Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 4.
Published in final edited form as: Dev Cell. 2019 Sep 19;51(3):387–398.e4. doi: 10.1016/j.devcel.2019.08.014

Syndecan-1 mediates sorting of soluble lipoprotein lipase with sphingomyelin-rich membrane in the Golgi apparatus

Emma L Sundberg 1, Yongqiang Deng 1, Christopher G Burd 1,§,†
PMCID: PMC6832887  NIHMSID: NIHMS1538875  PMID: 31543446

Abstract

In the secretory pathway, budding of vesicular transport carriers from the trans Golgi network (TGN) must coordinate specification of lipid composition with selection of secreted proteins. We elucidate a mechanism of soluble protein cargo sorting into secretory vesicles with a sphingomyelin rich membrane; the integral membrane proteoglycan Syndecan-1 (SDC1) acts as a sorting receptor, capturing the soluble enzyme Lipoprotein Lipase (LPL) during export from the TGN. Sorting of LPL requires bivalent interactions between LPL and SDC1-linked heparan sulfate chains and between LPL and Golgi membrane. Physical features of the SDC1 transmembrane domain, rather than specific sequence, confer targeting of SDC1 and bound LPL into the sphingomyelin secretion pathway. This study establishes that physicochemical properties of a protein transmembrane domain that drive lateral heterogeneity of the plasma membrane, also operate at the TGN to confer sorting of an integral membrane protein and its ligand within the biosynthetic secretory pathway.

eTOC Blurb

The trans Golgi network coordinates the specification of lipid composition with incorporation of secreted proteins into diverse vesicular carriers. Sundberg et al. demonstrate that the physicochemical properties of the Syndecan-1 transmembrane domain mediate sorting of this protein and a bound cargo, lipoprotein lipase, into secretory vesicles enriched in sphingomyelin.

Introduction

The trans Golgi network (TGN) is recognized as an important protein and lipid sorting center, responsible for packaging and targeting cargos to the various intracellular organelles or to the cell surface (Surma, et al., 2012). Protein cargos that transit the TGN differ widely in structure, composition, function, and ultimate resident organelle, implying the existence of multiple sorting pathways to accommodate diverse protein cargos (Boncompain and Weigel, 2018; De Matteis and Luini, 2008). However, the mechanisms by which the cell elaborates different sorting pathways are largely unknown. It is unclear how the lipid content of secretory vesicles is specified and the extent to which lipid composition influences the selection of proteins to be incorporated into vesicular carriers that bud from the TGN.

Early work exploring the interplay between membrane lipid composition and the sorting of integral membrane proteins related the length of membrane spanning segments to their organellar localization along the secretory pathway (Banfield, 2011). The plasma membrane is the thickest cellular membrane due to its high relative proportions of sphingolipid and cholesterol, and studies have shown that proteins with a longer TMD sequence have a higher rate of export from the Golgi, driving their accumulation at the plasma membrane (Dukhovny, et al., 2009). Secretory vesicles purified from yeast (Saccharomyces cerevisiae) cells are enriched in cholesterol and sphingolipid relative to Golgi membrane (Surma, et al., 2011; Klemm, et al., 2009), consistent with longstanding proposals of lipid sorting at the TGN during secretion (Simons and Ikonen, 1997; Bretscher and Munro, 1993). Importantly, the length and sequence of a TMD can also determine sorting of an integral membrane protein to the apical or basolateral domains of the plasma membrane of polarized epithelial cells (Barman and Nayak, 2000; Lin, et al., 1998). Thus, TMDs can dictate both sorting within and export from the TGN.

In addition, lipid modifications on proteins, especially glycosylphosphatidylinositol (GPI) and palmitate, promote their coalescence with sphingolipid and cholesterol-rich membrane domains in vitro and during export from the Golgi in vivo (Lorent, et al., 2017; Levental, et al., 2010; Brown and Rose, 1992). As the assembly of lipid-protein clusters in the plasma membrane drives important biological processes (Sezgin, et al., 2017), including signaling events (Simons and Toomre, 2000), endocytosis (Fuki, et al., 2000), and pathogen evasion (Manes, et al., 2003), much remains to be explored regarding the outcomes of protein-lipid nanodomains in intracellular compartments and the roles of such domains in molecular sorting in the secretory pathway.

An additional question in the field of protein secretion focuses on how soluble secreted proteins are captured and enriched in Golgi-derived secretory vesicles (Kienzle and von Blume, 2014). The paradigm for the segregation of specific soluble proteins from the bulk milieu within the secretory pathway is receptor-mediated sorting, exemplified by the capture of lysosomal proteins via mannose 6-phosphate receptors in the TGN (Hille-Rehfeld, 1995), followed by concentration of receptor-cargo complexes by oligomeric coat protein complexes that recognize sorting signals in the cytoplasmic domains of receptors. Secretory vesicles, however, are not coated, leading to the proposal that export of secreted proteins from the TGN is largely non-specific, termed “bulk flow” (Pfeffer and Rothman, 1987). Nonetheless, strong evidence indicates that some proteins are exported from the TGN in distinct carriers, a phenomenon that has been most studied in polarized cells where soluble and integral membrane proteins are separately packaged for targeting to either the basolateral or apical domain. The mechanisms for sorting of soluble secreted proteins are largely unknown.

Sphingomyelin (SM), the most abundant sphingolipid in cells, is synthesized in the lumenal leaflet of the membrane of the trans Golgi and TGN and trafficked to the plasma membrane (van Meer, et al., 2008). Using an engineered SM-binding protein (EQ-SM), we have previously shown that SM is enriched in a subset of secretory transport carriers that bud from the TGN, constituting a branch of the secretory pathway that is specialized for transporting SM from the Golgi to the cell surface; we term this the sphingomyelin secretion (SMS) pathway (Deng, et al., 2016). Sphingomyelin-rich secretory vesicles, identified by the presence of EQ-SM, are preferentially enriched in particular proteins, including GPI-anchored proteins, Cab45, and lysozyme C (Deng, et al., 2018; Deng, et al., 2016). Previously, we employed a proximity-based biotinylation approach to identify native proteins that are secreted via the SMS pathway, where an EQ-SM-APEX2 fusion protein was used to modify proteins in Golgi-derived vesicles containing EQ-SM-APEX2 (Deng, et al., 2018). One of the proteins that we identified is lipoprotein lipase (LPL), an enzyme that degrades triglycerides in triglyceride-rich lipoproteins circulating in the plasma. Many studies investigating the regulation of LPL activity have focused on the movement of secreted LPL through the interstitial space of tissues and across capillary endothelial cells to its site of function in the capillary lumen, (He, et al., 2018; Davies, et al., 2012a). However, the mechanisms that underpin the upstream sorting and trafficking of LPL through the secretory pathway of myocytes and adipocytes, the primary LPL producers, are not known.

Here we demonstrate that LPL is secreted via the SMS pathway. Sorting into the SMS pathway requires the interaction of LPL with heparan sulfate proteoglycans (HSPGs) within the Golgi. We identify Syndecan-1 (SDC1), a membrane-spanning HSPG, as the sorting receptor responsible for targeting LPL into the SMS pathway. Sorting of SDC1 and SDC1-LPL complexes into the SMS pathway is dependent on the physical properties, but not strictly on the sequence, of the SDC1 transmembrane domain, which is sufficient to drive its incorporation into sphingomyelin-rich secretory vesicles upon exit from the TGN.

Results

Identification of LPL as a cargo in the SMS pathway

To determine if LPL is preferentially secreted by vesicles of the SMS pathway, total internal reflection fluorescence (TIRF) microscopy time lapse imaging was used to observe cells co-expressing pHluorin-LPL (pH-LPL) and EQ-SM-mKate2 or EQ-sol-mKate2, a soluble control probe that is secreted via bulk flow (Deng, et al., 2016). Exocytic events were identified by the flash of fluorescence emitted by pHluorin-LPL when it encounters the neutral pH of cell culture medium. At the moment of fusion with the plasma membrane, the presence of EQ-SM-mKate2 or EQ-sol-mKate2 in the fused vesicle was then scored; example exocytic events are shown (Fig. 1A). In this assay, sorting of secreted proteins via the SMS pathway is indicated by increased proportion of exocytic events in which vesicles contain pHluorin-LPL and EQ-SM-mKate2, compared to EQ-sol-mKate2. We observed that 69 ± 14% (n=251; 13 cells) of pH-LPL vesicles also contained EQ-SM-mKate2, whereas only 35 ± 16% (n=141; 10 cells) of pH-LPL vesicles contained EQ-sol-mKate2 (Fig. 1B). This differential shows that LPL is secreted by vesicles containing EQ-SM more frequently (p< 0.0001) than by vesicles containing EQ-sol, confirming the previously published proteomics study of EQ-SM-containing vesicles (Deng et al., 2018) and indicating that LPL is secreted via the SMS pathway. Ongoing SM synthesis is not required for LPL sorting, as similar values for LPL co-secretion with EQ-SM and EQ-sol were observed with cells depleted of SMS1 and SMS2 SM synthases by siRNA treatment (Fig. S1). This contrasts with sorting of two other SMS pathway cargos, Cab45 and lysozyme C, that are sorted via a calcium-dependent mechanism that requires ongoing SM synthesis (Deng, et al., 2018).

Figure 1. LPL exocytosis in vesicles of the SMS pathway requires sorting determinants in both domains.

