Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Oct 1.
Published in final edited form as: Dev Biol. 2010 Jun 28;346(1):1–10. doi: 10.1016/j.ydbio.2010.06.024

dHIP14-dependent palmitoylation promotes secretion of the BMP antagonist Sog

Kyung-Hwa Kang 1,2, Ethan Bier 1
PMCID: PMC2937060  NIHMSID: NIHMS218209  PMID: 20599894

Abstract

Analysis of diverse signaling systems has revealed that one important level of control is regulation of membrane trafficking of ligands and receptors. The activities of some ligands are also regulated by whether they are membrane bound or secreted. In Drosophila, several morphogenetic signals that play critical roles in development have been found to be subject to such regulation. For example, activity of the Hedgehog (Hh) is regulated by Raspberry, which palmitoylates Hh. Similarly, the palmitoylases Porcupine and Raspberry increase the activities of Wingless (Wg) and the EGF-ligand Spitz (Spi) respectively. In contrast to its vertebrate homologues, which have typical N-terminal signal sequences, the precursor form of Drosophila Hh contains an internal type-II secretory signal motif. The Short Gastrulation (Sog) protein is another secreted Drosophila protein that contains a type-II signal and differs from its vertebrate ortholog Chordin which contains a standard signal peptide. In this study, we examine the regulation of Sog secretion and regulation by dHIP14, the ortholog of a mammalian palmitoylase first identified as Huntington Interacting Protein (HIP). We show that dHIP14 binds to Sog and that Sog is palmitoylated. In S2 cells, dHIP14 promotes secretion of Sog as well as stabilizing a membrane associated form of Sog. We examined the requirement for candidate Cysteine residues in the N-terminal predicted cytoplasmic domain of Sog and find that Cys27, one of two adjacent Cysteines (Cys27,28), is essential for the full activity of dHIP14 and its effect on Sog. Finally, we find that dHIP14 promotes the activity of Sog in vivo. These studies highlight the growing importance of lipid modification in regulating signaling at the level of ligand production and localization.

Introduction

During early phases of pattern formation, cell fates are determined by positional information conveyed by the graded distribution of morphogens. Gradients of morphogens result from extra cellular diffusion of the morphogens, which is tightly regulated by several mechanisms. Lipid modification of secreted ligands, such as cholesterolization or palmitoylation, has emerged as one important class of mechanisms for regulating the secretion, range, and activity of signals (Mann and Beachy, 2004; Miura and Treisman, 2006).

Palmitoylation is a post-translational modification resulting from the covalent attachments of a 16-carbon fatty acid, palmate, to specific Cysteine residues via thioester linkage (Resh, 2006). In contrast to other lipid modifications, such as myristoylation or farensylation, palmitoylation is reversible and dynamically regulated by extracellular signals. Palmitoylation modulates protein function by facilitating membrane association, interaction with other proteins, and targeting proteins to specific subcellular compartments. Many palmitoylated proteins were first identified in yeast using proteomics (Linder and Deschenes, 2004; Roth et al., 2006; Wan et al., 2007). While protein palmitoylation is emerging as an important mechanism for regulating the activity of ligands, receptors, and intracellular signaling mediators, the mechanisms by which this modification regulates the function of target proteins are just beginning to be appreciated.

In Drosophila, three critical secreted proteins, Hh, Wg, and Spitz are known to be palmitoylated by two different palmitoyl acyltransferases (PAT), Raspberry (RASP; also known as Skinny Hedgehog, Sightless and Central Missing) (Amanai and Jiang, 2001; Chamoun et al., 2001; Linder and Deschenes, 2004) and Porcupine (Porc), which are members of the membrane-bound O-acyltransferase (MBOAT) family of palmitoyl transferases. In the case of Wingless (Wg), palmitoylation of an internal Cysteine residue by Porc is required for targeting it to lipid rafts (Zhai et al., 2004), as well as its N-glycosylation in the endoplasmic reticulum, and ultimately for its secretion (Galli et al., 2007). Both Hh and Spitz are palmitoylated by RASP on their N-terminals, but these modifications have opposing effects. Palmitoylation enhances the activity of Hh in Drosophila and the long range effect of Sonic hedgehog (Buglino and Resh, 2008) by forming multimeric complexes of the ligand in a larger structures referred to as argosomes (Cadigan, 2002; Christian, 2002; Greco et al., 2001; Simpson et al., 2009), while palmitoylated Spitz is restricted to the membrane of producing cell leading to increased local concentration of the ligand that activates the EGF-R in adjacent cells (Miura et al., 2006).

Short Gastrulation (Sog), an antagonist of Dpp, is another secreted molecule which has a type II signal peptide (Francois et al., 1994). Sog is secreted from lateral cells comprising the neuroectoderm and forms a concentration gradient in the neighboring dorsal epidermis (Srinivasan et al., 2002). This Sog gradient extends over a range of 15–16 cells and results in a reciprocal Dpp activity gradient that is high dorsally and lower ventrally. Sog also plays a role during pupal stages to refine the wing vein pattern (Yu et al., 1996). During this stage, Sog, which is expressed in the intervein regions of the wing primordium, acts over relatively short distances to focus BMP signaling and hence vein formation to the center of 7–8 cell diameter provein domains.

dHIP14 is an Drosophila ortholog of mammalian Hip14, Huntingtin interacting protein 14, which was first identified as a Huntingtin (Htt) binding protein (Singaraja et al., 2002). Hip14 is a palmitoyl acyltransferase (PAT) with an integral transmembrane domain with six predicted membrane spanning domains as well as a Cysteine rich domain (CRD) including a DHHC motif (DHHC-CRD) (Linder and Deschenes, 2004). Hip14 has been shown to palmitoylate the Htt protein and inhibits its cytoplasmic aggregation. Several presynaptic proteins including PSD-95/DLG, Sty1, SNAP25, GAD43, and GAD65 have also been demonstrated to be palmitoylated by Hip14 in vitro (Huang et al., 2004; Singaraja et al., 2002; Yanai et al., 2006). In Drosophila, recent studies found that palmitoylation of the Cysteine string protein (CSP) protein and SNAP25 by dHIP14 were critical for synaptic localization of these two proteins (Bannan et al., 2008; Stowers and Isacoff, 2007), while localization of PSD-95 and Sty1 was not affected in dHIP14- mutants (Ohyama et al., 2007). dHIP14 is expressed ubiquitously in all tissues and is present in Golgi and presynaptic vesicles in neuronal cells. Its broad distribution suggests that dHIP14 might have other target proteins in Drosophila in addition to its well characterized nervous system target CSP.

In this study, we show that dHIP14 binds to the conserved BMP antagonist Sog. Sog is palmitoylated in S2 cells and dHIP14 promotes secretion of Sog as well as stabilizing a membrane associated form of it. We examined the requirement for candidate target Cysteine residues in the N-terminal predicted cytoplasmic domain of Sog and find that one of them (Cys27) is essential for mediating the full activity of dHIP14 on Sog. Finally, we find that dHIP14 promotes the activity of Sog in vivo. These studies suggest that palmitoylation of Sog on its cytoplasmic domain by dHIP14 promotes its secretion and highlight the growing importance of regulating signaling at the level of ligand production and localization.

