Sphingolipids as modulators of membrane proteins

Sphingolipids as modulators of membrane proteins

BBAMCB-57539; No. of pages: 6; 4C: 2 Biochimica et Biophysica Acta xxx (2013) xxx–xxx Contents lists available at ScienceDirect Biochimica et Biophy...

648KB Sizes 0 Downloads 2 Views

BBAMCB-57539; No. of pages: 6; 4C: 2 Biochimica et Biophysica Acta xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbalip

Review

Sphingolipids as modulators of membrane proteins☆ Andreas Max Ernst ⁎, Britta Brügger ⁎ Heidelberg University Biochemistry Center, Im Neuenheimer Feld 328, 69120 Heidelberg, Germany

a r t i c l e

i n f o

Article history: Received 25 July 2013 Received in revised form 22 September 2013 Accepted 28 October 2013 Available online xxxx Keywords: Sphingolipid Protein–lipid interaction Receptor activity Membrane domain Molecular lipid species Raft

a b s t r a c t The diversity of the transmembranome of higher eukaryotes is matched by an enormous diversity of sphingolipid classes and molecular species. The intrinsic properties of sphingolipids are not only suited for orchestrating lateral architectures of biological membranes, but their molecular distinctions also allow for the evolution of protein motifs specifically recognising and interacting with individual lipids. Although various reports suggest a role of sphingolipids in membrane protein function, only a few cases have determined the specificity of these interactions. In this review we discuss examples of specific protein–sphingolipid interactions for which a modulator-like dependency on sphingolipids was assigned to specific proteins. These novel functions of sphingolipids in specific protein–lipid assemblies contribute to the complexity of the sphingolipid classes and other molecular species observed in animal cells. This article is part of a Special Issue entitled New Frontiers in Sphingolipid Biology. © 2013 Published by Elsevier B.V.

1. Introduction 1.1. Co-evolution of transmembrane domain (TMD) and membrane lipid diversity An apparent difference in the composition of prokaryotic and eukaryotic membranes is the diversity and complexity of the lipid classes and molecular species observed, particularly in higher eukaryotes. The extent of TMD diversity also matches that of membrane lipids and correlates with the increased presence of sterols as membrane constituents. Comparisons of lipid packing in plasma membranes from prokaryotic and eukaryotic sources have shown a convergence of the respective surface orders, although their lipid compositions differ considerably [1]. While the relative membrane order of eukaryotes is influenced by sterol- and sphingolipid-based interactions, the membrane order in prokaryotes, which generally lack sterols, seems to be achieved by the action of transmembrane protein–lipid interactions. These findings suggest that the use of sterols as membrane constituents relieved evolutionary pressure on the transmembranome of eukaryotes and allowed diversification of both lipids and transmembrane domains. Likewise, sphingolipids have also diversified, and a steadily increasing number of different functions are associated with these lipids, including signalling and the formation of specialised nano-domains [2]. In the past decade, a novel field of research has emerged that focuses on specific interactions between distinct sphingolipid classes and molecular species with membrane proteins. It is now increasingly appreciated that certain ☆ This article is part of a Special Issue entitled New Frontiers in Sphingolipid Biology ⁎ Corresponding authors. Tel.: +49 6221 54 5426. E-mail addresses: [email protected] (A.M. Ernst), [email protected] (B. Brügger).

sphingolipids can act as effectors of membrane proteins by allosterically regulating protein function. Although several reports have suggested the involvement of sphingolipids in membrane protein function using indirect approaches, this review focuses on studies providing evidence of direct and specific protein–sphingolipid-interactions. 1.2. A role of sphingolipids in generating lateral heterogeneities in biological membranes Sphingolipids not only participate in various signalling pathways, but also play roles as membrane constituents, whereby their molecular diversity influences the physicochemical nature of biological membranes [3]. The biophysical aspects of sphingolipids are significantly different from glycerol-based membrane lipids in that both the 3′-hydroxyl group and the amide nitrogen represent hydrogen-bond donor and acceptor functions, which results in the propensity of these lipids to form networks with other sphingolipids [4]. Intramolecular hydrogen bonds and the predominantly saturated N-acylated fatty acid of sphingolipids result in increased acyl chain packing and a reduction in cis-gauche rotational freedom. This leads to a significant reduction in the molecular area per sphingolipid compared to glycerolipids with equally saturated acyl chains [5]. Thus, sphingolipids can be considered as rigid, networkforming molecules. For glycosphingolipids, the lipid–lipid interactions can extend towards their carbohydrate moieties (for a review see [6]), which may be why glycosphingolipid-enriched microdomains (GEMs) appear to assemble independently of cholesterol in vitro and result in detergent-insoluble platforms in cells [7]. An additional level of complexity appears to exist for sphingolipid– lipid interactions, as sphingolipid networks can form in liquid disordered phases even at concentrations below 1 mol% [8]. Furthermore, different sphingolipid molecular species differed in their propensity to

1388-1981/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.bbalip.2013.10.016

Please cite this article as: A.M. Ernst, B. Brügger, Sphingolipids as modulators of membrane proteins, Biochim. Biophys. Acta (2013), http:// dx.doi.org/10.1016/j.bbalip.2013.10.016