Figure 1.

Open in a new tab

(A) Time-lapse gallery of LPL, EQ-SM, and EQ-sol exocytosis. Galleries show example exocytic events of pHluorin-LPL and EQ-SM-mKate2 or EQ-sol-mKate2 captured by TIRFM. The corresponding graphs show total fluorescence intensity for each channel of the region of interest in each frame over time. (B) Full-length LPL, and not individual domains of LPL, is co-secreted with EQ-SM compared to EQ-sol. The mean proportions of exocytic events observed in at least 2 independent TIRFM experiments (± SD), where pHluorin-LPL, pHluorin-LPL*, pHluorin-lipase, or pHluorin-PLAT containing vesicles also contained EQ-SM-mKate2 or EQ-sol-mKate2 are indicated (251 events/13 cells for LPL+EQ-SM, 141 events/10 cells for LPL+EQ-sol, 249 events/13 cells for LPL*+EQ-SM, 246 events/13 cells for LPL*+EQ-sol, 219 events/15 cells for lipase+EQ-SM, 161 events/10 cells for lipase+EQ-sol, 554 events/17 cells for PLAT+EQ-SM, 357 events/12 cells for PLAT+EQ-sol). LPL* indicates a catalysis-deficient mutant form (S132A), and “lipase” and “PLAT” refer to LPL constructs containing residues 28–340 and 328–476, respectively. **** indicates p< 0.0001, n.s. indicates not significant (p>0.05). (C) Model of LPL domains. LPL is composed of two domains; lipase domain is yellow and PLAT domain is blue. The positions of four aromatic residues in the unstructured loop that were mutated (Y387A, W390A, W393A, W394A) in LPL-AR mutant protein are indicated with red circles. The heparin binding (HB) residues in clusters HB1 (K403, R405, D407), HB2 (K296, R297, K300), and HB3 (R279, K280, R282) are shown as magenta sticks. The structure was derived from PDB: 6E7K (Birrane, et al., 2019).

Due to concerns that over-expression of LPL might alter lipid and metabolic homeostasis because of its activity toward lipoprotein in the culture serum, a mutation in the catalytic triad was introduced to convert Serine 132 to Alanine, a mutation that was shown to reduce lipolytic activity to less than 8% of wild type function (Wong, et al., 1994). TIRFM sorting experiments showed that 71 ± 14% (n=249; 13 cells) of pH-LPL(S132A) vesicles also contained EQ-SM-mKate2, whereas only 38 ± 18% (n=246; 13 cells) of pH-LPL(S132A) vesicles contained EQ-sol-mKate2 (Fig. 1B). These data indicate that the (S132A) mutation does not affect sorting of LPL; hence, the S132A mutation was maintained in subsequent experiments.

Identification of structural elements in LPL that are required for sorting

We next sought to identify structural elements in LPL that promote its sorting into the SMS pathway. LPL contains two structural domains: a catalytic lipase domain, and a PLAT (polycystin-1, lipoxygenase, alpha toxin; LH2=lipoxygenase homology 2; also called, LH2) domain (Fig. 1C) (Birrane, et al., 2019). To test whether a single domain within LPL is sufficient to promote sorting into the SMS pathway, we generated truncation mutants of the lipase domain and the PLAT domain and introduced these constructs into TIRFM-based exocytic experiments (Fig. 1B). There was no difference between the proportions of pH-lipase exocytic events with either EQ-SM or EQ-sol, with 31 ± 16% (n=219; 15 cells) of pH-lipase vesicles containing EQ-SM and 33 ± 25% (n=161; 10 cells) of pH-lipase vesicles containing EQ-sol. In addition, there was no statistically significant difference in pH-PLAT sorting with EQ-SM compared to EQ-sol: 40 ± 14% (n=554; 17 cells) of pH-PLAT vesicles contained EQ-SM, and 30 ± 14% (n=357; 12 cells) of pH-PLAT vesicles contained EQ-sol. These data demonstrate that structural elements in both the lipase and PLAT domains of LPL are required to target the protein into the SMS pathway.

In other proteins, such as α-toxin, the PLAT domain is implicated in membrane binding (Naylor, 1998). Moreover, the PLAT domain bears structural resemblance to actinoporins, including equinatoxin (from which EQ-SM is derived). Of note, both the LPL PLAT domain and equinatoxin possess a cluster of aromatic residues positioned on a loop facing away from the core of the protein (Fig. 1C). In actinoporins, these aromatic residues form a membrane binding face that recognizes SM (Deng, et al., 2016; Athanasiadis, et al., 2001), leading us to speculate that the aromatic residues in the PLAT domain confer binding to membrane, possibly through recognition of SM. A liposome co-sedimentation assay was used to directly test this hypothesis. To this end, purified recombinant PLAT or PLAT-AR, in which the residues Y387, W390, W393, and W394 were mutated to Alanine, was incubated with liposomes containing 20 mol percent SM (and 60 mol percent phosphatidylcholine and 20 mol percent cholesterol) or 0 mol percent SM (with phosphatidylcholine substituted for SM). Strikingly, the PLAT domain binds membrane with high avidity and this interaction requires the loop of aromatic residues (Fig. 2A). However, no specificity for SM is observed; in the absence of SM, 84 ± 14% of PLAT binds liposomes compared to 80 ± 6% in the presence of SM (Fig. 2B). For PLAT-AR, this binding is reduced to 18 ± 7% and 4 ± 5% respectively. This result differs markedly from the binding behavior of EQ-SM, where there is 4-fold increase in the membrane-bound protein pool in the presence of SM (Fig. 2B), raising the question of whether the membrane interaction mediated via the PLAT domain is required for LPL sorting into the SMS pathway.

Figure 2. Interaction of the LPL PLAT domain with membrane is required for sorting into the SMS pathway.

Figure 2.

Open in a new tab

(A) LPL PLAT domain binds synthetic liposomes. Recombinant PLAT, PLAT-AR, and EQ-SM protein were incubated with liposomes containing 20% SM or PC (and 20% cholesterol), and liposomes were collected by centrifugation. Bound pellet (P) and unbound supernatant (S) fractions were visualized by Coomassie blue staining and quantified. (B) Quantification of liposome binding experiment. The proportion of bound protein out of the total is plotted as a mean value (± SD) for three independent experiments. (C) Aromatic residues in the PLAT domain are required for LPL co-secretion with EQ-SM. TIRFM-based exocytosis experiments were used to score the proportion of secretory vesicles containing pHluorin-LPL-AR and EQ-SM-mKate2 or EQ-sol-mKate2, without and with expression of GPIHBP1-Flag (means ± SD in at least 2 independent experiments; 185 events/10 cells for pH-LPL-AR+EQ-SM, 152 events/11 cells for pH-LPL-AR+EQ-sol, 179 events/12 cells for pH-LPL-AR+EQ-SM+GPIHBP1-Flag, 140 events/10 cells for pH-LPL-AR+EQ-sol+GPIHBP1-Flag).

To test the requirement of the PLAT aromatic loop in LPL sorting, the PLAT-AR mutation described above was incorporated into full-length LPL to yield the construct LPL-AR. In TIRFM sorting experiments, whereas 71 ± 14% of vesicles containing native pH-LPL also contained EQ-SM-mKate2 (Fig. 1B), only 57 ± 14% (n=185; 10 cells) containing mutant pH-LPL-AR also contained EQ-SM (Fig. 2C). Importantly, this level of sorting is similar to that observed for pH-LPLAR with EQ-sol (53 ± 14% (n=152; 11 cells)). These observations indicate that LPL targeting into the SMS pathway requires the aromatic loop in the PLAT domain. In combination with the liposome binding results, these data show that the PLAT domain likely promotes LPL sorting and secretion by binding membrane; however, given the PLAT domain’s lack of SM-specificity, interaction with membrane is not sufficient to direct LPL into nascent carriers of the SMS pathway in the TGN.

Binding of heparan sulfate proteoglycans by LPL is required for sorting into the SMS pathway

Given our observations that the PLAT domain is required but insufficient for sorting into the SMS pathway and that this domain does not specifically recognize SM-containing membrane, we considered the hypothesis that a sorting receptor is responsible for directing LPL into the SMS pathway. Recent studies of LPL trafficking have shown that LPL binds the GPI-anchored protein GPIHBP1 (Davies, et al., 2010), resulting in accumulation of LPL on the luminal surface of capillary endothelial cells (Davies, et al., 2012b). However, it is unclear if GPIHBP1 contributes to the secretion of LPL. Of interest, our previous work identified GPI-anchored proteins as cargoes of the SMS pathway (Deng, et al., 2018; Deng, et al., 2016), raising the possibility that GPIHBP1 could mediate LPL sorting in the secretory pathway. However, GPIHBP1 mRNA could not be detected by reverse transcription-polymerase chain reaction assays in the HeLa cell line used for our studies, indicating that it does not contribute to biosynthetic sorting and secretion of LPL in these cells. Interestingly, exogenous expression of GPIHBPI is nevertheless able to rescue the sorting defect observed for the LPL aromatic residue mutant, LPL-AR (Fig. 2C). This suggests that although GPIHBP1 is not the primary sorting receptor for LPL, when introduced into LPL-producing cells, GPIHBP1 is able to promote LPL sorting into the SMS pathway. Intracellular trafficking of LPL by GPIHBP1 supports the proposal that GPIHBP1 mediates transcytosis of LPL in capillary endothelial cells (Davies, et al., 2012b; Davies, et al., 2010).

