Zebrafish model of CEP develops bone phenotype resembling human disease. (A) 6dpf zebrafish larvae were injected with uro-I or vehicle and imaged by confocal microscopy at 7dpf. Porphyrin was detected only in the bones of uro-I-injected group. Arrowhead-operculum; box-vertebrae. (B) Larvae were treated as in (A) and injected with calcein prior to imaging. Arrowhead-operculum; arrow-4^th vertebra. (C) Quantification of bone volume in larvae from (B); bone volume was normalized to vehicle-injected larvae set to 100%. Symbols represent individual larvae (14–18/group) from 4–5 independent experiments. (D) Larvae were treated as in (A). At 7dpf bones were harvested and imaged by epifluorescence microscopy pre and post HCl bone demineralization, arrow-notochord. (E) Hydroxyapatite was incubated with calcein/uro-I/copro-I and imaged by epifluorescence microscopy. Scale bars: 200 µm (A-D); 50 µm (E). Three-dimensional image reconstruction (A, B) was performed using Imaris 3D software v7.7 (http://imaris.oxinst.com/). **p < 0.01.

Acitretin mitigates CEP bone phenotype in zebrafish. (A) 6dpf larvae were injected with uro-I and transferred to medium containing acitretin or DMSO. At 7dpf larvae were injected with calcein and imaged by confocal microscopy. Quantification of porphyrin fluorescence (B), porphyrin excretion (C) and operculum volume (D) from experiment in (A). Symbols represent individual larvae (12–25/group) from 3–4 independent experiments. (E) 6dpf larvae were injected with uro-I. At 7dpf they were transferred to medium containing acitretin or DMSO. At 8dpf larvae were injected with calcein and imaged by confocal microscopy. Quantification of porphyrin fluorescence (F), porphyrin excretion (G) and operculum volume (H) from experiment in (E). Arrowhead-operculum; box-vertebrae (A, E). Symbols represent individual larvae (18–64/group) from 3–4 independent experiments. (I-K) Larvae were treated as in (A) with the indicated retinoid or DMSO and porphyrin fluorescence (I), porphyrin excretion (J) and operculum volume (K) were assayed. Bone volume was normalized to DMSO-treated larvae set to 100%, (D, H, K). Porphyrin excretion was normalized to DMSO-treated larvae set to 100%, (C, G, J). Symbols represent individual larvae (7–44/group) from 2–4 independent experiments. Scale bars: 200 µm. Three-dimensional image reconstruction (A, E) was performed using Imaris 3D software v7.7 (http://imaris.oxinst.com/). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Saos-2 cells mimic CEP zebrafish model. (A, B) Mineralization in Saos-2 cells treated with MAC ± Uro-I was assayed using ARS staining (photograph, A; quantification, B). Staining was normalized to MAC only-treated cells (set to 100%). (C) Cell lysates from experiment in (A) were blotted with the indicated antibodies. (D) Quantification of LC3-II shown in (C). LC3-II level was normalized to MAC only (left panel) or vehicle-treated (right panel), set to 100%. (E) RT² Profiler PCR Array (left panel) and qPCR (right panel). Relative gene expression is represented as fold change normalized to housekeeping gene. Data are from 2 independent experiments. (F) Acitretin does not rescue reduced mineral matrix phenotype in uro-I-treated cells. ARS staining quantification as in (B). (G) Acitretin normalizes ER stress (BiP) and autophagy (LC3-II) markers. Dashed lines represent non-adjacent lanes in the gel. Coomassie-stained gel (C,G) shows equal protein loading. Full-length blots and gels (C,G) are presented in Fig.S4. (H) Quantification of LC3-II. LC3-II level was normalized to DMSO-treated cells set to 100%. (I) Gene expression profiling as in (E). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Proposed model of CEP pathogenesis. UROS inhibition leads to production of uro/copro-I mostly in erythrocytes and liver, which is transported through blood to the bones. Uro-I causes bone damage by binding to hydroxyapatite, causing oxidative and ER stress, protein aggregation and stalled autophagy. Acitretin partially rescues uro-I-induced bone damage by reducing oxidative and ER stress and restoring autophagic flux.