Results

dHIP14 and Sog physically interact and co-localize within the Golgi

In a genome-wide for protein-protein interactions in Drosophila a single Sog binding protein was identified as CG6017 (Giot et al., 2003). This gene encodes the single Drosophila ortholog of a mammalian protein known as Huntington Interacting Protein 14 (HIP14) (Singaraja et al., 2002), which share overall organization in primary sequence including a cluster of contiguous Ankyrin domains in the N-terminus and a more C-terminal DHHC domain (Fig. 1A), which has been shown to have enzymatic activity in HIP14. The DHHC domains are highly conserved between the fly and mammalian proteins, sharing a set of six Cysteine residues with identical spacing (Fig. 1B). To validate the two-hybrid protein-protein interaction we generated a GFP tagged form of dHIP14 and tested whether this protein could be co-immunoprecipitated (co-IP) with Myc-tagged Sog from S2 cells co-transfected with dHIP14-GFP and Sog-Myc expression constructs. We found that we could selectively co-IP dHIP14-GFP with anti-Myc antibodies and reciprocally that we could co-IP Sog-Myc with anti-GFP (Fig. 2A,B). Because dHIP14 is an intracellular protein localized in the Golgi (see below), we also asked whether the cytoplasmic domain of Sog was required for interaction with dHIP14. We found that dHIP14 interacted with various carboxy-truncated forms of Sog but not with a directly secreted form of Sog engineered with a standard type signal sequence (not shown). These data confirm that dHIP14 is an authentic Sog interacting protein that it most likely binds to sequences in the N-terminal cytoplasmic domain of Sog.

Figure 1. A, Diagram of dHIP14 domain structure.

Figure 1

Open in a new tab

A, dHIP14 is an ortholog of mammalian Huntingtin interacting protein 14(Hip14), which contains 5 Ankyrin domains in the N-terminal region and a more carboxy-terminal zinc finger DHHC-CRD domain (zf-DHHC), which carries out the enzymatic function. The DHHC domain is highly conserved from yeast to mammals. dHIP14 has six predicted Ankyrin domains and a DHHC domain.

Figure 2. Sog binds to dHIP14 and co-localizes with dHIP14 in the Golgi in S2 cells.

Figure 2

Open in a new tab

A,B, S2 cells were transfected with Sog-myc and dHIP14-GFP constructs for 48 hrs. Cell lysates were immunoprecipitated by an anti-GFP antibody followed by immunoblotting with an anti-Myc antibody (A) or alternatively, immunoprecipitated with an anti-Myc antibody and immunoblotted with an anti-GFP antibody (B). C, dHIP14 co-localizes with the Golgi marker DG13 (upper panels) and with Sog (lower panels) in S2R+ cells.

Since Sog and dHIP14 physically interact, we wondered how dHIP14 was localized in the cell and whether it co-localized with Sog. We examined the subcellular expression of a GFP tagged form of dHIP14 (dHIP14-GFP), which has an activity similar to that of the untagged enzyme in transgenic flies (see below), and found that it is primarily localized to the Golgi compartment as revealed by double labeling for GFP and the Golgi marker DG13 (Fig. 2C, upper panels). In addition, cells double transfected with Sog-Myc and dHIP14-GFP constructs revealed a high degree of co-localization between these two proteins, although the distribution of Sog in cells co-expressing dHIP14 is indistinguishable from that of cells expressing Sog alone, indicating that dHIP14 does not appreciably alter the subcellular distribution of Sog. Low levels of Sog were also observed on the cell surface. These data indicate that Sog and dHIP14 bind directly to each other in vitro and co-localize within cells, where they are concentrated in the Golgi. The predominant staining of Sog in the Golgi is consistent with Sog being readily secreted from S2 cells.

Sog is palmitoylated in S2 cells

Mammalian HIP14 has been shown to palmitoylate various substrates including neuronal proteins such as Huntington protein, SNAP-25, Synaptotagmin1, PSD-95, and GAD65 (Ducker et al., 2004; Huang et al., 2004; Singaraja et al., 2002; Yanai et al., 2006). The yeast ortholog of HIP known as Akr1P has also been shown to function as a palmitoylase and acts on substrates including the pheromone receptor and Casein Kinase (Babu et al., 2004; Givan and Sprague, 1997; Pryciak and Hartwell, 1996; Roth et al., 2002), and Drosophila dHIP14 has been shown to act as a palmitoylase on the synaptic protein CSP (Ohyama et al., 2007; Stowers and Isacoff, 2007). Since dHIP14 binds to Sog, we tested whether Sog might also be a substrate for palmitoylation by dHIP14. Palmitoylated proteins can be selectively labeled with Biotin by acyl-biotin exchange in which total protein extracts are treated sequentially with N-ethylmaleimide (NEM) (to block free sulfhydril groups), hydroxylamine (to remove thioester linked lipids), and then incubation with a reactive thiol-linked biotinylated reagent (e.g., Biotin-HPDP) (see materials and Methods for details) (Roth et al., 2006). We prepared whole protein extracts from S2 cells expressing Sog-Myc, labeled palmitoylated residues with Biotin, immune-precipitated with Streptavidin-linked beads, separated the proteins by PAGE, and then immunoblotted with an anti-Myc antibody (Fig. 3A). This experiment suggests that Sog is directly palmitoylated by dHIP14 since Sog can be detected as a silver stained band (Fig. 3A, lower panels) as well as a corresponding immunoreactive band (Fig. 3A, upper panels).

Figure 3. Inhibition of palmitoylation reduces Sog secretion.

Figure 3

Open in a new tab

A, Palmitoylation of Sog was assayed by the Acyl-biotin-exchange method. Briefly, total cell lysates were treated with NEM to block free Cysteine residues. One half of the sample was treated with Hydroxylamine to remove palmate from palmitoylated Cysteine residues. Both samples were treated with Biotin labeled Cysteine cross-linking reagent. Biotin labeled samples were captured by Streptavidin and subjected for the immunoblotting with anti-Myc antibody to detect Sog-Myc protein (upper lanes). Part of the samples were also electrophoresed by PAGE and the gel was silver stained (lower lanes, the mobility of Sog is indicated by the arrow). B, Secretion of Sog was blocked by palmitoylation inhibitor, 2Br-palmate. Sog-Myc expressing cells were treated with different concentrations of 2Br-palmate for 6 hrs. Secretion of Sog into the medium was assessed by immunoblotting with anti-Myc to detect Sog-Myc in the conditioned media (Media), and was compared with the amount of total intracellular Sog in cell lysates (Total Cell Lysate). Sog secretion was inhibited by 2Br-palmate in a dose dependent fashion. Blocking palmitoylation also decreased levels Sog total cell lysates and lead to accumulation of Sog in a TritonX soluble compartment (Ttx Soluble Lysate) at intermediate doses, presumably by preventing membrane association of what otherwise would have been the palmitoylated pool of Sog. C, Addition of the proteasome inhibitor MG132 stabilized intracellular Sog (Total Cell Lysate) and Sog in the soluble fraction (Ttx Soluble Lysate), even at high doses of 2Br-palmate.