LPL binds avidly to heparan sulfate proteoglycans (HSPGs), a group of glycoproteins with well-documented roles in the binding and capture of secreted LPL (Mead, et al., 2002) and many secreted growth factors (Nagarajan, et al., 2018). Curiously, we observe that secreted LPL is retained at the site of vesicle fusion (Fig. 3A), despite the absence of GPIHBP1 expression in the cells used. We speculate that binding of LPL to HSPGs within the extracellular matrix contributes to LPL retention on or near the plasma membrane after secretion. Supporting the role of HSPGs in this process, we show with TIRF microscopy that when cells expressing GFP-LPL are washed with free heparin (a more highly sulfated GAG than those found within the ECM), the cell surface-associated pool of LPL is lost (Fig. 3A).

Figure 3. Interaction with heparin sulfate proteoglycans retains LPL on the cell surface and is required for LPL secretion via the SMS pathway.

Figure 3.

Open in a new tab

(A) LPL localization in HeLa cells. Cells were transfected with GFP-LPL and visualized by TIRFM 16h after transfection. After secretion, LPL remains associated with the cell surface in untreated cells. Treatment of cells with 5 U/mL heparin for 5 minutes eliminates LPL cell surface localization. Micrograph shows one z-slice. Scale bars represent 10 μm. (B) Production of HSPGs is required for co-secretion of LPL and EQ-SM. TIRFM was used to score the proportion of pHluorin-LPL containing vesicles that also contained EQ-SM-mKate2 or EQ-sol-mKate2 in cells pretreated with DMSO or 2.5 mM xyloside for 48h. The means ± SD in at least 2 independent experiments are shown (222 events/12 cells for pH-LPL+EQ-SM+DMSO, 166 events/11 cells for pH-LPL+EQ-sol+DMSO, 160 events/11 cells for pH-LPL+EQ-SM+xyloside, 130 events/10 cells for pH-LPL+EQ-sol+xyloside). The diagrams below illustrate the effect of xyloside treatment on HSPG biosynthesis. (C) Mutation of heparin binding residues reduces LPL co-secretion with EQ-SM. The heparin binding residues in cluster 1 were mutated to Alanines and this construct, pHluorin-LPL-HB1, was used in TIRFM where co-secretion with EQ-SM-mKate2 and EQ-sol-mKate2 was scored. The means ± SD in at least 2 independent experiments are indicated (256 events/14 cells for LPL-HB1+EQ-SM, 190 events/11 cells for LPL-HB1+EQ-sol).

HSPG chains are elaborated in the Golgi apparatus, raising the possibility that sorting of LPL in the secretory pathway may be influenced by binding of LPL to HSPGs in this organelle. To address this, we first tested if chemical perturbation of HSPG synthesis affects co-secretion of pH-LPL with EQ-SM by TIRF microscopy (Fig. 3B). Cells expressing pH-LPL with EQ-SM were pretreated with xyloside, which inhibits addition of GAGs to HSPG protein cores (Chanat and Huttner, 1991) (Fig. 3B). In cells treated with xyloside (48 hours), co-secretion of pH-LPL and EQ-SM is reduced from 72 ± 14% (n=222; 12 cells) in control cells incubated with vehicle (DMSO) to 44 ± 18% (n=160; 11 cells) of exocytic events, approaching the level co-secretion observed for pH-LPL and EQ-sol (36 ± 14% (n=130; 10 cells)) (Fig. 3B). These data demonstrate that mature HSPGs are required for sorting of LPL into the SMS pathway.

Prior studies of LPL-HSPG interaction identified several clusters of positively charged residues on the surface of LPL that contribute additively to heparin binding in vitro (Birrane, et al., 2019; Sendak and Bensadoun, 1998; van Tilbeurgh, et al., 1994; Hata, et al., 1993) (Fig. 1C). We therefore tested if mutations that substitute Alanine and Asparagine in place of positively charged residues in these binding sites affect sorting of LPL into the SMS pathway. Mutations in all three heparin binding sites resulted in a mutant protein that is retained in the ER, so its secretion could not be assayed. However, mutations to Asparagine in heparin binding cluster 1 (K403, R405, and K407; termed LPL-HB1) were tolerated and in TIRFM sorting experiments pH-LPL-HB1 was co-secreted with EQ-SM in 46 ± 23% (n=256; 14 cells) of exocytic events and with EQ-sol 39 ± 14% (n=190; 11 cells) of exocytic events (Fig. 3C). Taken together, these results show that GAG-decorated HSPGs contribute to LPL sorting, likely due to a direct interaction between HSPGs and the heparin binding residues on LPL.

Syndecan-1 acts as a sorting receptor for LPL

The human genome encodes at least 17 HSPGs with diverse expression and cellular functions (Sarrazin, et al., 2011), so we wondered if all or a subset of proteoglycans are responsible for LPL sorting. We first focused on the two families of membrane-bound proteoglycans: the GPI-anchored glypicans and the membrane-spanning syndecans. As shown previously, heparin is effective at removing secreted LPL from the surface of cells, consistent with recognition of HSPGs by secreted LPL. To determine if glypicans contribute to LPL retention at the cell surface, we treated intact cells with Phospholipase C, which cleaves the phosphodiester headgroup bond of phospholipids, eliciting the release of GPI-anchored proteins from the cell surface. This treatment had no effect on LPL cell surface localization (Fig. S2).

We therefore focused on syndecans, a family of single pass TMD HSPGs, as potential sorting receptors for LPL. There are four members of the syndecan family whose expression is normally restricted to particular developmental stages and/or specific tissues. Quantitative PCR analysis revealed that all four syndecans (SDC1, 2, 3, 4) are expressed in the HeLa cell line used in these studies. To evaluate the role of individual syndecans in LPL sorting, siRNAs were designed to deplete each syndecan (Fig. S3) and then LPL secretion via the SMS pathway was scored. Only knockdown of SDC1 reduced LPL co-sorting with EQ-SM; co-secretion of pH-LPL and EQ-SM was equivalent to that with EQ-sol (59 ± 18% (n=223, 11 cells) versus 63 ± 15% (n=160, 10 cells), respectively) (Fig. 4A). As all four syndecans are decorated with heparan sulfate, this result suggests the syndecan-1 (SDC1) core protein also contributes to LPL sorting.

Figure 4: Identification of Syndecan-1 as an LPL sorting receptor.

Figure 4:

Open in a new tab

(A) Syndecan-1 is uniquely required for co-secretion of LPL and EQ-SM. Cells were transfected with a non-targeting siRNA or with siRNAs targeting Syndecan-1, Syndecan-2, Syndecan-3, or Syndecan-4 for 72h before TIRFM exocytosis experiments. Exocytic cargo loads were evaluated and the means ± SD in at least 2 independent experiments are shown (208 events/12 cells for pH-LPL+EQ-SM+si-C, 222 events/12 cells for pH-LPL+EQ-sol+si-C, 172 events/11 cells for pH-LPL+EQ-SM+si-SDC1, 160 events/10 cells for pH-LPL+EQ-sol+si-SDC1, 149 events/11 cells for pH-LPL+EQ-SM+si-SDC2, 196 events/12 cells for pH-LPL+EQ-sol+si-SDC2, 123 events/10 cells for pH-LPL+EQ-SM+si-SDC3, 140 events/10 cells for pH-LPL+EQ-sol+si-SDC3, 176 events/11 cells for pH-LPL+EQ-SM+si-SDC4, 154 events/12 cells for pH-LPL+EQ-sol+si-SDC4). (B) LPL and SDC1 copurify. Cells were transfected with plasmids encoding GFP-LPL and SDC1-Flag for 24h. Cell lysates were prepared and incubated with Flag beads for 1h. A representative Western blot shows input and immunipurified (IP) fractions for cells co-expressing GFP-LPL and SDC1-Flag, cells expressing only GFP-LPL, in vitro mixing of cells independently expressing GFP-LPL or SDC1-Flag. (C) LPL and SDC1 are co-secreted in the same vesicle population. TIRFM was used to score the co-secretion of pHluorin-LPL and pHluorin-hGH with RUSH-Ruby-SDC1. Cargo loads were scored 15–60 minutes after release of RUSH-Ruby-SDC1 by biotin addition. The means ± SD in at least 2 independent experiments are shown (157 events/11 cells for pH-LPL+RUSH-Ruby-SDC1, 192 events/11 cells for pH-hGH+RUSH-Ruby-SDC1). *** indicates p< 0.001

These data lead us to speculate that HSPG-decorated SDC1 functions as a sorting receptor for LPL in the secretory pathway. A prediction of this hypothesis is that SDC1 and LPL should be associated within the cell and this was evaluated by co-immunoprecipitation of exogenously expressed GFP-LPL and Flag epitope-tagged SDC1 from cell lysates (Fig. 4B). Indeed, co-purification of SDC1 with LPL was observed and this required co-expression of both proteins in the same cell. A second requirement of our hypothesis is that SDC1 and LPL should populate the same secretory carriers. We therefore used TIRFM experiments to quantify co-secretion of SDC1 and LPL (Fig. 4C). For these experiments, endogenous SDC1 was depleted by siRNA treatment prior to re-expression of fluorescently tagged siRNA-resistant Ruby-SDC1, and co-secretion with pH-LPL was determined. To aid in detection of Ruby-SDC1 exocytosis, we used the “retention under selective hooks” (RUSH) system to retain and release Ruby-SDC1 from the ER into the secretory pathway prior to TIRFM imaging (Sundberg, et al., 2019; Boncompain, et al., 2012). Using this approach, we found that pH-LPL and RUSH-Ruby-SDC1 are co-secreted in 70 ± 15% (n=157, 11 cells) of exocytic carriers. In comparison, a soluble pHluorin-human Growth Hormone fusion construct (pH-hGH) is co-secreted with RUSH-Ruby-SDC1 in only 41 ± 17% (n=192, 11 cells) of exocytic events. These data, showing that that SDC1 binds to and is co-secreted with LPL in secretory vesicles, support the hypothesis that SDC1 functions as a sorting receptor for LPL in the secretory pathway.

