Identification of a family of fatty-acid-speciated sonic hedgehog proteins, whose members display differential biological properties

Abstract

Hedgehog (HH) proteins are proteolytically processed into a biologically active form that is covalently modified by cholesterol and palmitate. However, most studies of HH biogenesis have characterized protein from cells in which HH is overexpressed. We purified Sonic Hedgehog (SHH) from cells expressing physiologically relevant levels and showed that it was more potent than SHH isolated from overexpressing cells. Furthermore, the SHH in our preparations was modified with a diverse spectrum of fatty acids on its amino termini, and this spectrum of fatty acids varied dramatically depending on the growth conditions of the cells. The fatty acid composition of SHH affected its trafficking to lipid rafts as well as its potency. Our results suggest that HH proteins exist as a family of diverse lipid-speciated proteins that might be altered in different physiological and pathological contexts in order to regulate distinct properties of HH proteins.

Figure 1. SHH-I cells produce endogenous-like levels of potent SHH-Np

(A) An immunoblot showing SHH-Np abundance in SHH-I cells, under conditions in which its expression was induced (+) or uninduced (−) with muristerone. SHH-I parental cells, which are not engineered to express SHH, were used as a control (Ctrl). GAPDH was used to verify protein normalization. (B) An aliquot of SHH-I cellular lysate was tested for SHH-Np associated activity using the Light-II reporter cell line. SHH-Np activity measurements were carried out in the linear range of this assay (see Figure S1B), and these activity results were then normalized to overall SHH-Np levels to determine potency. Error bars represent the standard deviation (SD) in one representative experiment. (C) SHH-Np levels from uninduced SHH-I cells and chick embryo limb buds were compared by immunoblotting. During development SHH-Np is produced in the posterior portion of limb buds (Post). Here, the anterior (Ant) portion of limb buds serves as a negative control for SHH-Np. As only approximately 20% of the posterior tissue consists of SHH producing cells (Riddle et al., 1993), we mixed 20% of SHH-I cell lysate with 80% of lysate from anterior limb bud tissue for this comparison. GAPDH was used to verify normalization.

Figure 2. SHH-Np purified from low-level SHH expressing cells is highly active

(A) SHH-I cells, or the SHH-I parental cell line (Ctrl), were dounce homogenized under isotonic conditions and total lysate (left panel) separated by ultracentrifugation at 100,000 × g (right panel) to generate a cytosol-enriched fraction (Cyt) and a membrane-enriched fraction (Mem). These fractions were volume normalized to that of the original cellular lysate and immunoblotted as indicated. Tubulin serves as a cytosolic protein control while the Na⁺/K⁺ transporter serves as a membrane protein control. (B) Aliquots of the indicated fractions from various steps of SHH-Np purification were separated by SDS-PAGE, followed by visualization of proteins by silver staining (TL: total lysate, S: 100,000 × g supernatant, M: combined 100,000 × g pellet detergent extract, FT: non-bound material, W: column wash, E: column eluate). Recombinant, unmodified, SHH-N is shown as a control (rSHH-N). Electrophoretic retardation of rSHH-N, relative to cholesterol modified SHH-Np, has been previously noted (Lee et al., 1994). (C) The indicated amounts of purified SHH-Np were incubated with C3H10T1/2 fibroblasts, which differentiate into osteoblasts in response to SHH (squares: purified SHH-Np, circles: rSHH-N). Alkaline phosphatase activity, which is an indirect, quantitative measurement of this differentiation, was then measured. (D) The indicated amounts of purified SHH-Np were incubated with C3H10T1/2 fibroblasts, followed by RNA extraction. The levels of Gli1 and GAPDH were then determined by q-RT-PCR. Error bars represent the SD in one representative experiment.

Figure 3. Fatty acid speciation of SHH-Np is cell context dependent

(A): A schematic showing the procedure used to identify the fatty acid modification on SHH-Np. (B-D): Pie charts showing the relative abundance of lipid species identified on SHH-Np purified isolated under three different cell contexts: 10% FBS without muristerone induction of SHH expression , 10% FBS and muristerone induction of SHH expression, and serum deprivation and muristerone induction of SHH expression.

Figure 4. Fatty acid speciation of SHH-Np alters its lipid raft enrichment

(A) An immunoblot of cell lysates and conditioned media, from SHH-I cells incubated in the presence of indicated fatty acids. GAPDH served as a normalization control for cellular lysates. The same volume of conditioned medium was subjected to TCA precipitation prior to loading. (B) Lysates from SHH-I cells incubated with the indicated lipids, or DMSO control, were separated over an OptiPrep gradient to isolate a lipid raft enriched fraction. Fractions from these various OptiPrep density gradients were resolved by SDS-PAGE then analyzed by immunoblotting for SHH-Np, GAPDH as a cytoplasmic protein marker, or the lipid raft marker flotillin. Note that flotillin localization does not change with various lipid additions. Quantification of SHH-Np lipid raft enrichment from cells incubated with the indicated fatty acids is shown in (C). Error bars represent the standard error of the mean (SEM) of three independent experiments. p values ≤ 0.05 are considered statistically significant, and indicated by a asterisk.

Figure 5. Modification of SHH-Np with distinct fatty acids alters its functionality

(A) Lysates from embryonic limb bud explants exposed to different fatty acids were separated over an OptiPrep density gradient to isolate the lipid raft fraction. Various gradient fractions were resolved by SDS-PAGE than analyzed by immunoblotting for SHH-Np. Caveolin-1 was used as a lipid raft marker while GAPDH served to label non-lipid raft associated subcellular fractions. Treatment with saturated fatty acids did not change SHH-Np localization pattern over the DMSO control (data not shown). (B) An immunoblot (upper panel) of limb bud lysates showing effect of lipid modifications on SHH-Np levels. Lysates contain similar amount of total protein as indicated by total protein silver staining (lower panel). (C) The potency of SHH-Np was determined by incubating lysate from treated limb buds with Light-II cells, and then normalizing this activity to SHH-Np levels. Error bars represent the SEM of three independent experiments. p values ≤ 0.05 are considered statistically significant, and indicated by an asterisk.