2

A.M. Ernst, B. Brügger / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

form such networks or subdomains in a concentration-dependent manner, suggesting the existence of critical domain concentrations for sphingolipids as postulated by Holthuis et al. [4]. Subdomain formation is also critically modified by the presence of cholesterol. Sphingolipids, and in particular sphingomyelin (SM), have a tendency to interact with cholesterol due to acyl chain packing, hydrogen bonding, and the so-called umbrella effect [9–14]. As a result, cholesterol strengthens contacts within sphingolipid subdomains in the plane of the lipid bilayer and increases fluidity in areas where sphingolipids are tightly packed, thereby inhibiting the transition to gel phases [15]. 2. Interactions of membrane proteins with sphingolipids 2.1. Sphingolipid-enriched membrane domains In addition to sphingolipid–cholesterol based interactions that lead to the formation of membrane subdomains (for recent reviews see [16–18]), other lipid classes and molecular species undergo specific interactions as well (for examples see [19–21]). Biological membranes are comprised of smaller and highly dynamic subdomains, and the lipid diversity is thought to help establish their complex structure [22]. Lipids contribute to the molecular identity of organelles, and their composition reflects the different requirements needed in different locations throughout eukaryotic cells. Due to the presence of highly dynamic subdomains, for each membrane protein, different types of specificity for distinct nano-environments most likely have to be distinguished. Sphingolipid-based subdomains are characterised by an increase in the membrane thickness as well as high lateral pressure. In addition, sphingolipid-enriched membrane areas provide a distinct landscape of polar sphingolipid head groups, particularly because of the complex oligosaccharide chains of higher glycosphingolipids [5,23]. Therefore, membrane proteins are likely attracted to these sphingolipid-enriched subdomains by different means due to their physical compatibility. For example, the TMD of a membrane protein may be compatible in hydrophobic length to the increased diameter of sphingolipid–cholesterol based subdomains, which has been observed for plasma membrane resident proteins [24].

and membrane proteins [29]. For soluble proteins containing SBDs, sphingolipids function as receptors and mediate the entry of toxins and pathogens into the cell or act as scaffolds to organise protein– protein interactions relevant for membrane trafficking and other processes. SBDs are capable of distinguishing between sphingolipids and glycerolipids; however, it is not clear whether they select specific molecular species of sphingolipids. More recently this ability was discovered in membrane proteins through the identification of a sphingolipid-binding motif in the TMD of p24 [30]. This molecular species-determining (MSD) motif was not only found to be sufficient to bind SM, but also was selective for a single molecular SM species. The specific interaction in this case was shown to allosterically modulate the oligomeric state of the protein, thereby altering its function in COPI vesicle biogenesis. In summary, specific interactions with individual sphingolipid species may confer affinity of proteins for distinct types of sphingolipid subdomains. In particular, the selectivity mediated by MSD-motifs of membrane proteins towards distinct molecular sphingolipid species implies that membrane proteins have evolved to distinguish between different classes and molecular species of sphingolipids.

2.2. Sphingolipid-based sorting of membrane proteins in the secretory pathway Plasma membrane proteins exhibit the highest predicted average lengths of TMDs, and various thicknesses and lipid environments are encountered when these proteins are transported through the secretory pathway. The gradient of sphingolipids across organelles of the secretory pathway (i.e., minimal content in the ER and maximal content in the plasma membrane) [25] is thought to be responsible for sorting of plasma membrane proteins [24]. Interactions with sphingolipid-based subdomains would stabilise and sequentially distill membrane proteins containing long TMDs towards compartments that are enriched with sphingolipids. In the trans-Golgi network, lipid domains are thought to function as sorting platforms for proteins [26]. Other examples include glycosylphosphatidylinositol (GPI)-anchored proteins, which contain predominantly saturated acyl chains and are more compatible with the rigid environment of sphingolipid-based subdomains [27]. In liquid disordered phases, lipids are characterised by a reduction in acyl chain packing (due to a higher degree of unsaturation) and are predominantly glycerol-based [28]. In this phase, sphingolipids may act as ‘transport chaperones’ in that they generate a unique physical nano-environment around ‘protein clients’. In contrast to interactions that are based on the physicochemical nature of sphingolipid-enriched subdomains, the unique chemical composition of sphingolipids allows for the evolution of recognition motifs on proteins that mediate stoichiometric and specific interactions with them, including sphingolipid-binding domains (SBDs) on soluble

Fig. 1. A 2 + 2 model of p24(TMD) and SM18. The transmembrane domains are depicted in blue, and residues V13, T16, Leu 17, and Tyr 21 are depicted in red. SM 18 molecules are depicted in grey. Contacts between the TMD helices are mediated by polar residues. In this dimer conformation, the duration of the helix–helix interactions match the duration of the protein–lipid interaction (N200 ns; [30]). Top: side view on the model; bottom: top view. Kindly provided by Erik Lindahl.

Please cite this article as: A.M. Ernst, B. Brügger, Sphingolipids as modulators of membrane proteins, Biochim. Biophys. Acta (2013), http:// dx.doi.org/10.1016/j.bbalip.2013.10.016

A.M. Ernst, B. Brügger / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

2.3. Sphingolipid binding domains (SBDs) — piercing underneath the head group of sphingolipids Numerous soluble proteins have evolved to recognise sphingolipids from the solvent-accessible head group region of the plasma membrane. For some sphingolipid-binding proteins, such as actinoporins, SM is selectively recognised and acts as a receptor for the poreforming toxin [31]. The role of SM in toxin activity was first described by Bernheimer and Avigad who showed that sphingomyelinasetreated erythrocytes were resistant to toxin-mediated cytolysis [32]. In a subsequent study, Bakrač et al. found that SM, but not the identical head group-carrying lipid phosphatidylcholine (PC), was specifically bound by the toxin [33]. Mutagenesis studies also identified two aromatic residues (Trp112 and Tyr113) that are critical determinants for the interaction of the toxin with SM, and structural data suggested that recognition of SM takes place below the phosphocholine moiety. Hence, these toxins are thought to pierce underneath the head group of the respective lipid and undergo hydrogen-bonding and/or amide-π interactions with the molecular distinctions present in the polar to apolar interface sphingolipids (the amide bond and the hydroxylgroup; reviewed by [34]). Alternatively, the recently elucidated structure of the SM-specific toxin lysenin suggests that two critical tyrosine residues, positioned outside the phosphocholine binding pocket, may undergo stacking-like interactions with the acyl chain present in SM [35]. The specificity of sphingolipid recognition has been observed in other soluble proteins as well, including a hairpin-like structure in the V3-loop of HIV-1 gp120 [29], which mediates binding to glycosphingolipids and SM via key aromatic (phenylalanine, tyrosine) and arginine residues [36]. For gp120, sphingolipids act as cofactors in promoting the lateral assembly of the fusion complex between HIV-1 and the plasma membrane. V3-like folds were also detected in the Alzheimer ß-amyloid peptide (Aß) and shown to specifically bind to glycosphingolipids [37–40]. In addition, a V3-like fold was identified on prion protein (PrP) that specifically interacts with galactosylceramide and SM [41]. Based on the structural similarities of the V3-loop of gp120, Aß, and PrP, a common SBD was delineated [29]. For Aß and PrP, sphingolipids are discussed as cofactors in structurally orienting the proteins with respect to the membrane. Interestingly, a mutation found in PrP correlates with a common familial form of Creutzfeld–Jakob disease and affects a residue that was shown to be involved in the interaction with SM [41]. Another example of a protein domain within a soluble protein recognising a sphingolipid is the C2 domain of cPLA2 alpha, which was shown to specifically bind to ceramide-1-phosphate [42]. Taken together, SBDs present on soluble proteins appear to use sphingolipids as receptors, scaffolding elements, building blocks in oligomeric assemblies, and even as chaperones, such as that described for the bile salt-dependent lipase [43].