Sorting of Syndecan-1 is conferred by its transmembrane domain

The identification of SDC1 as a sorting receptor that mediates LPL secretion via the SMS pathway raised a new question - what mechanism drives SDC1 sorting into the SMS pathway? SDC1 consists of three domains: an extracellular domain containing attachment sites for five GAG chains, a transmembrane domain (TMD), and a cytoplasmic domain (Fig. 5A). Inspection of the amino acid sequence of the SDC1 TMD reveals a set of features that have been identified as characteristics that promote partitioning of TMD sequences into liquid ordered domains of plasma membrane-derived vesicles (Lorent, et al., 2017): a preponderance of amino acids with small side chains, a (predicted) palmitoylated Cysteine, and an adequate length to span a liquid ordered bilayer in plasma membrane derived vesicles. We hypothesized that these features of the SDC1 TMD are required for sorting of the protein into the SMS pathway. To test this hypothesis, we compared the sorting of ‘native’ SDC1 to a mutant where the endogenous TMD is replaced by an equivalent number of Leucine residues (referred to as SDC1-TMD_L) and a Cysteine residue (Fig. 5A). The RUSH system was used to retain SDC1 constructs in the ER prior to release into the secretory pathway. Both ‘native’ SDC1 and SDC1-TMD_L are readily released from the ER and delivered to the PM after the addition of biotin (Fig. 5B). TIRFM sorting experiments show that SDC1 is preferentially co-secreted with EQ-SM (71 ± 12% (n= 264, 11 cells) of exocytic events) compared to EQ-sol (47 ± 19% (n= 281, 11 cells) of exocytic events) (Fig. 5C). In striking contrast, SDC1-TMD_L shows no preference for sorting into the SMS pathway; it is co-secreted with EQ-SM in 37 ± 22% (n= 260, 11 cells) exocytic events and with EQ-sol 44 ± 24% (n= 270, 11 cells) events. These data provide compelling evidence that the TMD sequence of SDC1 is required for secretion of SDC1 via the SMS pathway.

Figure 5: The TMD of SDC1 drives secretion via the SMS pathway.

Figure 5:

Open in a new tab

(A) Schematic of Syndecan-1. The core protein is shown in blue, with the TMD highlighted in purple. The amino acid sequences of SDC1’s TMD and TMD_L are shown. Small side chain residues are shown in purple and a putative palmitoylation site is shown in red. (B) Example micrographs of cells expressing RUSH-pHluorin-SDC1 or RUSH-pHluorin-SDC1-TMD_L. Both proteins predominantly localize to the perinuclear ER before release and show a prominent cell surface localization 30 minutes after the addition of biotin. Scale bars represent 10 μm. (C) The TMD_L sequence does not support co-secretion with EQ-SM. TIRFM scoring the co-secretion of RUSH-pHluorin-SDC1 and RUSH-pHluorin-SDC1-TMD_L with EQ-SM-mKate2 and EQ-sol-mKate2. The means ± SD in at least 2 independent experiments are shown (264 events/11 cells for pH-SDC1+EQ-SM, 281 events/11 cells for pH-SDC1+EQ-sol, 260 events/11 cells for pH-SDC1-TMD_L+EQ-SM, 270 events/11 cells for pH-SDC1-TMD_L+EQ-sol). ** indicates p< 0.01. (D) The Ala-Leu transmembrane domain sequence confers secretion by the SMS pathway. Cells were transfected with plasmids to express paired combinations of RUSH-TMD_A8L, RUSH-TMD_L, EQ-SM, and EQ-sol. TIRFM was used to score the mean proportion of co-secretion ± SD in at least 2 independent experiments (192 events/11 cells for TMD_A8L+EQ-SM, 291 events/12 cells for TMD_A8L+EQ-sol, 192 events/10 cells for TMD_L+EQ-SM, 330 events/15 cells for TMD_L+EQ-sol, 219 events/13 cells for pH-TMD_A8L+Ruby-TMD_L, and 245 events/12 cells for Ruby-TMD_A8L + pH-TMD_L).

We next asked if the SDC1 TMD is sufficient to direct proteins into the SMS pathway. To this end, we generated a pHluorin tagged RUSH construct containing only the TMD sequence of SDC1. Unfortunately, this fusion protein could not be released from the ER, making the construct unusable for secretion experiments. Instead, we introduced a pair of artificial TMD sequences (Lorent, et al., 2017) into the RUSH system. One TMD construct consists of alternating Alanine and Leucine residues, and a Cysteine (RUSH-TMD_A8L). Importantly, the sequence of this TMD is distinct from that of the native SDC1 TMD sequence, however, it approximates the surface area of native SDC1 TMD. The second TMD construct contains a series of Leucine residues and a Cysteine (RUSH-TMD_L), which failed to support sorting into the SMS pathway when this TMD sequence was swapped into full-length SDC1. We found that secretion of RUSH-TMD_A8L, but not RUSH-TMD_L, is mediated via the SMS pathway (Fig. 5D). That is, TMD_A8L was co-secreted with EQ-SM in 82 ± 13% (n= 192, 11 cells) of exocytic events, whereas co-secretion with EQ-sol occurred in only 47 ± 11% (n= 291, 12 cells) of events. In stark contrast, TMD_L showed no preference for the SMS pathway; co-secretion with EQ-SM 67 ± 17% (n= 192, 10 cells) or with EQ-sol 64 ± 20% (n= 330, 15 cells) was approximately equivalent. To confirm the propensity of the TMD_A8L TMD sequence to drive sorting into the SMS pathway, we co-expressed RUSH-TMD_A8L and RUSH-TMD_L (with pHluorin or Ruby tags in both configurations) and scored co-secretion (Fig. 5D) by TIRF microscopy. We found that TMD_A8L and TMD_L are co-secreted in just 38% or 43% of exocytic events, respectively. This proportion of co-secretion is similar to that with EQ-sol, a marker of bulk flow secretion.

LPL sorting requires the SDC1 TMD and heparan sulfate chains

Our earlier experiments show a loss of LPL sorting when HSPG maturation is disrupted with xyloside treatment (Fig. 3B) or when heparin binding residues on LPL are mutated (Fig. 3C), suggesting that the binding of LPL to the heparan chains of SDC1 is required for sorting. Additionally, we determined that the TMD sequence of SDC1 is necessary to drive SDC1 sorting. We therefore wanted to directly test whether these two features of SDC1, its TMD and the heparan sulfate chains, are required to sort LPL into the SMS pathway. To this end, we tested the ability of several SDC1 siRNA-resistant constructs to rescue LPL sorting in SDC1 depleted cells (Fig. 6A). Each construct contains a C-terminal Flag tag, allowing us to determine the relative expression levels and steady state distribution of each SDC1 variant; importantly, this analysis shows that the levels and localization of native SDC1 and of each SDC1-variant are indistinguishable (Fig. 6B, S4). The rescue constructs include wildtype SDC1, the bulky TMD mutant SDC1-TMD_L, and a heparan sulfate-deficient SDC1 where 3 Serines (S37, S45, S47) that receive GAG chains are mutated to Alanine, yielding the construct SDC1-HS (Yang and Friedl, 2016) (Fig. 6A). As shown in Fig. 4A, knockdown of SDC1 eliminates sorting of LPL into the SMS pathway (Fig. 6C) and introduction of an siRNA-resistant wildtype SDC1 restores LPL sorting, with LPL and EQ-SM co-secretion in 70 ± 18% (n= 185, 11 cells) of exocytic events compared to LPL and EQ-sol co-secretion in 47 ± 18% (n= 133, 11 cells) of events (Fig. 6C). Neither mutant is able to rescue LPL sorting. When SDC1-TMD_L is introduced into cells, LPL is co-secreted with EQ-SM in 56 ± 16% (n= 198, 11 cells) of events and with EQ-sol in 45 ± 15% (n= 139, 11 cells) of events. When SDC1-HS is introduced into cells, LPL is co-secreted with EQ-SM in 42 ± 21% (n= 191, 12 cells) of events and with EQ-sol in 32 ± 12% (n= 180, 12 cells) of events. These results confirm that both the TMD sequence and the heparan chains of SDC1 are required for LPL sorting.

Figure 6: SDC1 TMD and heparan chains are required for LPL secretion via the SMS pathway.