Another way to assess the biological relevance of Sog palmitoylation is to ask whether the palmitoylation inhibitor 2Br-palmate can alter the secretion or subcellular distribution of Sog. We treated sog-myc transfected S2 cells with 2Br-palmate for 16 hours and monitored Sog-Myc secretion into the media. We observed that levels of secreted Sog were reduced by 2Br-palmate in a dose dependent fashion (Fig 3B). We also examined the effect of inhibiting palmitoylation on intracellular Sog levels. Of the total Sog in the cell pellet in untreated cells (Total Cell Lysate), only trace amounts of intracellular Sog were in found in a Triton soluble fraction (Ttx Soluble Lysate, Fig 3B, lane 1). When 2Br-palmate was added to the media, it reduced total intracellular levels of Sog in a dose dependent fashion (Fig. 3B, lanes 2–4). At low doses of 2Br-palmate, we also observed increased levels of soluble intracellular Sog. These results suggest that inhibiting palmitoylation reduces the level of membrane associated Sog and may slow transit of Sog through the secretory pathway. One possible reason for the reduction in intracellular Sog levels by 2Br-palmate is that non-palmitoylated forms of Sog protein become unstable and are degraded by the proteasome. To test this possibility we added the proteasome inhibitor MG132 and found that Sog levels no longer changed in response to 2Br-palmate (Fig. 3C). These data suggest that palmitoylation of Sog both increases Sog secretion and targets intracellular forms of Sog to a Triton insoluble compartment. When palmitoylation is inhibited, the intracellular pool of Sog moves into a Triton soluble fraction, which is readily degraded in a proteasome dependent fashion.

dHIP14 promotes Sog secretion from S2 cells

As mentioned above, palmitoylation of Wg increases its secretion, while palmitoylation of Spitz has the converse effect, thereby concentrating this EGF-R ligand at the cell surface (Miura and Treisman, 2006). Since the palmitoylase dHIP14 binds Sog, we asked whether it played a role in secretion of Sog from S2 cells. We treated S2 cells with double stranded RNA oligonucleotides to inhibit expression of dHIP14 by RNA interference (RNAi). This dHIP14 RNAi treatment greatly reduced expression of transfected dHIP14-GFP (Fig. 4A, compare lanes 2 and 3). We then assayed the effect of reducing or increasing dHIP14 RNA levels on Sog secretion into the media. In cells transfected with a sog-myc expression construct alone, Sog-Myc protein was efficiently secreted (Fig. 4B, lane 2), and some Sog could also be detected within cells in both Triton soluble and insoluble fractions. Presumably, the soluble fraction represents Sog within an aqueous membrane compartment (e.g., vesicles) on route to the cell surface while the non-soluble Sog fraction is a membrane associated. Consistent with a component of Sog being membrane associated, Sog staining was observed on the surface of unfixed transfected cells (data not shown). When dHIP14 was co-transfected with sog, the levels of secreted Sog increased (Fig. 4B, lane 3), while co-transfection with dHIP14-RNAi reduced Sog secretion (Fig. 4B, lane 4). dHIP14-RNAi treatment also reduced the amount of detergent insoluble Sog, while not appreciably affecting the levels of soluble intracellular Sog (Fig. 4B, lane 4). These experiments indicate that dHIP14 promotes Sog secretion and, taken together with the results from the 2Br-palmate inhibition experiments, suggests that dHIP14-dependent palmitoylation of Sog is also required to stabilize a membrane associated form of Sog that may serve as a precursor for secreted Sog.

Figure 4. dHIP14 increases the secretion and activity of Sog.

Figure 4

Open in a new tab

A, dHIP14 RNAi virtually eliminated expression of a transfected dHIP14-GFP expression construct in S2 cells (indicated by arrow). B, dHIP14 RNAi blocked membrane targeting and secretion of transfected Sog, while transfection of S2 cells with dHIP14-GFP increase Sog secretion into the media (Media). We also assayed Sog levels in total cell lysates (Total Cell Lysate) and from both TritonX soluble (Ttx Soluble Lysate) and insoluble (Ttx Insoluble Lysate) fractions. dHIP14 transfection reduced levels of soluble Sog, while dHIP14-RNAi reduced Sog levels in the insoluble fraction (presumably membrane associated Sog). C, The ability of Sog, Tsg, or dHIP14 RNAi to inhibit BMP signaling in supernatants recovered from cells co-transfected with Sog for Dpp. Conditioned media from cells expressing a combination of Sog-Myc, Tsg-His, and Dpp-HA was added to cells transfected with MAD-FLAG (Flag) with or without dHIP14-RNAi treatment. After 6 hrs, phosphorylation of MAD was checked by immunoblotting with an anti-phospho-MAD (P-MAD) antibody. The BMP inhibitory activity of Sog, which blocks Dpp induced phosphorylation of MAD, was virtually eliminated when cells were treated with dHIP14-RNAi.

Secreted Sog has been shown to block BMP signaling mediated by the Dpp ligand as assayed by phosphorylation of the cytoplasmic signal transducer MAD to generate pMAD (Shimmi and O’Connor, 2003). We therefore asked whether supernatants collected from cells co-transfected with dHIP14-RNAi and sog would have a reduced ability to block Dpp signaling as might be expected based on the lower levels of Sog secretion. We found that this was indeed the case since dHIP14-RNAi eliminated the ability of Sog (with or without the co-BMP antagonist Tsg) to reduce Dpp dependent pMAD activation (Fig. 4C, compare lanes 3 and 5 to lanes 2 and 4).

Cysteines 27,28 mediate the effect of dHIP14

Since palmitoylation of proteins occurs on Cysteine residues (Linder and Deschenes, 2003; Resh, 1999; Smotrys and Linder, 2004) we asked whether the two adjacent Cysteine residues (Cys27 and Cys28) preceding the transmembrane domain of Sog in the predicted cytoplasmic type-II signal sequence, are required to mediate the effect of dHIP14. We mutated these residues individually and in combination, co-expressed these mutant forms of Sog with dHIP14 in S2 cells, and then assayed the amount of Sog secreted into the medium (Fig. 5). While mutation of Cys27 alone or both Cys27 and Cys28 to Serine (C27S and C27,28S respectively) did not greatly alter the baseline secretion of Sog (Fig. 5A - compare lanes 3 and 5 to lane 1), both of these mutations eliminated the increase in Sog secretion stimulated by co-transfection with dHIP14. These Cys->Ser mutants also caused a reduction in the levels of intracellular Sog (Fig. 5A - compare lanes 4 and 6 to lane 2). In addition, the level of Triton extractable intracellular Sog was greatly increased for the C27S and C27,28S mutant forms of Sog, and in the case of the double mutant, the amount of triton insoluble Sog was also significantly reduced (Fig. 5B). As the C28S single mutant behaved much as wild-type Sog, we conclude that Cys27 is the most critical of the two residues for mediating dHIP14 function, although C28S does seem to have an effect on the levels of intracellular Triton-insoluble Sog in combination with C27S in the double mutant. It is possible that an additional untested Cysteine residue (Cys70) within the TM domain of Sog also contributes to mediating the effect of dHIP14 since the C27,28S double mutant is secreted at basal levels similar to that of wild-type Sog. In contrast, in cells treated with dHIP14 RNAi or with the palmitoylation inhibitor 2Br-palmate, basal Sog secretion is also severely reduced. The inability of the C27S and C27,28S mutant Sog proteins to respond to elevated dHIP14 expression could also be observed at the level of BMP signaling as manifest by the reduction in the ability of these Sog mutants to block Dpp mediated phosphorylation of MAD (Fig. 5C). These results are consistent with the data described above regarding the dual effect of dHIP14 on promoting secretion of wild-type Sog and stabilizing a membrane associated form of Sog.