3

energy transfer (FRET)-based assay was established. Different molecular species of fluorescent pentaene-SM with acyl chains ranging from C14–24 were probed with the TMD of p24. Again, SM C18 was the only molecule to show significant FRET. An alanine mutant screen of the TMD led to the identification of four residues that critically influenced the specificity of interactions with SM 18 (TMD positions Val13, Thr16, Leu17, and Tyr21). This assembly of residues was termed the “molecular species-determining” (MSD) motif. Further experiments suggested that binding of p24 to SM 18 shifts the oligomeric state of p24 towards the active dimer. A variant of p24 compromised in sphingomyelin binding, p24(TMD)L17F, was found to generate fewer p24-oligomers both in vitro and in vivo. Interestingly, this variant exhibited reduced Golgiexit kinetics, presumably because of compromised COPI-mediated retrograde trafficking. Taken together, these data show that a single molecular SM species, SM 18:0, is bound by and allosterically enhances oligomer formation of p24 (Fig. 1), which in turns affects the efficiency of COPI vesicle-mediated retrograde trafficking. Considering the gradient of sphingolipids across the Golgi, it could be envisaged that the rate of retrograde transport increases towards the trans-Golgi network, which is a mechanism that could putatively contribute to the quality control of protein modifications in the Golgi. Bioinformatic analyses have identified a number of other membrane proteins that contained similar motifs in their TMDs to date. An increasing number of these candidates were identified as sphingolipid-binding proteins; however, the respective interacting molecular sphingolipid species still need to be resolved. In general, binding of distinct molecular sphingolipid species via MSD motifs to membrane proteins may, as observed for p24, modulate the oligomeric state of membrane proteins. Alternatively, binding of sphingolipids may guide multi-subunitcontaining membrane protein complexes to distinct sphingolipidcontaining nanodomains. Since most of these candidates are plasma membrane residents, specific binding of sphingolipids may also facilitate efficient transport across the secretory pathway as well as modulate their function at the plasma membrane. 3.2. Regulation of receptor tyrosine kinases (RTKs) by sphingolipids RTKs reside at the plasma membrane and are involved in essential processes, such as metabolism, proliferation, and cell-cycle control [46]. RTKs are type I transmembrane proteins that contain ligand-binding domains in their extracellular moieties as well as a transmembrane domain and tyrosine kinase domain in the cytoplasmic face of the receptor. RTKs can either dimerise upon binding of their respective (bivalent) ligand or can be present as covalently linked dimers (e.g., the insulin receptor, [47]). Interaction with their respective ligand leads to stimulation of the tyrosine kinase domains, autophosphorylation of the RTKs, and subsequent downstream signalling cascades [48]. In addition, several RTKs appear to be regulated by sphingolipids, and in particular by sialylated glycosphingolipids.

3. Sphingolipids as regulators of membrane protein function 3.1. A single molecular species of SM allosterically regulates p24 COPI- and COPII-dependent transport within the early secretory pathway involves p24 proteins, which exhibit type I topology and exist as homo- and hetero-oligomers [44]. Mass spectrometric analysis of the membrane lipids of COPI vesicles revealed a significant reduction of SM and cholesterol compared to donor Golgi membranes, except for a single molecular species of SM (SM 18:0), which was found to be significantly enriched in COPI vesicle membranes [45]. In order to determine whether sorting of this single lipid species is based on a specific protein–lipid interaction, membrane proteins enriched in COPI vesicles were probed for interactions with tritiated, photoactivatable sphingolipids [30]. Of the proteins tested, only the p24 protein family member p24 bound significant amounts of photo-SM. To investigate the selectivity of binding to individual SM species, a novel proteoliposomal Förster resonance

3.2.1. Functions of glycosphingolipids in regulating receptors for plateletderived growth factor (PDGF) and epidermal growth factor (EGF) The involvement of glycosphingolipids as allosteric inhibitors of RTKs was initially hypothesised due to the observed drastic reduction of GM3 and GM1 together with the loss of growth control during oncogenic transformation of animal cells (initially described in [49]). For example, inhibition of the PDGF and a concomitant reduction in tyrosine phosphorylation of the PDGF receptor was observed upon exogenous addition of GM1 and GM3 [50]. The authors of that study extended this approach and probed the inhibition of cells in the presence of GM3, GM1, and EGF [51]. Again, EGF-stimulated phosphorylation of the EGF receptor was significantly reduced by the addition of GM3, and to a much lower extent by GM1. In a more recent study, EGF receptor (EGFR) dimerisation and autophosphorylation was characterised in vitro to gain insight into the molecular mechanisms of GM3-mediated negative regulation of the receptor [52]. EGFR was reconstituted in proteoliposomes

Please cite this article as: A.M. Ernst, B. Brügger, Sphingolipids as modulators of membrane proteins, Biochim. Biophys. Acta (2013), http:// dx.doi.org/10.1016/j.bbalip.2013.10.016