Figure 6:

Open in a new tab

(A) Schematic diagram depicting wild type and mutant SDC1 proteins. Note the larger volume of the TMD in SDC1-TMD_L and the absence of the three heparan sulfate chains at the N-terminus of SDC1-HS. (B) Micrographs comparing the localization of wild type SCD1 and mutants. Cells were transfected with SDC1-Flag, SDC1-TMD_L-Flag, or SDC1-HS-Flag and antisera to Flag was used to detect each protein by immunofluorescence microscopy. Scale bars indicate 10 μm. (C) SDC1-TMD_L and SDC1-HS plasmids fail to rescue LPL sorting in cells depleted of endogenous SDC1. Cells were transfected with siRNAs targeting Syndecan-1 and with plasmid expressing si-RNA-resistant SDC1, SDC1-TMD_L, or SDC1-HS 72h before TIRFM sorting experiments. Exocytic cargo loads were evaluated and the means ± SD in at least 2 independent experiments are shown (185 events/11 cells for pH-LPL+EQ-SM+si-SDC1+SDC1, 133 events/11 cells for pH-LPL+EQ-sol+si-SDC1+SDC1, 198 events/11 cells for pH-LPL+EQ-SM+si-SDC1+SDC1-TMD_L, 139 events/11 cells for pH-LPL+EQ-sol+si-SDC1+SDC1-TMD_L, 191 events/12 cells for pH-LPL+EQ-SM+si-SDC1+SDC1-HS, and 180 events/12 cells for pH-LPL+EQ-sol+si-SDC1+SDC1-HS). Sorting data for pH-LPL+EQ-SM/EQ-sol in SDC1-depleted cells is reproduced from Figure 4A. (D) GFP-LPL co-immunopurification with SDC1 requires heparan sulfate chains. Cells were transfected with GFP-LPL and SDC1-Flag, SDC1-TMD_L-Flag, or SDC1-HS-Flag (shown in panel A) for 24h and SDC1-Flag was immunopurified as in Figure 4B. A representative Western blot shows input and IP.

Our model predicts that the SDC1 TMD confers targeting of SDC1-bound LPL into the SMS pathway. Therefore, we anticipate that two different mechanisms are responsible for the respective failures of SDC1-TMD_L or SDC1-HS to rescue LPL sorting. We predict that SDC1-TMD_L is able to capture newly synthesized LPL in the secretory pathway, but that the receptor-ligand complex fails to enter the SMS pathway. To test this, we used co-IP experiments to compare the LPL-binding capabilities of SDC1, SDC1-TMD_L, and SDC1-HS (Fig. 6D). We find that SDC1-TMD_L binds LPL with similar efficiency as wildtype SDC1. In a striking contrast, we detect almost no LPL binding by SDC1-HS. Combined, these results demonstrate that the heparan sulfate chains on SDC1 are required to capture LPL in the secretory pathway and that the TMD sequence of SDC1 is required for LPL sorting but not its capture.

Discussion

Most soluble secretory proteins are thought to be constitutively secreted via non-specific “bulk flow” (Pfeffer and Rothman, 1987). In contrast to this proposal, we find that LPL is sorted into and secreted via a distinct Golgi-to-plasma membrane route - the SMS pathway. Sorting of LPL into the SMS pathway is mediated in the Golgi apparatus by its association with Syndecan-1, which functions in this context as a sorting receptor (Fig. 7). To our knowledge, these findings establish a precedent for a receptor-mediated sorting mechanism operating for secretion of a soluble protein.

Figure 7: Model of SDC1 mediated LPL secretion.

Figure 7:

Open in a new tab

Aromatic residues (red) in the LPL PLAT domain localize LPL to the membrane, where it is captured by SDC1 via binding of LPL heparin binding residues to Syndecan-1’s heparan sulfate chains. The SDC1 TMD drives the concentration of Syndecan-1 and bound LPL in sphingomyelin-rich membrane, enriching these cargos in vesicles of the SMS pathway as they bud from the TGN.

Recognition of SDC1 by LPL requires mature heparan sulfate chains, but these are common to all HSPGs, suggesting that LPL must also interact with features of the core protein in order to achieve specificity for SDC1. Regarding sorting determinants within LPL, association with SDC1 is conferred by patches of positively charged residues on LPL’s surface, which were previously shown to bind to glycan chains (Sendak and Bensadoun, 1998; van Tilbeurgh, et al., 1994; Hata, et al., 1993), and membrane association conferred by its PLAT domain, a structural domain of poorly understood function. Remarkably, the PLAT domain of LPL and the EQ-SM reporter exhibit the same fold, a beta sandwich with aromatic residues located at equivalent positions of both domains, which mediate membrane interaction. In contrast to EQ-SM, the LPL PLAT domain does not recognize SM, explaining why this feature is insufficient to sort LPL into the SMS pathway. Hence, sorting of LPL into the SMS pathway is conferred by multi-valent interactions between LPL, the Golgi membrane, heparan sulfate, and SDC1 core protein.

What is the physiological significance of SDC1-LPL association?

LPL is highly expressed by parenchymal cells in a small subset of tissues (adipose, mammary, heart) and is widely expressed, but at low levels, in most other tissues (GTEX Portal). Obviously, the role of SDC1 as a secretory sorting receptor for LPL is restricted to those cell types that express both proteins. In adult tissues, SDC1 is chiefly expressed in epithelial cells, where it localizes to the basolateral domain. Nevertheless, SDC1 expression has been observed at low levels in heart tissue (Asundi, et al., 1997) and adipocytes (Zaragosi, et al., 2015), where it is tempting to speculate that low SDC1 expression levels could be responsible for limiting the rate of LPL secretion. Indeed, heparin treatment is known to induce a massive exocytosis of LPL from cultured adipocytes (Vannier and Ailhaud, 1989). Furthermore, there is precedent for the co-trafficking of SDC1 and LPL; during endocytosis in hepatocytes, lipoprotein remnants enriched in LPL are captured by SDC1 and internalized to prevent accumulation of triglycerides in the plasma (Stanford, et al., 2009).

After secretion by parenchymal cells, LPL is retained on the cell surface through interaction with HSPGs prevalent in the extracellular matrix. Importantly, the interactions between LPL and HSPGs are dynamic (Allan, et al., 2017), allowing LPL to navigate the interstitial space until captured on the basolateral surface of nearby capillary endothelial cells. SDC1 expressed on the basolateral surface of capillary endothelial cells would be positioned to capture interstitial LPL and may aid in LPL transcytosis by promoting its association with GPIHBP1, the GPI-anchored protein identified as the transcytosis receptor that shuttles LPL from the basolateral to the apical surface of capillary endothelial cells (Beigneux, 2010; Beigneux, et al., 2007).

Sorting of Syndecan-1 into the SMS pathway

The role of SDC1 as a receptor that mediates sorting of LPL into the SMS pathway begs the question, what confers sorting of SDC1, and integral membrane proteins in general, into the SMS pathway? We find that the transmembrane domain of SDC1, as well as an unrelated, synthetic transmembrane domain (TMD_A8L) composed of alternating Alanine and Leucine residues, is necessary and sufficient to confer sorting into the SMS pathway. The composition of the transmembrane domain of SDC1 is unusual; nearly half of its amino acids possess short, unbranched side chains (e.g., Gly, Ala), a feature that is shared with other integral membrane proteins that have a propensity to partition into liquid ordered membrane domains and exemplified by TMD_A8L (Lorent, et al., 2017). Interestingly, the distribution of short chain amino acids within the SDC1 TMD is skewed to favor the exoplasmic leaflet, where SM resides, a feature common to many proteins of the plasma membrane (Sharpe, et al., 2010). Although this could be an indication that the TMD of SDC1 recognizes SM, and there is precedent for specific binding of SM to a TMD (Contreras, et al., 2012), we have been unable to observe labeling of SDC1 by sphingolipid photoaffinity crosslinking (Haberkant, et al., 2016). When we replaced the native sequence of the SDC1 TMD with a sequence consisting of only Leucine residues, which possess large branched side chains, SDC1 was no longer sorted into the SMS pathway. Likely, the small unbranched side chains of SDC1 TMD are better accommodated by the tightly packed lipids found in membranes rich in sphingolipids and cholesterol. Interestingly, the importance of lipid environment has been observed in studies of SDC1 endocytosis, where SDC1 clustering on the cell surface precedes endocytosis, and both steps require the presence of cholesterol-rich membrane (Fuki, et al., 2000). Our results suggest that sorting of SDC1 in the TGN similarly involves concentration of SDC1 (and bound LPL) via association with SM-rich membrane at the TGN. Thus, the organizational principles of plasma membrane heterogeneity (Jacobson, et al., 2019) are not unique to this membrane; they are also harnessed to drive protein and lipid sorting in the TGN.

STAR METHODS

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to and will be fulfilled by Lead Contact, Christopher Burd (christopher.burd@yale.edu).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

HeLa cells were maintained in 5% CO2 at 37°C in DMEM supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY, United States).

METHOD DETAILS

DNA manipulations

SS-pHluorin-LPL was generated by insertion of the LPL sequence (amplified from HeLa cDNA without endogenous SS) into the vector SS-GFP-FM4-CD8α (Deng, 2016) to replace FM4-CD8α. GFP was then replaced with pHluorin, amplified from the N1-pHluorin vector (Deng, 2018). Site directed mutagenesis was used to generate LPL mutants: LPL* (mutation S132A), LPL-AR (Y387A, W390A, W393A, W394A), LPL-HB1 (K403N, R405N, K407N). LPL truncation constructs pH-lipase included LPL residues 28–340, and pH-PLAT included LPL residues 328–476. The construct SS-6His-GPIHBP1 was generated by insertion of 6His-GPIHBP1 (amplified from DNA synthesized from Integrated DNA technologies) into the SS-GFP vector described above. Oligonucleotides used for DNA sequencing and mutagenesis were synthesized by the Keck Biotechnology Resource Laboratory of Yale University. The sequences of all new constructs were confirmed by DNA sequencing.