Figure 5. Cysteines 27,28 mediate part of dHIP14 activity.

Figure 5

Open in a new tab

A, Sog has a two adjacent Cysteine residues (Cys27 and Cys28) facing cytoplasm near the transmembrane domain, which are predicted for the targets of palmitoylation. These two Cysteine residues were mutated to Serine (C27S and C28S). After transfection of wild-type or mutant Sog-Myc expression constructs into S2 cells, secretion of Sog into the culture media (Media) or retention of Sog in the cell (Pellet) was determined by SDS-PAGE followed by immunoblotting with an anti-Myc antibody. While mutation of these Cysteine residues had little effect on the basal level of Sog secretion into to media, the C27S single mutant as well as the C27,28S double mutant did eliminate dHIP14 stimulated Sog secretion. B, Analysis of the intracellular distribution of Sog mutant proteins. A much large proportion of the intracellular C27S mutant Sog protein is Triton soluble than is wild-type Sog. Cells transfected with wild-type Sog-Myc or mutant Sog-Myc constructs were separated into a Triton-X insoluble phase (Ttx Insoluble) and a soluble aqueous phase (Ttx Soluble) and aliquots of these fractions were separated by SDS-PAGE and immunoblotted using an anti-Myc antibody. C, Supernatants recovered from cells co-transfected with Dpp-HA and wild-type or mutant Sog-Myc constructs were added to BMP responsive cells transfected with MAD-FLAG. P-MAD levels were assayed by immunoblotting with an anti-P-MAD antibody (arrow), total MAD with an anti-FLAG antibody (MAD-Flag, arrow), and Sog-Myc and Dpp-HA were detected respectively by immunoblotting with anti-Myc and anti-HA antibodies. The C27S single mutant and the C27,28S double mutant were unable to inhibit Dpp induced phosphorylation of MAD, while the C28S mutant inhibited with the same efficacy as wild-type Sog. D, The subcellular distributions of wild-type and mutant forms of Sog-Myc were determined in conjunction with dHIP14-GFP in transfected S2 cells. The C27S and C27,28S Sog mutants stained enlarged vesicular structures relative to the wild-type Sog or the C28S Sog mutant. In addition, some of the C27S and C27,28S mutant Sog protein did not co-localize with dHIP14 (arrows).

Since secretion of the C27S and C27,28S mutant forms of Sog does not increase in response to co-expression with dHIP14, we checked whether there was a change in the subcellular co-localization of these Sog mutants with dHIP14 relative to the strict co-localization observed for wild-type Sog (Fig. 5D). The C28S single mutant had a distribution indistinguishable from wild-type and co-localized strongly with dHIP14 (Fig. 5E, third row of panels), consistent with this mutation alone having little effect on Sog secretion or activity. In contrast, the levels of the Sog C27S and C27,28S mutant proteins as well as dHIP14 were elevated within this compartment relative to that observed for wild-type Sog (Fig. 5D, second and fourth rows of panels). In addition, there was no longer complete co-localization of the mutant Sog proteins with dHIP14 (arrows indicate Sog staining without dHIP14). These results suggest that Cys27 plays an important role in dHIP14-dependent Sog secretion and in the association of dHIP14 with Sog in the Golgi.

dHIP14 over-expression mimics Sog over-expression

The analysis of dHIP14 function in S2 cells described above suggests that palmitoylation promotes Sog secretion as well as stabilization of a membrane associated form of Sog. Both of these effects should increase the overall activity of Sog. We therefore examined the effect of over-expressing dHIP14 in vivo and compared to that of over-expressing sog. In situ analysis revealed that dHIP14 is expressed ubiquitously during embryogenesis and in larval imaginal discs (data not shown), suggesting that the activity of this enzyme is most likely not regulated by its pattern of expression. Nonetheless, since co-expression of dHIP14 and Sog in cell culture increased Sog secretion, we tested whether over-expressing dHIP14 might similarly increase Sog activity in vivo. As the early embryo and pupal wing are two developmental contexts in which Sog plays a well characterized role in inhibiting Dpp signaling (Francois et al., 1994; Yu et al., 2004; Yu et al., 1996), we assayed the effect of expressing dHIP14 in these two contexts. Ubiquitous expression of dHIP14 throughout the wing (Fig. 6B), or restricted expression in the central region between the L3 and L4 vein primordia (Fig. 6E), resulted in modest vein-loss phenotypes that closely mimic those caused by mis-expression of Sog (Fig. 6C - compare with 6B; Fig. 6F -compare with 6E). Similar wing phenotypes were observed using wild-type or a C-terminally-GFP-tagged form of GFP indicating that this tag did not appreciably alter the function of dHIP14 (data not shown). We also co-expressed dHIP14 and Sog and found that the phenotype was the same as with expression of either transgene alone (Fig. 6D). The concordance of the dHIP14 and sog mis-expression phenotypes is consistent with dHIP14 acting to promote Sog secretion. The fact that co-expressing sog with dHIP14 did not result in a stronger phenotype than that of either sog or dHIP14 alone is consistent with prior studies indicating that only low levels of sog are required to achieve its full effect (e.g., similar wing phenotypes were observed in response to expressing one versus eight copies of sog (Yu et al., 2004; Yu et al., 1996)).

Figure 6. dHIP14 mimics Sog activity in vivo and acts upstream of the BMP receptor.

Figure 6

Open in a new tab

A, A wild-type wing. B–F, Over-expression of Sog (B), or dHIP14 (C) with the ubiquitous wing-GAL4 driver (ubi = MS1096) or the ptc-GAL4 driver (E,F) resulted in a similar vein loss patterns, which was unchanged by co-expressing both transgenes (D). Arrows indicate the stereotyped pattern of vein loss in these wings. G–I, Co-expression Sog (H) or dHIP14 (I), suppressed the ectopic vein phenotype caused by expression of Gbb ligand alone (G) with the ptc-GAL4 driver. J–L, In contrast to their ability to suppress the effect of Gbb, neither Sog (K) nor dHIP14 (L) could block the ectopic vein phenotype resulting from expression of an active form of the Sax* BMP type-I receptor (J). M,N, pMAD staining in early blastoderm state embryos. Embryos are viewed from a dorso-lateral perspective. Over-expression of dHIP14 in the anterior region of embryo driven by Bcd-GCN4GAL4 driver reveal reduction of pMAD staining in the region of dHIP14 expression (bracket) (M). Similar but stronger suppression of pMAD staining has been observed when expressing sog with this same driver (Yu et al., 2004).

