The P4-phospholipid flippase Atp11a is required for maintenance of eye and ear structure in zebrafish

Abstract

The atp11a gene encodes a phospholipid flippase protein required to flip phosphatidylserine (PS) and phosphatidylethanolamine (PE) from the outer leaflet of the cytoplasmic membrane to the inner leaflet. Mutations in ATP11A have been described in individuals with sensorineural hearing loss and neurological deterioration; however, little is known regarding the mechanism by which loss of atp11a results in such phenotypes. To this end, we created loss-of-function atp11a mutant zebrafish to characterize potential disease states. We demonstrate that mutant atp11a zebrafish display a reduced number of stereocilia in the larval ear and a reduced number of hair cells in some sensory neuromasts, indicating that these fish represent an ideal model for studying atp11a-attributable hearing loss. In addition, atp11a mutant zebrafish raised in a standard light cycle have reduced photoreceptor outer segments, the severity of which is lessened when mutant larvae are raised in the dark. Photoreceptors that do remain in homozygous atp11a mutants undergo mitochondrial fission and produce an increased number of mitochondria, suggesting that defects in energy homeostasis may contribute to or result from outer segment degradation.

Keywords: atp11a; Hair cell; Photoreceptor; Retinal pigment epithelium; Stereocilia; Zebrafish.

Fig. 1.

Generation of atp11a loss-of-function mutants. (A–C) Two mutant strains were isolated using the same gRNA, creating a 7 bp deletion (GTTCTGT) highlighted by the blue bar and a 5 bp deletion (TAAAG) highlighted by the red bar. (D–K) At both 4 and 6 dpf, larvae are morphologically wild type, inflate their swim bladders and have reduced pigment. (L–N) In situ hybridization using an antisense atp11a probe demonstrates reduced transcript levels in heterozygous and homozygous mutant embryos at 24 hpf.

Fig. 2.

Reduced number of stereocilia in atp11a mutants. (A) Phalloidin staining of F-actin filaments in the hair cells of the atp11a^nl1005 zebrafish inner ear at 5 dpf, illustrating significant hair cell loss in the medial crista (MC), posterior crista (PC) and macula (PM), as well as the anterior macula (AM), resulting in an overall decrease in the inner ear. (B) Hair cell quantification for the anterior crista (AC), anterior macular, medial crista, posterior crista and posterior macula demonstrates a reduction in stereocilia in heterozygous and homozygous mutants. Data are mean±s.e.m. with significance testing (*P<0.05, **P<0.01, ***P<0.001) calculated using a one-factor ANOVA with Tukey's post-hoc analysis for multiple comparisons. Wild type, n=17; heterozygous, n=22; homozygous mutant, n=7.

Fig. 3.

Reduced number of hair cells in the O1 neuromast. (A–C) Nuclear staining of four neuromasts in 5 dpf atp11a^nl1007 larvae with Yo-Pro-1, demonstrating a decrease in the number of positively stained cells in the O1 neuromast in homozygous and heterozygous mutants compared with wild type. (D) Neuromast hair cell quantification for one otic (O1) and three supraoptic (SO1, SO2 and SO3) neuromasts demonstrates a significant reduction in hair cell number for the O1 neuromast in homozygous and heterozygous mutants. Data are mean±s.e.m. with significance calculated using a one-factor ANOVA with Tukey's post-hoc analysis for multiple comparisons. Wild type, n=15; heterozygous, n=36; homozygous mutant, n=12.

Fig. 4.

Photoreceptor loss in atp11a^nl1007 mutants is partially light dependent. (A,C,E) Wild-type larvae display regular retinal lamination with a distinct photoreceptor layer. (B,D,F) atp11a^nl1007 larvae raised in ambient light display a reduced photoreceptor layer. (G–L) When raised in the dark, both wild-type (G,I,K) and atp11a^nl1007 larvae (H,J,L) have a more wild-type appearing photoreceptor layer. Wild type light, n=9; homozygous mutant light, n=7; wild type dark, n=8; homozygous mutant dark, n=9. All larvae are at 6 dpf. Scale bars: 100µm in A for A–D,G-J; 50 µm in E for E,F,K,L.

Fig. 5.

Ultrastructure analysis of atp11a^nl1007 mutant photoreceptors. (A–D) Photoreceptors in ultrathin sections show a reduction of outer segments in homozygous mutants when compared to wild-type siblings. (E–H) This phenotype partially resolves when mutant larvae are raised in the dark. Quantification of outer segment number (I), length (J) and mitochondrial number (K) confirm a partial rescue via dark-rearing conditions. Data are mean±s.e.m. ***P<0001, **P<0.001, *P<0.01 (one-factor ANOVA with Tukey’s analysis for multiple comparisons). Wild type light, n=6; mutant light, n=6; wild type dark, n=3; mutant dark, n=3. Larvae are at 6 dpf. Scale bars: 2 µm in F for A,B,E,F; 1 µm in H for C,D,G,H.

Fig. 6.

Cell death in atp11a^nl1007 mutant eyes. (A,B) Few dying cells (Acridine Orange positive, arrows) are observed in wild-type sibling eyes in light and dark conditions. (C,D) In atplla homozygous mutants, an increase in the number of Acridine Orange-positive cells (arrows) is observed in light and dark conditions. (E) Quantification of cell death (number of Acridine Orange-positive cells), presented as mean±s.e.m. Wild type light, n=9; homozygous mutant light, n=9; wild type dark, n=10; homozygous mutant dark, n=11. Significance testing was carried out using two-factor ANOVA. All larvae are at 6 dpf.

Fig. 7.

atp11a mutants have altered swimming behavior. When transitioning from light to dark, wild-type larvae have a stereotypical increase in swimming activity, which is not observed in atp11a^nl1007 mutant larvae. Wild type, n=11; mutant, n=7. Data are mean±s.d. Significance testing was carried out using a one-factor ANOVA with Tukey's analysis of multiple comparisons. Analysis was performed at 7 dpf.