For expression of 6His-PLAT and 6His-PLAT-AR in BL21 cells, the PLAT domain was amplified from the SS-pH-LPL or SS-pH-LPL-AR vector and inserted into an N-terminal His-tag containing pET28a-derived vector.

To make a RUSH-SDC1 construct, the vector Ii-str-HA_IRES_ST-SBP-EGFP (gift from Franck Perez) was first modified to replace the ST sequence with the signal sequence from hGH. Next, EGFP was replaced with pHluorin to generate the vector Ii-str-HA_IRES_SS-SBP-pH used in subsequent cloning. SDC1 was amplified from HeLa cDNA and inserted downstream of pHluorin. The Ii-str hook was then replaced with the str-KDEL hook amplified from the plasmid str-KDEL_ST-SBP-eGFP (gift from Julia von Blume). To make str-KDEL_SS-SBP-Ruby-SDC1, Ruby was amplified from the vector pKanCMV-mRuby3–18aa-actin (gift from Felix Rivera-Molina). To make str-KDEL_SS-SBP-pH-SDC1-TMD_L, primers were designed that spanned the TMD, mutating this sequence to 22 Leucines and a Cysteine. To make RUSH-TMD_A8L and RUSH-TMD_L, the TMD_A8L and TMD_L sequences were amplified from plasmids tr_allA8L-mRFP and tr_allL-mRFP (gift from Ilya Levental) and inserted into the Ii-str_IRES_SS-SBP-pH vector.

To make SS-SDC1-Flag, SDC1 (starting after the endogenous signal sequence) was amplified by PCR from cDNA and inserted into a vector with an N-terminal hGH signal sequence and a C-terminal Flag tag (Deng, 2016). The SDC1-TMD_L-Flag and SDC1-HS-Flag mutants were generated by site directed mutagenesis. siRNA-resistant SDC1-Flag, SDC1-TMD_L-Flag, SDC1-HS-Flag, and str-KDEL_SS-SBP-Ruby-SDC1 constructs were generated by mutating the siRNA binding site 5’ -aagauaucaccuugucacagcagac - 3’ to 5’ - aaAatCAGCccGtgCcaTTCAagac - 3’.

The plasmid pC4S1-SS-pH-FCS-hGH was a gift from Felix Rivera-Molina (Kukic, et al., 2016).

Cell culture

Hela cells were maintained in 5% CO2 at 37 °C in DM EM supplemented with 10% (vol/vol) FBS (Gibco). Cells were transfected with Fugene HD (Promega) or Lipofectamine 2000 (Thermo Fisher Scientific). All analyses of transfected cells were initiated at 16–20 h after transfection.

For siRNA experiments, HeLa cells were reverse-transfected with 30 nM nontargeting control siRNA or 10 nM each of three siRNA (combined) directed against the target using lipofectamine RNAiMax. Two days later, cells were transfected again with expression plasmids using Fugene HD and then analyzed 16–20 h later. The siRNA sequences are listed below. The effectiveness of siRNAs was determined by quantitative RT-PCR of total RNA purified by Qiagen RNAeasy reagents. RT-PCR was performed with iTaq Universal SYBR Green Supermix (Bio-Rad) and the CFX96 Touch Real-Time PCR Detection System (Bio-Rad).

Non-targeting control: CGUUAAUCGCGUAUAAUACGCGUAT

SDC1: AAGAUAUCACCUUGUCACAGCAGAC; GUAGACCUUGUUACUUCUGAGGUAA; GUACUUGUCAUUUCGGGCAAAAAAA

SDC2: GAUAAAGACAUGUACCUUGACAACA; CAUGUCUCAGAUUGACCUUACCAAG; GCAACACUUGGAACAGUGUUUACTT

SDC3: GCCAAAACCUUAAAUAAGAAAAACA; GUCAUGGUCACAUGACAGUGACAGT; AAAUCAUCUAGACACUGCAACCUCT

SDC4: GAGGUCAACCUAAUACUGACUUGTC; CAAUGAGUUCUACGCGUGAAGCUTG; AAUGGGUACUUGUGAUCACACUACG

SMS1: GACGGCAGCUUCAGCAUCAAGAUUA

SMS2: UCAAUAGUGGGACGCAGAUUCUGUU

For heparin wash experiments, cells were incubated with live cell imaging solution (Molecular Probes) supplemented with 10 mM glucose and 5 U/mL heparin sulfate sodium salt (Sigma H7640). For xyloside treatment, cells were plated and incubated with 2.5 mM 4-Nitrophenyl β-D-xylopyranoside (Sigma N2132) for 48 hours prior to imaging.

Fluorescence microscopy and image analysis

For live cell deconvolution fluorescence microscopy, cells were washed twice with PBS, and the medium was replaced with live cell imaging solution (Molecular Probes) supplemented with 10 mM glucose. 3D image stacks were collected at 0.3 μm z increments on a DeltaVision Elite workstation (Applied Precision) based on an inverted microscope (IX-70; Olympus) using a 60×, 1.4 NA oil immersion lens. Images were captured with a sCMOS camera (CoolSnap HQ; Photometrics) and deconvolved with softWoRx (v.6.0) software using the iterative-constrained algorithm and the measured point spread function.

Total internal reflection microscopy was done using a microscope (IX-70; Olympus) equipped with argon (488 nm) and argon/krypton (568 nm) laser lines, a TIRFM condenser (Olympus or custom condenser), a 60× 1.45 NA oil immersion objective lens (Olympus), and an EMCCD camera (iXon887; 0.18 μm per pixel, 16 bits; Andor Technology). The TIRFM system was controlled by iQ software (Andor Technology). HeLa cells were grown in MatTek dishes and imaged 16–20 h after transfection. All experiments were done at 37°C in live cell imaging solution (Molecular Probes) containing 10 mM glucose, pH 7.4. Cells were imaged in one channel at 5 Hz or two channels by sequential excitation at 2 Hz.

Analysis of TIRF images was done as described in Sundberg et al (Sundberg, 2018). Briefly, each image stack was manually reviewed to identify putative vesicle fusion events that released pHluorin. The coordinates of the fusion events were labeled and a small region of interest around each exocytic event in each channel was used for further analysis. Circular regions with diameters of 4 pixels were used to calculate the intensity of a single vesicle. Colocalization of proteins in the same vesicle was determined manually based on the coincident appearance and release of fluorescence signals by each fusion protein. No nonlinear adjustments were made to alter fluorescent signals.

Purification of recombinant proteins and liposome-binding experiments

PLAT, PLAT-AR, and EQ-SM proteins were expressed in Escherichia coli BL21 (DE3) cells with an N-terminal His tag from pET28a-derived vectors. The cells were grown at 37 °C to an OD600 of ~0.8–1.0, after which protein expression was induced with lactose at a final concentration of 0.2% to induce at RT. Cells were harvested ~16h after induction, then mechanically lysed in buffer consisting of 20 mM Na2HPO4/NaH2PO4, pH 7.4, 500 mM NaCl, and 25 mM imidazole supplemented with cOmplete EDTA-free protease inhibitor (Roche), using a cell disruptor (Avenstin). Cleared cell extracts were applied to a HisTrap column using an AKTA Prime liquid chromatography system (GE Healthcare), and bound PLAT, PLAT-AR, and EQ-SM were eluted with an imidazole gradient.

Liposomes used for vesicle sedimentation assays were produced from pure synthetic lipids (Avanti Polar Lipids or Echelon Biosciences) by mixing 60 mole percent (80% for liposomes not containing SM) dioleoyl-phosphatidylcholine, 20 mole percent SM (porcine brain), and 20 mole percent cholesterol in buffer (10 mM Hepes and 100 mM NaCl, pH 6.5) by extrusion through 1-μm-pore filters using a mini-extruder (Avanti Polar Lipids). For binding assays, liposomes (2 mM lipid) were incubated with purified proteins (10 μM) for 30 min at 37 °C, after which liposomes were collected by centrifugation at 100,000 × g for 10 min at 37 °C. Supernatants were collected, and pellets were suspended in an equal volume of sample buffer. The protein band densities were determined using Image Lab software (Bio-Rad).

Co-Immunoprecipitation experiments

HeLa cells cultured in 6-well plates were transfected with a SDC1-Flag construct (WT, TMD_L, or HS) and GFP-LPL or with GFP-LPL alone. After 24h, cells were washed with 1x dPBS and harvested by scraping in 1mL dPBS. Cells were collected by centrifugation at 1,500 × g and lysed in buffer containing 50mM Tris pH 8.0, 150mM NaCl, 0.5% Triton X-100, and 250 U/mL benzonase (Sigma, E1014). Cell lysate was clarified by centrifugation at 14,000 × g. Protein concentrations were measured by Bradford and volumes were adjusted for equal protein concentrations. 200 μl cell lysate was added to 10 μl Flag beads (ThermoFisher, 36797) and incubated for 1h at 37 °C. Beads were washed 3x in lysis buffer and protein was eluted by boiling in SDS page protein loading buffer. 5% input and 50% IP are loaded in gel for Western Blot. Proteins were detected with anti-Flag (1:5000; Sigma, 1804) and anti-GFP (1:1000; Roche, 11814460001) antibodies.

