Functional analysis of the zebrafish ortholog of HMGCS1 reveals independent functions for cholesterol and isoprenoids in craniofacial development

Abstract

There are 8 different human syndromes caused by mutations in the cholesterol synthesis pathway. A subset of these disorders such as Smith-Lemli-Opitz disorder, are associated with facial dysmorphia. However, the molecular and cellular mechanisms underlying such facial deficits are not fully understood, primarily because of the diverse functions associated with the cholesterol synthesis pathway. Recent evidence has demonstrated that mutation of the zebrafish ortholog of HMGCR results in orofacial clefts. Here we sought to expand upon these data, by deciphering the cholesterol dependent functions of the cholesterol synthesis pathway from the cholesterol independent functions. Moreover, we utilized loss of function analysis and pharmacological inhibition to determine the extent of sonic hedgehog (Shh) signaling in animals with aberrant cholesterol and/or isoprenoid synthesis. Our analysis confirmed that mutation of hmgcs1, which encodes the first enzyme in the cholesterol synthesis pathway, results in craniofacial abnormalities via defects in cranial neural crest cell differentiation. Furthermore targeted pharmacological inhibition of the cholesterol synthesis pathway revealed a novel function for isoprenoid synthesis during vertebrate craniofacial development. Mutation of hmgcs1 had no effect on Shh signaling at 2 and 3 days post fertilization (dpf), but did result in a decrease in the expression of gli1, a known Shh target gene, at 4 dpf, after morphological deficits in craniofacial development and chondrocyte differentiation were observed in hmgcs1 mutants. These data raise the possibility that deficiencies in cholesterol modulate chondrocyte differentiation by a combination of Shh independent and Shh dependent mechanisms. Moreover, our results describe a novel function for isoprenoids in facial development and collectively suggest that cholesterol regulates craniofacial development through versatile mechanisms.

Fig 1. Mutation of hmgcs1 causes craniofacial abnormalities.

Alcian staining was performed to visualize the developing cartilage in hmgcs1 wildtype or heterozygous (Sibling) or homozygous hmgcs1 mutant larvae (Mutant). Embryos were stained at 4 days post fertilization and manual dissection of the viscerocranium and neurocranium was performed. Viscerocranium is depicted in (A) and (B) and neurocranium is depicted in (A’) and (B’). Asterisk denotes basihyal. n = 38 including 8 wildtype embryos, 6 homozygous mutants, and 24 heterozygous mutants. No significant defects were present in heterozygous carriers (S1 Fig).

Fig 2. hmgcs1 does not regulate neural crest cell specification or migration.

(A&B) In situ hybridization to detect sox10 expression was performed in hmgcs1 homozygous mutants or siblings at the 18 somite stage. Dorsal view with anterior to the left. (n = 15 including 8 wildtype/heterozygous siblings and 7 homozygous mutants). Wildtype Sibling is depicted in A, but no significant differences were observed between wildtype and heterozygous carriers. Arrowheads indicated two positive streams of Sox10+ positive cells. (C&D) Sagittal view of wildtype hmgcs1 siblings (hmgcs1 Sibling) or hmgcs1 homozygous mutant larvae (hmgcs1 MT). Larvae were maintained in a Tg(sox10:tagRFP) background and RFP was visualized using traditional microscopy at 1 day post fertilization (dpf) between Prim-5 and Prim-15. n = 20 including 15 wildtype and heterozygous individuals and 5 homozygous mutant individuals. Sibling depicted is a wildtype individual. No significant difference between wildtype embryos was detected.

Fig 3. hmgcs1 modulates cranial neural crest cell differentiation.

(A&B) Whole mount in situ hybridization (ISH) was performed at 1 day post fertilization (dpf) between Prim-5 and Prim- 15 on wildtype/heterozyous hmgcs1 siblings (Sibling) or hmgcs1 homozygous mutant larvae (Mutant) using a riboprobe specific to dlx2a. n = 20 including 16/20 wildtype and heterozygous siblings and 4/20 homozygous mutants. Panel A depicts a wildtype sibling and no significant difference between wildtype embryos was detected. (C&D) ISH was performed at 3 dpf (protruding mouth stage) on hmgcs1 siblings (wildtype and heterozygous individuals) or hmgcs1 homozygous mutant larvae with a riboprobe specifically detecting col2a1a expression. Numbers denote the pharyngeal arch patterns present a 3dpf. Arrowhead denotes a loss of defined expression of col2a1a in hmgcs1 mutant larvae. n = 36 including 29 wildtype and heterozygous individuals with 7 homozygous mutants. Panel C depicts a wildtype sibling with no significant differences observed between wildtype and heterozygous carriers. (E&F) Tg(sox10:memRFP) hmgcs1 siblings or homozygous hmgcs1 mutant larvae were imaged for RFP expression at 3dpf at the protruding mouth stage. Numbers indicate structures that will give rise to the putative ceratobranchial cartilages. n = 20 including 14/20 wildtype and heterozygous siblings and 6/20 homozygous mutants. Panel E depicts a wildtype sibling with no significant differences observed in heterozygous carriers.

Fig 4. hmgcs1 regulates facial development by a cholesterol independent mechanism.

(A-C) Wildtype AB embryos were treated with 0.01% DMSO (vehicle), 8uM lonafarnib (Lona) or 3 uM Ro 48 8071 (Ro-48) beginning at 5 hours post fertilization and for a period of 4 days post fertilization (dpf). Alcian blue was used to analyze the cartilage structures of the developing zebrafish head and neck. (A-C) demonstrate the dissected viscerocranium of 4 day old larvae. (DMSO n = 30, Ro-48 n = 14, and Lona n = 8) (D-F) Whole mount in situ hybridization (ISH) was performed at 3 days post fertilization (protruding mouth stage) with embryos treated with 0.01% DMSO (vehicle), 8uM lonafarnib (Lona) or 2.5 uM Ro 48 8071 (Ro-48) (DMSO n = 20, Ro 48 n = 27, and Lona n = 30) (G) Schematic representation of viscerocranium. The extension of the Meckel’s cartilage (mandible) was measured as a detection for the presence of a facial cleft phenotype. Distance is indicated by the red line. (G’) Average was normalized and represented relative to the vehicle control group. Asterisk denotes statistical significance of p<0.001. Statistical analysis was performed using a T-test.

Fig 5. Inhibition of isoprenoids and cholesterol has not effect on sonic hedgehog (Shh) activity.

(A-C) Whole mount in situ hybridization (ISH) was performed at 3 days post fertilization (protruding mouth stage) (dpf) with embryos treated with 0.01% DMSO (vehicle), 8uM lonafarnib (Lona) or 2.5 uM Ro 48 8071 (Ro-48). (DMSO n = 10, Ro-48 n = 13, and Lona n = 16). (D) Sybr green based quantitative real time PCR was performed on hmgcs1 mutant embryos (Homozygous MT) or their wildtype siblings (WT Sibling) at the indicated days post fertilization (DPF) to detect mRNA expression of gli1. Error bars represent standard deviation of technical replicates.