We next asked whether dHIP14 acted like Sog to block the effect of the BMP ligand Gbb upstream of BMP-receptors (Yu et al., 2000). We found that co-expressing dHIP14 with Gbb between the L3 and L4 vein primordia (Fig. 6I) could reverse the vein fusion and ectopic vein phenotype caused by Gbb alone (Fig. 6G) as has been observed in the case of co-expressing Sog with Gbb (Yu et al., 2000) (Fig. 6H). In contrast to the Gbb mis-expression phenotype, dHIP14 and Sog were unable to suppress the sporadic ectopic vein phenotype caused ubiquitous expression of an activated form of the Sax BMP type-I receptor (Sax*) (Fig. 6K,L - compare to 6J). Similar results have been observed previously indicating that Sog is unable to block the effect of Sax* using other GAL4 drivers (Yu et al., 2000). These data are consistent with the view that dHIP14 functions upstream of the Sax receptor to promote secretion and activity of Sog during wing development.

We also examined the effect of dHIP14 mis-expression in the early embryo. Elevated or ectopic expression of sog during this stage of development leads to inhibition of BMP signaling dorsally as revealed most directly by the reduction of pMAD activation in a sharp dorsal midline stripe (Francois et al., 1994; Mizutani et al., 2005; Yu et al., 2004). When we expressed dHIP14 in the head region of the embryo using the bcd-GCN4GAL4 driver, we observed a significant reduction in pMAD staining in the head region (Fig. 6N - compare to 6M). A similar, although somewhat stronger, effect is observed when expressing Sog with this same driver (Yu et al., 2004). Although ideally one would wish to complement these gain-function studies with dHIP14 loss of function in the embryo, where even subtle effects of sog dosage can be assayed, a prior developmental role of dHIP14 in establishing oocyte polarity during oogenesis (K. Kang, manuscript in preparation) precludes such an analysis at this time. Overall, the results of the mis- expression studies in the wing and embryo support the view that dHIP14 promotes the effect of Sog.

Discussion

Lipid modification of many ligands plays an important role in regulating their activity and range of action (Mann and Beachy, 2004; Miura and Treisman, 2006; Resh, 2006) (Fig. 7). In this current study, we provide evidence that the type-II signal sequence of the BMP antagonist Sog is a target for palmitoylation by dHIP14. Reduction and gain-of-function studies in cell culture suggest this pos-translational modification is important for secretion of Sog and for targeting intracellular Sog to a Triton insoluble fraction, which may represent a membrane cargo compartment, possibly the Golgi, involved efflux of Sog. Mutational analysis suggests further that Cys27 in the cytoplasmic tail of the type-II signal sequence plays an important role in mediating dHIP14 activity. The elevated level of Sog secretion observed in cells over-expressing dHIP14 is active as a BMP antagonist since supernatants from dHIP14 transfected cells are more effective than control supernatants in inhibiting BMP mediated pMAD phosphorylation. In vivo experiments also indicate that over-expression of dHIP14 mimics the effect of over-expression of Sog in the wing and, like Sog, can inhibit Gbb-dependent vein formation upstream of the BMP receptor during wing development. Localized ectopic expression of dHIP14 in the early embryo likewise phenocopies the effect of by reducing pMAD activation along the dorsal midline. Cumulatively, these results suggest that palmitoylation increases secretion of an active form of Sog from cells, and may do so by elevating Sog flux through a membrane associated compartment.

Figure 7. The role of palmitoylation in regulating the activity of secreted ligands.

Figure 7

Open in a new tab

Diagram summarizing the effects of palmitoylation on the activity of secreted ligands in Drosophila. A, Palmitoylation can increase the secretion of morphogens as in the case of N-terminal palmitoylation of Hedgehog (Hh) mediated by RASP, which promotes long range diffusion and activity of this ligand in a lipid structure referred to as the Argosome or Exosome. Similarly, Porcupine (Porc) dependent palmitoylation of Wg increases its secretion and range of action by targeting it to lipid rafts in the membrane. B, Spitz (Spi), like Hh, is also palmitoylated by RASP, but in this case, palmitoylation reduces secretion of the ligand by linking it to the membrane, which raises its local concentration and short range ability to activate the EGF-R. C, Palmitoylation of full length Sog (FL-Sog) increases its secretion as a selective antagonist of heteromeric BMP ligands (e.g., Dpp:Scw or Dpp:Gbb), while truncated Supersog forms of Sog (S-Sog) may be retained in the membrane where they can inhibit the activity of a broader array of BMP ligands (e.g., Dpp:Dpp homodimers as well as heterodimeric BMP ligands). Potential sites of differential Sog cleavage to generate secreted Sog versus Supersog are indicated by the scissor symbol. Tld and Tok proteases may contribute to generating Supersog forms (Yu et al., 2000), while other proteases (e.g., ADAMS) most likely would be involved in release of full length Sog. The complementary C-terminal fragment of Sog produced by cleavage at the Supersog site (not shown) may also diffuse over long distances and may act to potentiate BMP signaling (Yu et al., 2004).

Dual effect of palmitoylation on Sog activity and subcellular compartmentalization

The role that dHIP14 plays in increasing Sog secretion and it activity as a BMP antagonist at moderate-to-long range is similar to the positive regulatory role of palmitoylation on Wg or Hh secretion and activity (Miura and Treisman, 2006; Resh, 2006), rather than to the negative effect this post-translational modification has on secretion of the EGF-R ligand, Spi (Miura et al., 2006). In the case of Spi, however, palmitoylation does increase the short range activity of Spi, presumably by concentrating on the cell surface, thereby increasing its likelihood of engaging EGF-R receptors on neighboring cells. Thus in all cases, palmitoylation plays a role in promoting ligand activity, albeit at differing ranges.

Another effect of dHIP14, which may also play a role in the increasing secretion of Sog, is that it results in a stable Triton-insoluble form of Sog. This stable intracellular form of Sog may represent a membrane bound intermediate within the secretory pathway or possibly also a pool of membrane tethered Sog at the cell surface. When palmitoylation is blocked pharmacologically or cells are treated with dHIP14 RNAi, the proportion of Triton soluble Sog increases. This soluble fraction of Sog is unstable and its degradation is proteasome-dependent, although this effect may be indirect since the soluble Sog would be expected to lie within a vesicular compartment where it would not come into direct contact with the cytoplasmically localized proteasome machinery. Since the ability of Sog to co-localize with dHIP14 in the Golgi compartment is compromised by the C27S mutation, one model to account for the data would be that palmitoylation is required to integrate Sog into the Golgi membrane thereby protecting it from degradation prior to reaching the cell surface, where it may be rapidly cleaved and liberated into the extracellular space.

Possible role of palmitoylation on regulating the activity of distinct forms of Sog

Although we focused in this study on analyzing the effect of dHIP14 on full length Sog, and found that it played a role increasing its secretion, palmitoylases can also limit the range of action of other ligands such as Spi (Miura et al., 2006) (Fig. 7). In this latter case, palmitoylation functions instead to retain the ligand on the cell surface, thereby increasing its local concentration. It is interesting to consider whether such an alternative potential role of palmitoylation might differentially regulate the activity of shorter processed forms of Sog that have been observed to be produced during several stages of development including oogenesis (Carneiro et al., 2006), embryogenesis (Yu et al., 2000), and pupal wing development (Araujo et al., 2003). Evidence obtained using antibodies raised to different portions of Sog and from genetic clonal analysis indicate that different forms of Sog diffuse to differing extents within the follicle cell epithelium during oogenesis (Carneiro et al., 2006), and potentially also in the wing epithelium where Sog diffuses from intervein cells into broad provein territories during the vein refinement process (Araujo et al., 2003).