Protein detection by immunoblotting or immunofluorescence

For immunoblotting, proteins were transferred to nitrocellulose membrane and then blocked in PBS-T with 5% milk. After immunoblotting, proteins were detected by chemiluminescence. Images of immunoblots were acquired by ChemiDoc Imaging System (Bio-Rad).

For immunostaining, cells were cultured in glass-bottom MatTeK dishes, fixed for 10 min with 4% paraformaldehyde, washed with PBS and subsequently permeabilized for 5 min in 0.2% Triton-X 100 and 0.5% SDS in 4% BSA solution. After washing with PBS and blocking of slides for 1 h in 4% BSA, cells were incubated with primary and secondary antibody for 1 h at room temperature in blocking buffer in the dark. SDC1-Flag constructs were detected with anti-Flag (1:500; Sigma, 1804), and Alexa Fluor 405 anti-mouse (1:1000) secondary antibody.

QUANTIFICATION AND STATISTICAL ANALYSIS

For statistical evaluation GraphPad Prism version 7.01 for PC (GraphPad Software, La Jolla California USA) was used. TIRF exocytosis assays were analyzed using Student’s unpaired t-test. The following P-value style was used: ≤ 0.05 (*), ≤ 0.01 (**), ≤ 0.001 (***), ≤ 0.0001 (****).

DATA AND CODE AVAILABILITY

This study did not generate datasets/code.

Supplementary Material

Supplemental Figures
Table S1

Table S1, related to STAR Methods. List of siRNAs and oligonucleotides used for knockdowns and PCR-based determination of knockdown levels.

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
Mouse monoclonal anti-FLAG Sigma-Aldrich Cat# F1804, RRID: AB_262044
Mouse monoclonal anti-β-actin Sigma-Aldrich Cat# A5441, RRID: AB_476744
Mouse monoclonal anti-GPF Sigma-Aldrich Cat# 11814460001, RRID: AB_390913
Goat anti-Mouse IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 647 Thermo Fisher Scientific Cat# A28181, RRID: AB_2536165
Goat anti-Mouse IgG, HRP-linked Antibody Cell Signaling Cat# 7076, RRID: AB_330924
Bacterial and Virus Strains
Biological Samples
Chemicals, Peptides, and Recombinant Proteins
Heparin sodium salt from porcine intestinal mucosa Sigma-Aldrich Cat# H3393
4-Nitrophenyl β-D-xylopyranoside Sigma-Aldrich Cat# N2132
Critical Commercial Assays
Deposited Data
Experimental Models: Cell Lines
Human: T-REx™-HeLa Cell Line Invitrogen R71407; RRID:CVCL_D587
Experimental Models: Organisms/Strains
Oligonucleotides
siRNAs- see Table S1
Primers for qPCR- see Table S1
Recombinant DNA
N1-EQ-SM-mKate2 Deng et al., 2016 N/A
N1-EQ-sol-mKate2 Deng et al., 2016 N/A
C4S1- SS-pHluorin-LPL This paper N/A
C4S1-SS-pHluorin-LPL-S132A This paper N/A
C4S1-SS-pHluorin-LPL-AR This paper N/A
C4S1-SS-pHluorin-LPL-HB1 This paper N/A
C4S1-SS-pHluorin-lipase This paper N/A
C4S1-SS-pHluorin-PLAT This paper N/A
C4S1-SS-6His-GPIHBP1 This paper N/A
pET28a-6His-PLAT This paper N/A
pET28a-6His-PLAT-AR This paper N/A
Ii-str _IRES_SS-SBP-pH-TMD_L This paper N/A
Ii-str _IRES_SS-SBP-pH-TMD_A8L This paper N/A
Str-KDEL_IRES_SS-SBP-pH-SDC1 This paper N/A
Str-KDEL_IRES_SS-SBP-pH-SDC1-TMD_L This paper N/A
Str-KDEL_IRES_SS-SBP-Ruby-SDC1 This paper N/A
N1-SS-SDC1-Flag This paper N/A
N1-SS-SDC1-TMD_L-Flag This paper N/A
N1-SS-SDC1-HS-Flag This paper N/A
pC4S1-SS-pH-FCS-hGH Kukic et al, 2016 N/A
Software and Algorithms
ImageJ NIH https://imagej.nih.gov/ij/
Prism Graphpad https://www.graphpad.com/scientific-software/prism/
Other
Pierce™ Anti-DYKDDDDK Magnetic Agarose Beads Thermo Scientific Cat# A36797
Open in a new tab

Highlights.

  • Syndecan-1 acts as a sorting receptor for lipoprotein lipase at the TGN

  • Syndecan-1 heparan sulfate chains and core protein bind lipoprotein lipase

  • Properties of the Syndecan-1 TMD target it to sphingomyelin-rich membrane

  • Syndecan-1 and LPL are co-secreted in secretory vesicles enriched in sphingomyelin

Acknowledgements

We are grateful to Ilya Levental for reagents and advice regarding TMD experiments. We thank Andreas Ernst, Julia von Blume, and Xiaolei Su for discussions. The data referenced as sourced from GTEX portal in this manuscript were obtained from the Genotype-Tissue Expression, GTEx Portal (www.gtexportal.org) on 05/01/2019. Research reported in this publication was supported by the National Institute of General Medical Sciences of the National Institutes of Health under awards, R01GM095766 and R01GM098498. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors declare no competing financial interests.

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.

Declaration of Interests

The authors declare no competing interests.