4

A.M. Ernst, B. Brügger / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

and assessed for modulation by chemically-defined lipid compositions. In the presence of GM3, EGF binding was not affected, but a strong reduction in dimerisation and autophosphorylation was observed. Moreover, enzymatic removal of neuraminic acid from GM3 restored autophosphorylation and dimerisation of the receptor. In accordance with the juxtamembrane lysine residue found to be critical for the interaction of GM3 with the insulin receptor (see below), a juxtamembrane lysine residue (Lys642) was identified in EGFR that was close in proximity to the exoplasmic leaflet. This residue indeed appeared to be important for GM3 binding, as mutation of Lys642 did not affect EGF binding, yet led to insensitivity of the receptor towards GM3. Moreover, a recent molecular dynamics simulation study suggested that an interaction of anionic lipids (in this case POPS) in the cytosolic leaflet occurs with basic residues of the juxtamembrane segments of EGFR, thus leading to the formation of an active kinase dimer [53]. In addition, this basic juxtamembrane segment of EGFR was suggested to mediate binding of PIP2 to the receptor, suggesting a novel role of PIP2 as a positive modulator of EGFR activity [54].

3.2.2. Glycosphingolipids as allosteric effectors of the insulin receptor (IR) Another well-studied example of glycosphingolipids acting as allosteric effectors is with the IR, which is required for the regulation of insulinmediated metabolic control in animal cells. Employing an in vitro phosphorylation assay and detergent-solubilised IR from B lymphoblast IM9 cells, it was shown that insulin-dependent autophosphorylation of the receptor was fully inhibited in the presence of unbranched neolacto series gangliosides possessing a NeuAc2➝3Gal terminus [55]. Partial inhibition was observed in the presence of GM3 and GD3, but not other gangliosides. Similarly, a correlation between TNF-α-mediated insulin resistance and increased levels of GM3 synthase was reported [56]. The addition of GM3 to adipocytes inhibited autophosphorylation of IR, whereas depletion of GM3 rendered the adipocytes unresponsive to TNF-α-mediated resistance to insulin. In this context, another report described a correlation between GM3 and insulin-mediated signalling, as mice lacking GM3 synthase exhibited a heightened sensitivity to insulin [57]. Further insight into the molecular mechanism of the observed inhibition of IR by GM3 were gained from a detailed study showing that IR localises to two distinct types of microdomains: caveolae through interactions with caveolin-1, and GEMs containing GM3 [58]. Interestingly, the amount of GEM-localised IR is higher in a state of insulin resistance. Again, a lysine residue (Lys956) in proximity of the exoplasmic leaflet appears to influence the association of IR to GEMs. It has been suggested that successful insulin signalling requires the IR to localise to caveolae through interactions with caveolin-1. During a state of insulin resistance, autophosphorylation of IR is inhibited due to its sequestration from caveolae to GEMs.

3.2.3. Other receptors regulated by glycosphingolipids Fibroblast growth factor-2 (FGF-2) is another example of negative regulation of RTKs by glycosphingolipids. It has been shown that a number of glycosphingolipids, including GT1b, GD1b, GM1, GM2, GM3, sulfatide, and galactosylceramide can prevent binding of FGF-2 to its high-affinity FGF receptor, both in a free or cell-associated state [59,60]. In this case, sphingolipids indirectly regulate a RTK by modulating the accessibility of the ligand. In contrast, signalling via nerve growth factor (NGF) and the high-affinity nerve growth factor receptor TrkA provides an example of positive regulation of RTKs by glycosphingolipids (in this case GM1) [61–64]. Although the molecular mechanisms leading to enhanced signalling have not yet been elucidated, it has been suggested that GM1 may stimulate phosphatidylinositol-3-kinase activity [65]. Taken together, several studies have shown that RTKs can be delicately modulated by binding to glycosphingolipids, which can lead to sequestration of the receptors in different types of nano-domains or prevention of receptor dimerisation.

3.3. Tetraspanin- and ganglioside-containing microdomains Tetraspanins (the transmembrane four superfamily) are cell surface membrane proteins that contain four transmembrane domains as well as a small (EC1) and large (EC2) extracellular loop. The sequences of their transmembrane domains as well as CCG and PXSC motifs in EC2 are highly conserved and distinguish tetraspanins from other fourtransmembrane domain-containing proteins [66]. Tetraspanins are expressed in a variety of tissues and act as inner membrane scaffolds or hubs in organising cellular adhesion, motility, immune responses, and signalling [67,68]. Interestingly, tetraspanins appear to possess an intrinsic affinity for glycosphingolipids and are key organisers of GEMs, which, together with adhesive and/or signalling proteins, result in supramolecular complexes at the cell surface, termed “glycosynapses” [69]. Type 3 glycosynapses, which are defined as microdomains enriched with gangliosides, tetraspanins, and adhesion receptors (such as integrins), are of particular interest. Similar to integrin–ganglioside–tetraspanin interactions, it has been shown that the binding of fibronectin to integrins depends on the presence of GM3 [70]. Cell lines lacking GM3 exhibited a weak interaction between fibronectin and integrins, whereas binding was restored upon addition of GM3 to GM3-deficient cells. Cell motility and adhesion are tightly interconnected processes, and therefore Ono and colleagues (2001) focused on the integrin-associated tetraspanin CD9 [71]. Using photoactivatable, tritiated GM3, specific binding of GM3 to CD9 was shown. The motility of cells expressing CD9 at high levels but lacking GM3 was inhibited by the addition of GM3, but not by other gangliosides. In addition, a complex consisting of the tetraspanin CD82 and the integrins alpha3 and 5 has been identified in GEMs [72]. For CD82, its interaction with integrins appears to depend on three sites of N-linked glycans. A subsequent study unequivocally demonstrated that CD82 specifically and exclusively interacts with GM2 [73], and this complex was found to regulate cell motility by inhibiting hepatocyte growth factor (HGF)-induced activation of Met tyrosine kinase, which functionally interacts with alpha3beta1 integrin. In addition, another study found that the addition of nanospheres coated with Ca2+-dependent GM2/GM3 dimers to highly malignant YTS-1 cells inhibited cMet activity and cellular motility, suggesting that this approach may have therapeutic potential [74]. Together, these results show that tetraspanins organise GEMs and are required for cellular motility and signalling in animal cells. 4. Conclusion The diversity of sphingolipids and membrane lipids in general is matched by a tremendous diversity of the membrane proteome of higher eukaryotes. Sphingolipids are remarkable membrane constituents, as they bear self-organising potential and unique physicochemical properties that result in a specific lipid interactome and the generation of lateral heterogeneities in biological membranes. Although several reports hint at the involvement of sphingolipids as regulators of membrane protein function, only a few have studied the specificity of interactions of sphingolipids with membrane proteins and revealed functional consequences for the interacting protein (Table 1). This indicates that membrane proteins have adapted to and are specifically regulated by sphingolipids. In particular, the signalling of some RTKs, such as the IR and EGFR, is critically modulated by gangliosides. Other examples include the superfamily of tetraspanins, which utilise gangliosides as cofactors to organise GEMs and are necessary for a multitude of signalling networks involved in cell motility and growth factor responses. Other soluble proteins possess SBDs that specifically recognise sphingolipids in the plasma membrane of higher eukaryotes and employ them as receptors or for organisation of fusion complexes formed by viruses. A recent report suggests that selectivity of binding to different sphingolipids is not only restricted to the composition of their polar head group moiety, as the specificity of p24 interaction was restricted to a single molecular sphingomyelin species (C18:0). In