The most well characterized truncated forms of Sog, which contain only the first of four cysteine repeat (CR) domains plus some adjacent stem sequences, are referred as Supersog molecules since they have the ability to inhibit signaling mediated by Dpp:Dpp homodimers as well as by Dpp:Scw or Dpp:Gbb heterodimers (Yu et al., 2004; Yu et al., 2000). In contrast, full length Sog (with all four CR domains intact) only blocks signaling mediated by BMP heterodimers (Yu et al., 2000). Supersog fragments may be generated at least in part by alternative cleavage by the Tld protease, which can also cleave Sog immediately after the CR1 to inactivate it (Yu et al., 2000). In addition, a C-terminal fragment of Sog including the CR2-CR4 domains, which is complementary to Supersog has a mild BMP activating function (Yu et al., 2004). It has been anecdotally noted in both our lab and that of M. O’Connor (personal communication) that the shorter Supersog fragments are not as readily secreted from cells in culture as full length Sog. There is also evidence that N-terminal fragments of Sog have reduced mobility in vivo in comparison to full length or C-terminal fragments during oogenesis (Carneiro et al., 2006), and clonal analysis of sog and integrin gene function during vein formation suggest that different forms of Sog have similar differing mobilities in the pupal wing (Araujo et al., 2003). While further experimental analysis will be required to resolve this question, it is interesting to speculate that if Supersog-like molecules indeed have a greater tendency to remain tethered to the surface of lateral neuroectodermal cells in which sog is expressed, they might be able to more effectively inhibit the activity of Dpp:Dpp homodimers, which are thought to diffuse in from the adjacent dorsal epidermal ectoderm. This localized deployment of Supersog-like fragments would help explain how Sog acts as potently as it does to prevent BMP autoactivation within the neuroectoderm given that secreted full length Sog would only be able to block the activity of incoming Dpp:Scw heterodimers. The restriction of such Dpp:Dpp inhibiting forms of Sog to the neuroectoderm would also account for why Sog acts selectively to block peak BMP signaling mediated by Dpp:Scw heterodimers in the epidermal ectoderm (Biehs et al., 1996).

If truncated forms of Sog are less diffusible than full length or C-terminal Sog fragments, then another question would be how palmitoylation might contribute to this difference. Perhaps there are two competing pathways for processing palmitoylated forms Sog at the cell surface, one which results in cleavage after the TM domain, thereby generating freely diffusible full length Sog, and another which results in Sog cleavage (by Tld or other proteases) some distance after CR1 to generate a tethered Supersog form and, potentially, a freely diffusible BMP promoting C-terminal fragment. Additional studies will be needed to address these various possibilities.

Methods

Fly Stocks and genetics

UAS-dHIP14 and UAS-dHIP14-GFP transgenic flies were made by general method. UAS-Sog constructs and GAL4 driver lines used in this study including MS1096- GAL4, ptc-GAL4, and, bcd-GCN4GAL4 have been described previously (Yu et al., 2004; Yu et al., 2000). Other standard stocks and balancer chromosomes were obtained from the Bloomington Stock Center.

S2 cell culture and transfection

S2 cells were cultured in Schneider’s cell medium (GIBCO, Carlsbad, CA) containing 10 % Fetal bovine serum (GIBCO, Carlsbad, CA). Cells were transfected with CellFectin (Invitrogen) according to manufacturer’s instructions. pRmHa vector was used for expression Sog and dHIP14 in S2 cells. Single or double Sog mutants (Cysteine to Serine) were made generated by PCR using the pfu-ultra (Stratagene, La Jolla, CA) with pRm-HA-Sog-Myc as a template and the following primers, For pRmHA-SogC27S, 5′-CTG GAA AGG AGC TCC TGC CAC AGC GAG GAC-3′, 5′-GTC CTC GCT GTG GCA GGA GCT CCT TTC CAG-3′, pRmHA-SogC28S; 5′-CTG GAA AGG AGC TGC TCC CAC AGC GAG GAC-3′, 5′-GTC CTC GCT GTG GGA GCA GCT CCT TTC CAG-3′, and for pRmHA-SogC27,28S; 5′-CTG GAA AGG AGC TCC TCC CAC AGC GAG GAC-3′, 5′-GTC CTC GCT GTG GGA GGA GCT CCT TTC CAG-3′. Protein expression was induced by treatment of 0.7 mM CuSO4 after 12 hrs of transfection.

Immunoblotting, immunoprecipitation, and immunohistochemistry

S2 cells were transfected with pRmHa-Sog-myc and pRmHa-dHIP14-GFP constructs for 48 hrs. Cells were harvested and lysed with 200 ul of TNN buffer (10 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.1 % NP-40). After preclearing of total cell lysates with protein-A agarose beads for 1 hr, the lysates were incubated with 1 mg of a polyclonal anti-GFP antibody (1: 20,000, Santa Cruz) and 30 l of protein-A agarose at 4° C for overnight. Beads were washed with PBS containing 0.1 % Triton X-100 for 3 times, boiled at 95° C for 5 minutes, and electrophoresed on NuPAGE 10% Bis-Tris gel (Invitrogen). Sog protein was detected by immunoblotting using an anti-Myc monoclonal antibody (1:1000, Roche). 1 g of anti-Myc antibody was used for immunoprecipitation of Sog. S2R+ cells grown on chamber slides were performed for immunofluorescence. After 48 hrs of transfection, cells were fixed for 10 min in 4% PFA in PBS, pH 7.0 and permeabilized in ice-cold buffer (20 mM Hepes, pH 7.4, 0.5% Triton X-100, 50 mM NaCl, 3 mM MgCl2, and 300 mM sucrose). Cells were blocked with 5% normal goat serum for 10 min and incubated with primary antibody (diluted in 5% normal goat serum in PBST) for 20 min. Secondary antibody was incubated with cells for 15 min. The fluorescence was detected by confocal microscopy using a Leica SP2-OBS microscope (Leica Microsystems, Wetzlar, Germany). For pMAD assays, cells were transfected with various combination of gene expression constructs for 48 hrs, and the media were collected and added to the cells expressing Mad-Flag. Cells were harvested and immunoblots were performed using a polyclonal antibody against phospho-Mad (1:1000, generous gift from P. Ten Dijke).

TritonX-114 Phase Separation

Phase separation was performed as described (Bordier, 1981).

Acyl-biotin-exchange assay(ABE)

We detected palmitoylation of Sog using the Acyl-biotin-exchange assay as described in (Wan et al., 2007). Briefly, S2 cells transfected with Sog-myc and dHIP14-GFP or dHIP14 RNAi were lysed with TNN buffer containing 1% Triton X-100 and 25 mM N-ethylmaleimide to quench free Cysteines for 30 min at 4°C, and proteins were precipitated with methanol/chloroform. The pellet was resuspended in 100 μl of LB (2% SDS, 100 mM NaCl, 50 mM Tris·HCl, pH 7.5), diluted with 600 μl of 1 M hydroxylamine (pH 7.4) and 300 μM biotin- HPDP (Pierce), and rotated for 2 h at 4°C. As a control, hydroxylamine was replaced by PBS. Proteins were then precipitated with methanol/chloroform, dried, and resuspended in LB. LB containing 0.1% Triton X-100 (1 ml) was added to each sample, aliquots (60 μl) were removed for loading controls, and the remaining reactions were incubated with 100 μl of Streptavidin-agarose beads (Pierce) for 1 h at room temperature on a nutator. Streptavidin beads were washed with PBS containing 0.5 M NaCl and 0.1% Triton X-100 and then once with PBS. Proteins were eluted by boiling for 5 min in 20 μl of LB containing 4x SDS sample buffer lacking 2-mercaptoethanol. Samples were analyzed by SDS/PAGE and immunoblotting.

