikfyve^Δ8 mutants in Tg(mpeg1:dsredx) background after LysoSensor staining. The results were confirmed in more than three different individuals. The scale bars represent 20 μm.

Figure 6.. Baf-A1 partially rescued the vacuole defect in the lens of pikfyve^Δ8 mutant zebrafish.

(A) Representative confocal images of the lens of 4-dpf pikfyve^Δ8 mutants treated with DMSO or Baf-A1 for 4.5 hr. (B) Quantification of the vacuole numbers in the lens of 4-dpf siblings and pikfyve^Δ8 mutant embryos treated with DMSO or Baf-A1 (n=10 for sibling groups; n=7 for mutant groups). (C) Confocal images of the lens of 4-dpf pikfyve^Δ8 mutants with no treatment or after 4.5 hr treatment with DMSO or Baf-A1. (D) Quantification of the vacuole numbers in (C) (n=6 for each group). All experiments were repeated three times. All the scale bars represent 20 μm. *, p<0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001, Student’s t-test. See Figure 6—source data 1 for details.

Figure 6—figure supplement 1.. Overexpression of trpml1 failed to rescue the lens defects in pikfyve ^Δ8 mutants.

(A) Confocal images of the lens of 4-dpf pikfyve^Δ8 mutants and pikfyve^Δ8 mutants injected with trpml1 mRNA. The scale bars represent 20 μm. (B) Quantification of the vacuole number in the lens of 4-dpf pikfyve^Δ8 mutants and pikfyve^Δ8 mutants injected with trpml1 mRNA. n.s., not significant, p>0.05, Student’s t-test. See Figure 6—figure supplement 1—source data 1 for details.