Please cite this article as: A.M. Ernst, B. Brügger, Sphingolipids as modulators of membrane proteins, Biochim. Biophys. Acta (2013), http:// dx.doi.org/10.1016/j.bbalip.2013.10.016

A.M. Ernst, B. Brügger / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

Cluster of Excellence (EXC81). We thank Erik Lindahl (Stockholm) for molecular dynamics simulations of the p24–SM interaction.

Table 1 Examples of direct protein–sphingolipid interactions. Protein

Sphingolipid

Equinatoxin II (EqtII) Lysenin

sphingomyelin

gp120

ß-amyloid peptide (Aß) Prion protein (PrP) Bile salt-dependent lipase (BSDL) p24 Insulin receptor (IR) Platelet-derived growth factor receptor (PDGFR) Epidermal growth factor receptor (EGFR) Fibroblast growth factor receptor (FGFR) TrkA (high-affinity nerve growth factor receptor) CD9

CD82

Cofactor in

binding to target cells; cytolysis sphingomyelin binding to target cells; cytolysis glycosphingolipids; promoting fusion sphingomyelin complex between host membranes/HIV-1 glycosphingolipids orientation in non-pathological conformations galactosyceramide; orientation in sphingomyelin non-pathological conformations glucosylceramide; protein folding galactosylceramide SM 18:0 dimerisation GM3 negative regulation of autophosphorylation GM1; GM3 negative regulation of autophosphorylation

Selected references

References

[31] [33] [37,38]

[36]

[39]

[41] [28] [53–55] [48]

GM3

negative regulation of autophosphorylation

[49,50]

glycosphingolipids

inhibition of FGF-2-binding

[56]

GM1

enhancement of signal transduction

[57]

GM3

complex formation with integrins and tetraspanins; regulation of cellular motility inhibition of integrins via cMet; regulation of cellular motility

[68]

GM2

5

[70,71]

this case, the protein–lipid interaction promoted dimerisation, thereby allosterically stabilising the functional state of p24 involved in COPIvesicle-mediated trafficking. The unexpected high specificity of lipid recognition by transmembrane proteins opens exciting and yet mostly unexplored targets for therapeutics since these motifs have been identified on receptors involved in immune responses and inflammation. Specific protein–sphingolipid interactions are unlikely to be rare events, and in this context, p24 serves well as a precedent. A combination of mass spectrometric, in vitro, in silico, and cellular approaches support the molecular species selectivity of p24. Although multiple reports suggest that sphingolipids are involved in membrane protein function, the specificity of these interactions has not been extensively explored at the molecular level to date. To address this issue, chemical biology tools continue to emerge that can be used to explore these mechanisms (reviewed by Haberkant et al. [75]), such as photoactivatable lipids or fluorescent lipid analogues, which will likely identify additional membrane proteins that have adapted to distinct molecular lipid species. Membrane biologists may now begin to appreciate the diversity of membrane lipids as a source of highly specific allosteric effectors of membrane protein function. Moreover, molecular insight into specific protein–sphingolipid interactions opens a new avenue in pharmaceutical research with novel classes of specific lipid molecular drug targets.

Acknowledgements B.B. is supported by grants from the Deutsche Forschungsgemeinschaft within SPP 1175 and TRR 83 and is investigator of the CellNetworks