Immunohistochemistry in embryos

Embryos were fixed in fixation buffer (4 ml PBS/0.05M EGTA, 8% formaldehyde, and 4 ml Heptane) for 20 minutes. After 3 times of washing with PBST (PBS containing 0.1% triton X-100), embryos were blocked with 5% normal goat serum in PBS-T for 1 hr and incubated with the anti-P-MAD antibody (1:1000) for overnight at 4° C. The secondary antibody, Alexa Fluor 488 goat anti-rabbit (1:500, Invitrogen-Molecular probe), was incubated for 1 hr and the fluorescence was detected by confocal microscopy using a Leica SP2-OBS microscope (Leica Microsystems, Wetzlar, Germany).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Amanai K, Jiang J. Distinct roles of Central missing and Dispatched in sending the Hedgehog signal. Development. 2001;128:5119–27. doi: 10.1242/dev.128.24.5119. [DOI] [PubMed] [Google Scholar]
  2. Araujo H, Negreiros E, Bier E. Integrins modulate Sog activity in the Drosophila wing. Development. 2003;130:3851–64. doi: 10.1242/dev.00613. [DOI] [PubMed] [Google Scholar]
  3. Babu P, Deschenes RJ, Robinson LC. Akr1p-dependent palmitoylation of Yck2p yeast casein kinase 1 is necessary and sufficient for plasma membrane targeting. J Biol Chem. 2004;279:27138–47. doi: 10.1074/jbc.M403071200. [DOI] [PubMed] [Google Scholar]
  4. Bannan BA, Van Etten J, Kohler JA, Tsoi Y, Hansen NM, Sigmon S, Fowler E, Buff H, Williams TS, Ault JG, Glaser RL, Korey CA. The Drosophila protein palmitoylome: Characterizing palmitoyl-thioesterases and DHHC palmitoyl-transferases. Fly (Austin) 2008:2. doi: 10.4161/fly.6621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Biehs B, Francois V, Bier E. The Drosophila short gastrulation gene prevents Dpp from autoactivating and suppressing neurogenesis in the neuroectoderm. Genes Dev. 1996;10:2922–34. doi: 10.1101/gad.10.22.2922. [DOI] [PubMed] [Google Scholar]
  6. Bordier C. Phase separation of integral membrane proteins in Triton X-114 solution. J Biol Chem. 1981;256:1604–7. [PubMed] [Google Scholar]
  7. Buglino JA, Resh MD. Hhat is a palmitoylacyltransferase with specificity for N-palmitoylation of Sonic Hedgehog. J Biol Chem. 2008;283:22076–88. doi: 10.1074/jbc.M803901200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cadigan KM. Regulating morphogen gradients in the Drosophila wing. Semin Cell Dev Biol. 2002;13:83–90. doi: 10.1016/s1084-9521(02)00014-9. [DOI] [PubMed] [Google Scholar]
  9. Carneiro K, Fontenele M, Negreiros E, Lopes E, Bier E, Araujo H. Graded maternal short gastrulation protein contributes to embryonic dorsal-ventral patterning by delayed induction. Dev Biol. 2006;296:203–18. doi: 10.1016/j.ydbio.2006.04.453. [DOI] [PubMed] [Google Scholar]
  10. Chamoun Z, Mann RK, Nellen D, von Kessler DP, Bellotto M, Beachy PA, Basler K. Skinny hedgehog, an acyltransferase required for palmitoylation and activity of the hedgehog signal. Science. 2001;293:2080–4. doi: 10.1126/science.1064437. [DOI] [PubMed] [Google Scholar]
  11. Christian JL. Argosomes: intracellular transport vehicles for intercellular signals? Sci STKE. 2002;2002:pe13. doi: 10.1126/stke.2002.124.pe13. [DOI] [PubMed] [Google Scholar]
  12. Ducker CE, Stettler EM, French KJ, Upson JJ, Smith CD. Huntingtin interacting protein 14 is an oncogenic human protein: palmitoyl acyltransferase. Oncogene. 2004;23:9230–7. doi: 10.1038/sj.onc.1208171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Francois V, Solloway M, O’Neill JW, Emery J, Bier E. Dorsal-ventral patterning of the Drosophila embryo depends on a putative negative growth factor encoded by the short gastrulation gene. Genes Dev. 1994;8:2602–16. doi: 10.1101/gad.8.21.2602. [DOI] [PubMed] [Google Scholar]
  14. Galli LM, Barnes TL, Secrest SS, Kadowaki T, Burrus LW. Porcupine-mediated lipid-modification regulates the activity and distribution of Wnt proteins in the chick neural tube. Development. 2007;134:3339–48. doi: 10.1242/dev.02881. [DOI] [PubMed] [Google Scholar]
  15. Giot L, Bader JS, Brouwer C, Chaudhuri A, Kuang B, Li Y, Hao YL, Ooi CE, Godwin B, Vitols E, Vijayadamodar G, Pochart P, Machineni H, Welsh M, Kong Y, Zerhusen B, Malcolm R, Varrone Z, Collis A, Minto M, Burgess S, McDaniel L, Stimpson E, Spriggs F, Williams J, Neurath K, Ioime N, Agee M, Voss E, Furtak K, Renzulli R, Aanensen N, Carrolla S, Bickelhaupt E, Lazovatsky Y, DaSilva A, Zhong J, Stanyon CA, Finley RL, Jr, White KP, Braverman M, Jarvie T, Gold S, Leach M, Knight J, Shimkets RA, McKenna MP, Chant J, Rothberg JM. A protein interaction map of Drosophila melanogaster. Science. 2003;302:1727–36. doi: 10.1126/science.1090289. [DOI] [PubMed] [Google Scholar]
  16. Givan SA, Sprague GF., Jr The ankyrin repeat-containing protein Akr1p is required for the endocytosis of yeast pheromone receptors. Mol Biol Cell. 1997;8:1317–27. doi: 10.1091/mbc.8.7.1317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Greco V, Hannus M, Eaton S. Argosomes: a potential vehicle for the spread of morphogens through epithelia. Cell. 2001;106:633–45. doi: 10.1016/s0092-8674(01)00484-6. [DOI] [PubMed] [Google Scholar]
  18. Huang K, Yanai A, Kang R, Arstikaitis P, Singaraja RR, Metzler M, Mullard A, Haigh B, Gauthier-Campbell C, Gutekunst CA, Hayden MR, El-Husseini A. Huntingtin-interacting protein HIP14 is a palmitoyl transferase involved in palmitoylation and trafficking of multiple neuronal proteins. Neuron. 2004;44:977–86. doi: 10.1016/j.neuron.2004.11.027. [DOI] [PubMed] [Google Scholar]
  19. Linder ME, Deschenes RJ. New insights into the mechanisms of protein palmitoylation. Biochemistry. 2003;42:4311–20. doi: 10.1021/bi034159a. [DOI] [PubMed] [Google Scholar]
  20. Linder ME, Deschenes RJ. Model organisms lead the way to protein palmitoyltransferases. J Cell Sci. 2004;117:521–6. doi: 10.1242/jcs.00989. [DOI] [PubMed] [Google Scholar]