References

  1. Allan CM, Larsson M, Jung RS, Ploug M, Bensadoun A, Beigneux AP, Fong LG, and Young SG (2017). Mobility of “HSPG-bound” LPL explains how LPL is able to reach GPIHBP1 on capillaries. J Lipid Res 58, 216–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Asundi VK, Keister BF, Stahl RC, and Carey DJ (1997). Developmental and cell-type-specific expression of cell surface heparan sulfate proteoglycans in the rat heart. Exp Cell Res 230, 145–53. [DOI] [PubMed] [Google Scholar]
  3. Athanasiadis A, Anderluh G, Macek P, and Turk D (2001). Crystal structure of the soluble form of equinatoxin II, a pore-forming toxin from the sea anemone Actinia equina. Structure 9, 341–6. [DOI] [PubMed] [Google Scholar]
  4. Banfield DK (2011). Mechanisms of protein retention in the Golgi. Cold Spring Harb Perspect Biol 3, a005264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barman S, and Nayak DP (2000). Analysis of the transmembrane domain of influenza virus neuraminidase, a type II transmembrane glycoprotein, for apical sorting and raft association. J Virol 74, 6538–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beigneux AP (2010). GPIHBP1 and the processing of triglyceride-rich lipoproteins. Clin Lipidol 5, 575–582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Beigneux AP, Davies BS, Gin P, Weinstein MM, Farber E, Qiao X, Peale F, Bunting S, Walzem RL, Wong JS, et al. (2007). Glycosylphosphatidylinositol-anchored high-density lipoprotein-binding protein 1 plays a critical role in the lipolytic processing of chylomicrons. Cell Metab 5, 279–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Birrane G, Beigneux AP, Dwyer B, Strack-Logue B, Kristensen KK, Francone OL, Fong LG, Mertens HDT, Pan CQ, Ploug M, et al. (2019). Structure of the lipoprotein lipase-GPIHBP1 complex that mediates plasma triglyceride hydrolysis. Proc Natl Acad Sci U S A 116, 1723–1732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Boncompain G, Divoux S, Gareil N, de Forges H, Lescure A, Latreche L, Mercanti V, Jollivet F, Raposo G, and Perez F (2012). Synchronization of secretory protein traffic in populations of cells. Nat Methods 9, 493–8. [DOI] [PubMed] [Google Scholar]
  10. Boncompain G, and Weigel AV (2018). Transport and sorting in the Golgi complex: multiple mechanisms sort diverse cargo. Curr Opin Cell Biol 50, 94–101. [DOI] [PubMed] [Google Scholar]
  11. Bretscher MS, and Munro S (1993). Cholesterol and the Golgi apparatus. Science 261, 1280–1. [DOI] [PubMed] [Google Scholar]
  12. Brown DA, and Rose JK (1992). Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cell surface. Cell 68, 533–44. [DOI] [PubMed] [Google Scholar]
  13. Chanat E, and Huttner WB (1991). Milieu-induced, selective aggregation of regulated secretory proteins in the trans-Golgi network. J Cell Biol 115, 1505–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Contreras FX, Ernst AM, Haberkant P, Bjorkholm P, Lindahl E, Gonen B, Tischer C, Elofsson A, von Heijne G, Thiele C, et al. (2012). Molecular recognition of a single sphingolipid species by a protein’s transmembrane domain. Nature 481, 525–9. [DOI] [PubMed] [Google Scholar]
  15. Davies BS, Beigneux AP, Barnes RH 2nd, Tu Y, Gin P, Weinstein MM, Nobumori C, Nyren R, Goldberg I, Olivecrona G, et al. (2010). GPIHBP1 is responsible for the entry of lipoprotein lipase into capillaries. Cell Metab 12, 42–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Davies BS, Beigneux AP, Fong LG, and Young SG (2012a). New wrinkles in lipoprotein lipase biology. Curr Opin Lipidol 23, 35–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Davies BS, Goulbourne CN, Barnes RH 2nd, Turlo KA, Gin P, Vaughan S, Vaux DJ, Bensadoun A, Beigneux AP, Fong LG, et al. (2012b). Assessing mechanisms of GPIHBP1 and lipoprotein lipase movement across endothelial cells. J Lipid Res 53, 2690–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. De Matteis MA, and Luini A (2008). Exiting the Golgi complex. Nat Rev Mol Cell Biol 9, 273–84. [DOI] [PubMed] [Google Scholar]
  19. Deng Y, Pakdel M, Blank B, Sundberg EL, Burd CG, and von Blume J (2018). Activity of the SPCA1 Calcium Pump Couples Sphingomyelin Synthesis to Sorting of Secretory Proteins in the Trans-Golgi Network. Dev Cell 47, 464–478 e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Deng Y, Rivera-Molina FE, Toomre DK, and Burd CG (2016). Sphingomyelin is sorted at the trans Golgi network into a distinct class of secretory vesicle. Proc Natl Acad Sci U S A 113, 6677–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Dukhovny A, Yaffe Y, Shepshelovitch J, and Hirschberg K (2009). The length of cargo-protein transmembrane segments drives secretory transport by facilitating cargo concentration in export domains. J Cell Sci 122, 1759–67. [DOI] [PubMed] [Google Scholar]
  22. Fuki IV, Meyer ME, and Williams KJ (2000). Transmembrane and cytoplasmic domains of syndecan mediate a multi-step endocytic pathway involving detergent-insoluble membrane rafts. Biochem J 351 Pt 3, 607–12. [PMC free article] [PubMed] [Google Scholar]
  23. Haberkant P, Stein F, Hoglinger D, Gerl MJ, Brugger B, Van Veldhoven PP, Krijgsveld J, Gavin AC, and Schultz C (2016). Bifunctional Sphingosine for Cell-Based Analysis of Protein-Sphingolipid Interactions. ACS Chem Biol 11, 222–30. [DOI] [PubMed] [Google Scholar]
  24. Hata A, Ridinger DN, Sutherland S, Emi M, Shuhua Z, Myers RL, Ren K, Cheng T, Inoue I, Wilson DE, et al. (1993). Binding of lipoprotein lipase to heparin. Identification of five critical residues in two distinct segments of the amino-terminal domain. J Biol Chem 268, 8447–57. [PubMed] [Google Scholar]
  25. He PP, Jiang T, OuYang XP, Liang YQ, Zou JQ, Wang Y, Shen QQ, Liao L, and Zheng XL (2018). Lipoprotein lipase: Biosynthesis, regulatory factors, and its role in atherosclerosis and other diseases. Clin Chim Acta 480, 126–137. [DOI] [PubMed] [Google Scholar]
  26. Hille-Rehfeld A (1995). Mannose 6-phosphate receptors in sorting and transport of lysosomal enzymes. Biochim Biophys Acta 1241, 177–94. [DOI] [PubMed] [Google Scholar]
  27. Jacobson K, Liu P, and Lagerholm BC (2019). The Lateral Organization and Mobility of Plasma Membrane Components. Cell 177, 806–819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kienzle C, and von Blume J (2014). Secretory cargo sorting at the trans-Golgi network. Trends Cell Biol 24, 584–93. [DOI] [PubMed] [Google Scholar]
  29. Klemm RW, Ejsing CS, Surma MA, Kaiser HJ, Gerl MJ, Sampaio JL, de Robillard Q, Ferguson C, Proszynski TJ, Shevchenko A, et al. (2009). Segregation of sphingolipids and sterols during formation of secretory vesicles at the trans-Golgi network. J Cell Biol 185, 601–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kukic I, Rivera-Molina F, and Toomre D (2016). The IN/OUT assay: a new tool to study ciliogenesis. Cilia 5, 23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Levental I, Lingwood D, Grzybek M, Coskun U, and Simons K (2010). Palmitoylation regulates raft affinity for the majority of integral raft proteins. Proc Natl Acad Sci U S A 107, 22050–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lin S, Naim HY, Rodriguez AC, and Roth MG (1998). Mutations in the middle of the transmembrane domain reverse the polarity of transport of the influenza virus hemagglutinin in MDCK epithelial cells. J Cell Biol 142, 51–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lorent JH, Diaz-Rohrer B, Lin X, Spring K, Gorfe AA, Levental KR, and Levental I (2017). Structural determinants and functional consequences of protein affinity for membrane rafts. Nat Commun 8, 1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Manes S, del Real G, and Martinez AC (2003). Pathogens: raft hijackers. Nat Rev Immunol 3, 557–68. [DOI] [PubMed] [Google Scholar]
  35. Mead JR, Irvine SA, and Ramji DP (2002). Lipoprotein lipase: structure, function, regulation, and role in disease. J Mol Med (Berl) 80, 753–69. [DOI] [PubMed] [Google Scholar]
  36. Nagarajan A, Malvi P, and Wajapeyee N (2018). Heparan Sulfate and Heparan Sulfate Proteoglycans in Cancer Initiation and Progression. Front Endocrinol (Lausanne) 9, 483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Pfeffer SR, and Rothman JE (1987). Biosynthetic protein transport and sorting by the endoplasmic reticulum and Golgi. Annu Rev Biochem 56, 829–52. [DOI] [PubMed] [Google Scholar]
  38. Sarrazin S, Lamanna WC, and Esko JD (2011). Heparan sulfate proteoglycans. Cold Spring Harb Perspect Biol 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sendak RA, and Bensadoun A (1998). Identification of a heparin-binding domain in the distal carboxyl-terminal region of lipoprotein lipase by site-directed mutagenesis. J Lipid Res 39, 1310–5. [PubMed] [Google Scholar]
  40. Sezgin E, Levental I, Mayor S, and Eggeling C (2017). The mystery of membrane organization: composition, regulation and roles of lipid rafts. Nat Rev Mol Cell Biol 18, 361–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sharpe HJ, Stevens TJ, and Munro S (2010). A comprehensive comparison of transmembrane domains reveals organelle-specific properties. Cell 142, 158–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Simons K, and Ikonen E (1997). Functional rafts in cell membranes. Nature 387, 569–72. [DOI] [PubMed] [Google Scholar]
  43. Simons K, and Toomre D (2000). Lipid rafts and signal transduction. Nat Rev Mol Cell Biol 1, 31–9. [DOI] [PubMed] [Google Scholar]
  44. Stanford KI, Bishop JR, Foley EM, Gonzales JC, Niesman IR, Witztum JL, and Esko JD (2009). Syndecan-1 is the primary heparan sulfate proteoglycan mediating hepatic clearance of triglyceride-rich lipoproteins in mice. J Clin Invest 119, 3236–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sundberg EL, Deng Y, and Burd CG (2019). Monitoring Sphingolipid Trafficking in Cells using Fluorescence Microscopy. Curr Protoc Cell Biol 82, e67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Surma MA, Klose C, Klemm RW, Ejsing CS, and Simons K (2011). Generic sorting of raft lipids into secretory vesicles in yeast. Traffic 12, 1139–47. [DOI] [PubMed] [Google Scholar]
  47. Surma MA, Klose C, and Simons K (2012). Lipid-dependent protein sorting at the trans-Golgi network. Biochim Biophys Acta 1821, 1059–67. [DOI] [PubMed] [Google Scholar]
  48. van Meer G, Voelker DR, and Feigenson GW (2008). Membrane lipids: where they are and how they behave. Nat Rev Mol Cell Biol 9, 112–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. van Tilbeurgh H, Roussel A, Lalouel JM, and Cambillau C (1994). Lipoprotein lipase. Molecular model based on the pancreatic lipase x-ray structure: consequences for heparin binding and catalysis. J Biol Chem 269, 4626–33. [PubMed] [Google Scholar]
  50. Vannier C, and Ailhaud G (1989). Biosynthesis of lipoprotein lipase in cultured mouse adipocytes. II. Processing, subunit assembly, and intracellular transport. J Biol Chem 264, 13206–16. [PubMed] [Google Scholar]
  51. Wong H, Davis RC, Thuren T, Goers JW, Nikazy J, Waite M, and Schotz MC (1994). Lipoprotein lipase domain function. J Biol Chem 269, 10319–23. [PubMed] [Google Scholar]
  52. Yang N, and Friedl A (2016). Syndecan-1-Induced ECM Fiber Alignment Requires Integrin alphavbeta3 and Syndecan-1 Ectodomain and Heparan Sulfate Chains. PLoS One 11, e0150132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Zaragosi LE, Dadone B, Michiels JF, Marty M, Pedeutour F, Dani C, and Bianchini L (2015). Syndecan-1 regulates adipogenesis: new insights in dedifferentiated liposarcoma tumorigenesis. Carcinogenesis 36, 32–40. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Figures
NIHMS1538875-supplement-Supplemental_Figures.pdf (772.2KB, pdf)
Table S1

Table S1, related to STAR Methods. List of siRNAs and oligonucleotides used for knockdowns and PCR-based determination of knockdown levels.

NIHMS1538875-supplement-Table_S1.xlsx (11.9KB, xlsx)

Data Availability Statement

This study did not generate datasets/code.

RESOURCES