[1] H.J. Kaiser, M.A. Surma, F. Mayer, I. Levental, M. Grzybek, R.W. Klemm, S. Da Cruz, C. Meisinger, V. Muller, K. Simons, D. Lingwood, Molecular convergence of bacterial and eukaryotic surface order, J. Biol. Chem. 286 (2011) 40631–40637. [2] Y.A. Hannun, L.M. Obeid, Principles of bioactive lipid signalling: lessons from sphingolipids, Nat. Rev. Mol. Cell Biol. 9 (2008) 139–150. [3] J.P. Slotte, Biological functions of sphingomyelins, Prog. Lipid Res. 52 (2013) 424–437. [4] J.C. Holthuis, T. Pomorski, R.J. Raggers, H. Sprong, G. Van Meer, The organizing potential of sphingolipids in intracellular membrane transport, Physiol. Rev. 81 (2001) 1689–1723. [5] P.S. Niemela, M.T. Hyvonen, I. Vattulainen, Atom-scale molecular interactions in lipid raft mixtures, Biochim. Biophys. Acta 1788 (2009) 122–135. [6] K. Simons, E. Ikonen, Functional rafts in cell membranes, Nature 387 (1997) 569–572. [7] N.M. Hooper, Detergent-insoluble glycosphingolipid/cholesterol-rich membrane domains, lipid rafts and caveolae (review), Mol. Membr. Biol. 16 (1999) 145–156. [8] A.M. Ernst, F.X. Contreras, C. Thiele, F. Wieland, B. Brügger, Mutual recognition of sphingolipid molecular species in membranes, Biochim. Biophys. Acta 1818 (2012) 2616–2622. [9] W. Guo, V. Kurze, T. Huber, N.H. Afdhal, K. Beyer, J.A. Hamilton, A solid-state NMR study of phospholipid–cholesterol interactions: sphingomyelin–cholesterol binary systems, Biophys. J. 83 (2002) 1465–1478. [10] J. Huang, G.W. Feigenson, A microscopic interaction model of maximum solubility of cholesterol in lipid bilayers, Biophys. J. 76 (1999) 2142–2157. [11] M. Lonnfors, J.P. Doux, J.A. Killian, T.K. Nyholm, J.P. Slotte, Sterols have higher affinity for sphingomyelin than for phosphatidylcholine bilayers even at equal acyl-chain order, Biophys. J. 100 (2011) 2633–2641. [12] B. Ramstedt, J.P. Slotte, Interaction of cholesterol with sphingomyelins and acyl-chain-matched phosphatidylcholines: a comparative study of the effect of the chain length, Biophys. J. 76 (1999) 908–915. [13] B.Y. van Duyl, D.T. Rijkers, B. de Kruijff, J.A. Killian, Influence of hydrophobic mismatch and palmitoylation on the association of transmembrane alpha-helical peptides with detergent-resistant membranes, FEBS Lett. 523 (2002) 79–84. [14] C. Wolf, K. Koumanov, B. Tenchov, P.J. Quinn, Cholesterol favors phase separation of sphingomyelin, Biophys. Chem. 89 (2001) 163–172. [15] K.J. Tierney, D.E. Block, M.L. Longo, Elasticity and phase behavior of DPPC membrane modulated by cholesterol, ergosterol, and ethanol, Biophys. J. 89 (2005) 2481–2493. [16] A. Kusumi, T.K. Fujiwara, R. Chadda, M. Xie, T.A. Tsunoyama, Z. Kalay, R.S. Kasai, K.G. Suzuki, Dynamic organizing principles of the plasma membrane that regulate signal transduction: commemorating the fortieth anniversary of Singer and Nicolson's fluid-mosaic model, Annu. Rev. Cell Dev. Biol. 28 (2012) 215–250. [17] K. Simons, J.L. Sampaio, Membrane organization and lipid rafts, Cold Spring Harb. Perspect. Biol. 3 (2011) a004697. [18] S. Sonnino, A. Prinetti, Membrane domains and the “lipid raft” concept, Curr. Med. Chem. 20 (2013) 4–21. [19] B. Cannon, A. Lewis, J. Metze, V. Thiagarajan, M.W. Vaughn, P. Somerharju, J. Virtanen, J. Huang, K.H. Cheng, Cholesterol supports headgroup superlattice domain formation in fluid phospholipid/cholesterol bilayers, J. Phys. Chem. B 110 (2006) 6339–6350. [20] Z.T. Graber, Z. Jiang, A. Gericke, E.E. Kooijman, Phosphatidylinositol-4,5-bisphosphate ionization and domain formation in the presence of lipids with hydrogen bond donor capabilities, Chem. Phys. Lipids 165 (2012) 696–704. [21] S. Sennato, F. Bordi, C. Cametti, C. Coluzza, A. Desideri, S. Rufini, Evidence of domain formation in cardiolipin-glycerophospholipid mixed monolayers. A thermodynamic and AFM study, J. Phys. Chem. B 109 (2005) 15950–15957. [22] D.M. Engelman, Membranes are more mosaic than fluid, Nature 438 (2005) 578–580. [23] S. Sonnino, L. Mauri, M.G. Ciampa, A. Prinetti, Gangliosides as regulators of cell signaling: ganglioside–protein interactions or ganglioside-driven membrane organization? J. Neurochem. 124 (2013) 432–435. [24] H.J. Sharpe, T.J. Stevens, S. Munro, A comprehensive comparison of transmembrane domains reveals organelle-specific properties, Cell 142 (2010) 158–169. [25] A. Colbeau, J. Nachbaur, P.M. Vignais, Enzymic characterization and lipid composition of rat liver subcellular membranes, Biochim. Biophys. Acta 249 (1971) 462–492. [26] M.A. Surma, C. Klose, K. Simons, Lipid-dependent protein sorting at the trans-Golgi network, Biochim. Biophys. Acta 1821 (2012) 1059–1067. [27] N. Kahya, D.A. Brown, P. Schwille, Raft partitioning and dynamic behavior of human placental alkaline phosphatase in giant unilamellar vesicles, Biochemistry 44 (2005) 7479–7489. [28] G. van Meer, D.R. Voelker, G.W. Feigenson, Membrane lipids: where they are and how they behave, Nat. Rev. Mol. Cell Biol. 9 (2008) 112–124. [29] R. Mahfoud, N. Garmy, M. Maresca, N. Yahi, A. Puigserver, J. Fantini, Identification of a common sphingolipid-binding domain in Alzheimer, prion, and HIV-1 proteins, J. Biol. Chem. 277 (2002) 11292–11296. [30] F.X. Contreras, A.M. Ernst, P. Haberkant, P. Bjorkholm, E. Lindahl, B. Gonen, C. Tischer, A. Elofsson, G. von Heijne, C. Thiele, R. Pepperkok, F. Wieland, B. Brügger, Molecular recognition of a single sphingolipid species by a protein's transmembrane domain, Nature 481 (2012) 525–529. [31] K.C. Kristan, G. Viero, M. Dalla Serra, P. Macek, G. Anderluh, Molecular mechanism of pore formation by actinoporins, Toxicon 54 (2009) 1125–1134.