  21. Mann RK, Beachy PA. Novel lipid modifications of secreted protein signals. Annu Rev Biochem. 2004;73:891–923. doi: 10.1146/annurev.biochem.73.011303.073933. [DOI] [PubMed] [Google Scholar]
  22. Miura GI, Buglino J, Alvarado D, Lemmon MA, Resh MD, Treisman JE. Palmitoylation of the EGFR ligand Spitz by Rasp increases Spitz activity by restricting its diffusion. Dev Cell. 2006;10:167–76. doi: 10.1016/j.devcel.2005.11.017. [DOI] [PubMed] [Google Scholar]
  23. Miura GI, Treisman JE. Lipid modification of secreted signaling proteins. Cell Cycle. 2006;5:1184–8. doi: 10.4161/cc.5.11.2804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mizutani CM, Nie Q, Wan FY, Zhang YT, Vilmos P, Sousa-Neves R, Bier E, Marsh JL, Lander AD. Formation of the BMP activity gradient in the Drosophila embryo. Dev Cell. 2005;8:915–24. doi: 10.1016/j.devcel.2005.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ohyama T, Verstreken P, Ly CV, Rosenmund T, Rajan A, Tien AC, Haueter C, Schulze KL, Bellen HJ. Huntingtin-interacting protein 14, a palmitoyl transferase required for exocytosis and targeting of CSP to synaptic vesicles. J Cell Biol. 2007;179:1481–96. doi: 10.1083/jcb.200710061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Pryciak PM, Hartwell LH. AKR1 encodes a candidate effector of the G beta gamma complex in the Saccharomyces cerevisiae pheromone response pathway and contributes to control of both cell shape and signal transduction. Mol Cell Biol. 1996;16:2614–26. doi: 10.1128/mcb.16.6.2614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Resh MD. Fatty acylation of proteins: new insights into membrane targeting of myristoylated and palmitoylated proteins. Biochim Biophys Acta. 1999;1451:1–16. doi: 10.1016/s0167-4889(99)00075-0. [DOI] [PubMed] [Google Scholar]
  28. Resh MD. Palmitoylation of ligands, receptors, and intracellular signaling molecules. Sci STKE. 2006;2006:re14. doi: 10.1126/stke.3592006re14. [DOI] [PubMed] [Google Scholar]
  29. Roth AF, Feng Y, Chen L, Davis NG. The yeast DHHC cysteine-rich domain protein Akr1p is a palmitoyl transferase. J Cell Biol. 2002;159:23–8. doi: 10.1083/jcb.200206120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Roth AF, Wan J, Bailey AO, Sun B, Kuchar JA, Green WN, Phinney BS, Yates JR, 3rd, Davis NG. Global analysis of protein palmitoylation in yeast. Cell. 2006;125:1003–13. doi: 10.1016/j.cell.2006.03.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Shimmi O, O’Connor MB. Physical properties of Tld, Sog, Tsg and Dpp protein interactions are predicted to help create a sharp boundary in Bmp signals during dorsoventral patterning of the Drosophila embryo. Development. 2003;130:4673–82. doi: 10.1242/dev.00684. [DOI] [PubMed] [Google Scholar]
  32. Simpson F, Kerr MC, Wicking C. Trafficking, development and hedgehog. Mech Dev. 2009;126:279–88. doi: 10.1016/j.mod.2009.01.007. [DOI] [PubMed] [Google Scholar]
  33. Singaraja RR, Hadano S, Metzler M, Givan S, Wellington CL, Warby S, Yanai A, Gutekunst CA, Leavitt BR, Yi H, Fichter K, Gan L, McCutcheon K, Chopra V, Michel J, Hersch SM, Ikeda JE, Hayden MR. HIP14, a novel ankyrin domain-containing protein, links huntingtin to intracellular trafficking and endocytosis. Hum Mol Genet. 2002;11:2815–28. doi: 10.1093/hmg/11.23.2815. [DOI] [PubMed] [Google Scholar]
  34. Smotrys JE, Linder ME. Palmitoylation of intracellular signaling proteins: regulation and function. Annu Rev Biochem. 2004;73:559–87. doi: 10.1146/annurev.biochem.73.011303.073954. [DOI] [PubMed] [Google Scholar]
  35. Srinivasan S, Rashka KE, Bier E. Creation of a Sog morphogen gradient in the Drosophila embryo. Dev Cell. 2002;2:91–101. doi: 10.1016/s1534-5807(01)00097-1. [DOI] [PubMed] [Google Scholar]
  36. Stowers RS, Isacoff EY. Drosophila huntingtin-interacting protein 14 is a presynaptic protein required for photoreceptor synaptic transmission and expression of the palmitoylated proteins synaptosome-associated protein 25 and cysteine string protein. J Neurosci. 2007;27:12874–83. doi: 10.1523/JNEUROSCI.2464-07.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wan J, Roth AF, Bailey AO, Davis NG. Palmitoylated proteins: purification and identification. Nat Protoc. 2007;2:1573–84. doi: 10.1038/nprot.2007.225. [DOI] [PubMed] [Google Scholar]
  38. Yanai A, Huang K, Kang R, Singaraja RR, Arstikaitis P, Gan L, Orban PC, Mullard A, Cowan CM, Raymond LA, Drisdel RC, Green WN, Ravikumar B, Rubinsztein DC, El-Husseini A, Hayden MR. Palmitoylation of huntingtin by HIP14 is essential for its trafficking and function. Nat Neurosci. 2006;9:824–31. doi: 10.1038/nn1702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Yu K, Kang K-W, Heine P, Pyati U, Srinivasan S, Biehs B, Kimelman D, Bier E. Cysteine repeat domains and adjacent sequences determine distinct BMP modulatory activities of the Drosophila Sog protein. Genetics. 2004;166:1323–36. doi: 10.1534/genetics.166.3.1323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Yu K, Srinivasan S, Shimmi O, Biehs B, Rashka KE, Kimelman D, O’Connor MB, Bier E. Processing of the Drosophila Sog protein creates a novel BMP inhibitory activity. Development. 2000;127:2143–54. doi: 10.1242/dev.127.10.2143. [DOI] [PubMed] [Google Scholar]
  41. Yu K, Sturtevant MA, Biehs B, Francois V, Padgett RW, Blackman RK, Bier E. The Drosophiladecapentaplegic and short gastrulation genes function antagonistically during adult wing vein development. Development. 1996;122:4033–44. doi: 10.1242/dev.122.12.4033. [DOI] [PubMed] [Google Scholar]
  42. Zhai L, Chaturvedi D, Cumberledge S. Drosophila Wnt-1 undergoes a hydrophobic modification and is targeted to lipid rafts, a process that requires porcupine. J Biol Chem. 2004;279:33220–7. doi: 10.1074/jbc.M403407200. [DOI] [PubMed] [Google Scholar]

RESOURCES