Please cite this article as: A.M. Ernst, B. Brügger, Sphingolipids as modulators of membrane proteins, Biochim. Biophys. Acta (2013), http:// dx.doi.org/10.1016/j.bbalip.2013.10.016

6

A.M. Ernst, B. Brügger / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

[32] A.W. Bernheimer, L.S. Avigad, Properties of a toxin from the sea anemone Stoichacis helianthus, including specific binding to sphingomyelin, Proc. Natl. Acad. Sci. U. S. A. 73 (1976) 467–471. [33] B. Bakrac, I. Gutierrez-Aguirre, Z. Podlesek, A.F. Sonnen, R.J. Gilbert, P. Macek, J.H. Lakey, G. Anderluh, Molecular determinants of sphingomyelin specificity of a eukaryotic pore-forming toxin, J. Biol. Chem. 283 (2008) 18665–18677. [34] A.M. Ernst, F.X. Contreras, B. Brügger, F. Wieland, Determinants of specificity at the protein–lipid interface in membranes, FEBS Lett. 584 (2010) 1713–1720. [35] L. De Colibus, A.F. Sonnen, K.J. Morris, C.A. Siebert, P. Abrusci, J. Plitzko, V. Hodnik, M. Leippe, E. Volpi, G. Anderluh, R.J. Gilbert, Structures of lysenin reveal a shared evolutionary origin for pore-forming proteins and its mode of sphingomyelin recognition, Structure 20 (2012) 1498–1507. [36] D. Hammache, N. Yahi, M. Maresca, G. Pieroni, J. Fantini, Human erythrocyte glycosphingolipids as alternative cofactors for human immunodeficiency virus type 1 (HIV-1) entry: evidence for CD4-induced interactions between HIV-1 gp120 and reconstituted membrane microdomains of glycosphingolipids (Gb3 and GM3), J. Virol. 73 (1999) 5244–5248. [37] L.P. Choo-Smith, W. Garzon-Rodriguez, C.G. Glabe, W.K. Surewicz, Acceleration of amyloid fibril formation by specific binding of Abeta-(1–40) peptide to gangliosidecontaining membrane vesicles, J. Biol. Chem. 272 (1997) 22987–22990. [38] L.P. Choo-Smith, W.K. Surewicz, The interaction between Alzheimer amyloid beta(1–40) peptide and ganglioside GM1-containing membranes, FEBS Lett. 402 (1997) 95–98. [39] D. Hammache, G. Pieroni, N. Yahi, O. Delezay, N. Koch, H. Lafont, C. Tamalet, J. Fantini, Specific interaction of HIV-1 and HIV-2 surface envelope glycoproteins with monolayers of galactosylceramide and ganglioside GM3, J. Biol. Chem. 273 (1998) 7967–7971. [40] D. Hammache, N. Yahi, G. Pieroni, F. Ariasi, C. Tamalet, J. Fantini, Sequential interaction of CD4 and HIV-1 gp120 with a reconstituted membrane patch of ganglioside GM3: implications for the role of glycolipids as potential HIV-1 fusion cofactors, Biochem. Biophys. Res. Commun. 246 (1998) 117–122. [41] S.B. Prusiner, Prions, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 13363–13383. [42] K.E. Ward, N. Bhardwaj, M. Vora, C.E. Chalfant, H. Lu, R.V. Stahelin, The molecular basis of ceramide-1-phosphate recognition by C2 domains, J. Lipid Res. 54 (2013) 636–648. [43] E. Aubert-Jousset, N. Garmy, V. Sbarra, J. Fantini, M.O. Sadoulet, D. Lombardo, The combinatorial extension method reveals a sphingolipid binding domain on pancreatic bile salt-dependent lipase: role in secretion, Structure 12 (2004) 1437–1447. [44] V. Popoff, F. Adolf, B. Brugger, F. Wieland, COPI budding within the Golgi stack, Cold Spring Harb. Perspect. Biol. 3 (2011) a005231. [45] B. Brügger, R. Sandhoff, S. Wegehingel, K. Gorgas, J. Malsam, J.B. Helms, W.D. Lehmann, W. Nickel, F.T. Wieland, Evidence for segregation of sphingomyelin and cholesterol during formation of COPI-coated vesicles, J. Cell Biol. 151 (2000) 507–518. [46] M.A. Lemmon, J. Schlessinger, Cell signaling by receptor tyrosine kinases, Cell 141 (2010) 1117–1134. [47] C.W. Ward, M.C. Lawrence, V.A. Streltsov, T.E. Adams, N.M. McKern, The insulin and EGF receptor structures: new insights into ligand-induced receptor activation, Trends Biochem. Sci. 32 (2007) 129–137. [48] A. Ullrich, J. Schlessinger, Signal transduction by receptors with tyrosine kinase activity, Cell 61 (1990) 203–212. [49] R.A. Laine, S. Hakomori, Incorporation of exogenous glycosphingolipids in plasma membranes of cultured hamster cells and concurrent change of growth behavior, Biochem. Biophys. Res. Commun. 54 (1973) 1039–1045. [50] E.G. Bremer, S. Hakomori, D.F. Bowen-Pope, E. Raines, R. Ross, Ganglioside-mediated modulation of cell growth, growth factor binding, and receptor phosphorylation, J. Biol. Chem. 259 (1984) 6818–6825. [51] E.G. Bremer, J. Schlessinger, S. Hakomori, Ganglioside-mediated modulation of cell growth. Specific effects of GM3 on tyrosine phosphorylation of the epidermal growth factor receptor, J. Biol. Chem. 261 (1986) 2434–2440. [52] U. Coskun, M. Grzybek, D. Drechsel, K. Simons, Regulation of human EGF receptor by lipids, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 9044–9048. [53] A. Arkhipov, Y. Shan, R. Das, N.F. Endres, M.P. Eastwood, D.E. Wemmer, J. Kuriyan, D.E. Shaw, Architecture and membrane interactions of the EGF receptor, Cell 152 (2013) 557–569. [54] I.E. Michailidis, R. Rusinova, A. Georgakopoulos, Y. Chen, R. Iyengar, N.K. Robakis, D.E. Logothetis, L. Baki, Phosphatidylinositol-4,5-bisphosphate regulates epidermal growth factor receptor activation, Pflugers Arch. 461 (2011) 387–397.

[55] H. Nojiri, M. Stroud, S. Hakomori, A specific type of ganglioside as a modulator of insulin-dependent cell growth and insulin receptor tyrosine kinase activity. Possible association of ganglioside-induced inhibition of insulin receptor function and monocytic differentiation induction in HL-60 cells, J. Biol. Chem. 266 (1991) 4531–4537. [56] S. Tagami, J. Inokuchi Ji, K. Kabayama, H. Yoshimura, F. Kitamura, S. Uemura, C. Ogawa, A. Ishii, M. Saito, Y. Ohtsuka, S. Sakaue, Y. Igarashi, Ganglioside GM3 participates in the pathological conditions of insulin resistance, J. Biol. Chem. 277 (2002) 3085–3092. [57] T. Yamashita, A. Hashiramoto, M. Haluzik, H. Mizukami, S. Beck, A. Norton, M. Kono, S. Tsuji, J.L. Daniotti, N. Werth, R. Sandhoff, K. Sandhoff, R.L. Proia, Enhanced insulin sensitivity in mice lacking ganglioside GM3, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 3445–3449. [58] K. Kabayama, T. Sato, K. Saito, N. Loberto, A. Prinetti, S. Sonnino, M. Kinjo, Y. Igarashi, J. Inokuchi, Dissociation of the insulin receptor and caveolin-1 complex by ganglioside GM3 in the state of insulin resistance, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 13678–13683. [59] M. Rusnati, E. Tanghetti, C. Urbinati, G. Tulipano, S. Marchesini, M. Ziche, M. Presta, Interaction of fibroblast growth factor-2 (FGF-2) with free gangliosides: biochemical characterization and biological consequences in endothelial cell cultures, Mol. Biol. Cell 10 (1999) 313–327. [60] M. Rusnati, C. Urbinati, E. Tanghetti, P. Dell'Era, H. Lortat-Jacob, M. Presta, Cell membrane GM1 ganglioside is a functional coreceptor for fibroblast growth factor 2, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 4367–4372. [61] S.J. Rabin, I. Mocchetti, GM1 ganglioside activates the high-affinity nerve growth factor receptor trkA, J. Neurochem. 65 (1995) 347–354. [62] T. Mutoh, T. Hamano, S. Yano, H. Koga, H. Yamamoto, K. Furukawa, R.W. Ledeen, Stable transfection of GM1 synthase gene into GM1-deficient NG108-15 cells, CR-72 cells, rescues the responsiveness of Trk-neurotrophin receptor to its ligand, NGF, Neurochem. Res. 27 (2002) 801–806. [63] T. Mutoh, A. Tokuda, J. Inokuchi, M. Kuriyama, Glucosylceramide synthase inhibitor inhibits the action of nerve growth factor in PC12 cells, J. Biol. Chem. 273 (1998) 26001–26007. [64] T. Mutoh, A. Tokuda, T. Miyadai, M. Hamaguchi, N. Fujiki, Ganglioside GM1 binds to the Trk protein and regulates receptor function, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 5087–5091. [65] A.M. Duchemin, Q. Ren, N.H. Neff, M. Hadjiconstantinou, GM1-induced activation of phosphatidylinositol 3-kinase: involvement of Trk receptors, J. Neurochem. 104 (2008) 1466–1477. [66] M.D. Wright, J. Ni, G.B. Rudy, The L6 membrane proteins—a new four-transmembrane superfamily, Protein Sci. 9 (2000) 1594–1600. [67] A.R. Todeschini, S.I. Hakomori, Functional role of glycosphingolipids and gangliosides in control of cell adhesion, motility, and growth, through glycosynaptic microdomains, Biochim. Biophys. Acta 1780 (2008) 421–433. [68] J.M. Tarrant, L. Robb, A.B. van Spriel, M.D. Wright, Tetraspanins: molecular organisers of the leukocyte surface, Trends Immunol. 24 (2003) 610–617. [69] S. Hakomori, The glycosynapse, Proc. Natl. Acad. Sci. U. S. A. 8 (2002) 225–232. [70] M. Zheng, H. Fang, T. Tsuruoka, T. Tsuji, T. Sasaki, S. Hakomori, Regulatory role of GM3 ganglioside in alpha 5 beta 1 integrin receptor for fibronectin-mediated adhesion of FUA169 cells, J. Biol. Chem. 268 (1993) 2217–2222. [71] M. Ono, K. Handa, S. Sonnino, D.A. Withers, H. Nagai, S. Hakomori, GM3 ganglioside inhibits CD9-facilitated haptotactic cell motility: coexpression of GM3 and CD9 is essential in the downregulation of tumor cell motility and malignancy, Biochemistry 40 (2001) 6414–6421. [72] M. Ono, K. Handa, D.A. Withers, S. Hakomori, Glycosylation effect on membrane domain (GEM) involved in cell adhesion and motility: a preliminary note on functional alpha3, alpha5-CD82 glycosylation complex in ldlD 14 cells, Biochem. Biophys. Res. Commun. 279 (2000) 744–750. [73] A.R. Todeschini, J.N. Dos Santos, K. Handa, S.I. Hakomori, Ganglioside GM2-tetraspanin CD82 complex inhibits met and its cross-talk with integrins, providing a basis for control of cell motility through glycosynapse, J. Biol. Chem. 282 (2007) 8123–8133. [74] A.R. Todeschini, J.N. Dos Santos, K. Handa, S.I. Hakomori, Ganglioside GM2/GM3 complex affixed on silica nanospheres strongly inhibits cell motility through CD82/cMet-mediated pathway, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 1925–1930. [75] P. Haberkant, G. van Meer, Protein-lipid interactions: paparazzi hunting for snapshots, Biological chemistry 390 (2009) 795–803.

Please cite this article as: A.M. Ernst, B. Brügger, Sphingolipids as modulators of membrane proteins, Biochim. Biophys. Acta (2013), http:// dx.doi.org/10.1016/j.bbalip.2013.10.016