Extended-Synaptotagmins (E-Syts); the extended story

Extended-Synaptotagmins (E-Syts); the extended story

Pharmacological Research 107 (2016) 48–56 Contents lists available at ScienceDirect Pharmacological Research journal homepage: www.elsevier.com/loca...

2MB Sizes 2 Downloads 37 Views

Pharmacological Research 107 (2016) 48–56

Contents lists available at ScienceDirect

Pharmacological Research journal homepage: www.elsevier.com/locate/yphrs

Invited review

Extended-Synaptotagmins (E-Syts); the extended story Chelsea Herdman, Tom Moss ∗ Laboratory of Growth and Development, St-Patrick Research Group in Basic Oncology, Cancer Division of the Québec University Hospital Research Centre, Department of Molecular Biology, Medical Biochemistry and Pathology, Faculty of Medicine, Laval University, Edifice St Patrick, 9 rue McMahon, Québec, QC G1R 3S3, Canada

a r t i c l e

i n f o

Article history: Received 22 January 2016 Received in revised form 24 January 2016 Accepted 24 January 2016 Available online 27 February 2016 Keywords: Extended Synaptotagmin (E-Syt,ESYT) FGF Signaling Calcium Signaling Phospholipid Metabolism EndoplasmicReticulum (ER) - Plasma Membrane(PM) Junctions ER-PM Tethering

a b s t r a c t The Extended-Synaptotagmin (E-Syt) membrane proteins were only recently discovered, but have already been implicated in a range of interrelated cellular functions, including calcium and receptor signaling, and membrane lipid transport. However, despite their evolutionary conservation and detailed studies of their molecular actions, we still have little idea of how and when these proteins are required in cellular and organism physiology. Here we review our present understanding of the E-Syts and discuss the molecular functions and in vivo requirements for these proteins. © 2016 Elsevier Ltd. All rights reserved.

Contents 1. 2.

3.

4. 5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 E-Syt structure-function relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 2.1. Membrane anchoring and E-Syt dimerization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 2.2. The synaptotagmin-like mitochondrial-lipid binding protein (SMP) domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 2.3. The C2 domains and Ca2+ coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 2.4. Lipid-binding properties of the E-Syt C2 domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 2.5. PM association via combined C2 action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 E-Syts and cell signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.1. E-Syt functions in receptor signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.2. E-Syts in the formation of ER-PM contact sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.3. E-Syt functions in lipid transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.4. model for E-Syt function in FGF receptor signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 E-Syt functions in vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

1. Introduction The families of multiple C2 domain proteins, such as the synaptotagmins, Multiple C2 Transmembrane Proteins (MCTPs), ferlins, and most recently the yeast tricalbins and the mammalian

∗ Corresponding author. E-mail address: [email protected] (T. Moss). http://dx.doi.org/10.1016/j.phrs.2016.01.034 1043-6618/© 2016 Elsevier Ltd. All rights reserved.

extended-synaptotagmins have been linked with membrane trafficking, cell signaling and/or membrane tethering and fusion events, especially those involving Ca2+ and phospholipids. They are all integral membrane proteins containing distinct membrane spanning or trans-membrane (TM) sequences, but while the MCTPs and ferlins contain a C-terminal TM domain, the TM domains of the synaptotagmins, extended-synaptotagmins and tricalbins are Nterminal. In particular, the synaptotagmins contain an N-terminal TM domain followed by two closely spaced C2 domains. Thus, when

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

49

Fig. 1. Diagram of the extended-synaptotagmin and tricalbin domain structures. The C2 domains of the human E-Syts, but not the yeast Tcbs, are shaded to demonstrate domain homologies. The N-terminus of E-Syt2b is shaded to show the variable region between the splice variants. The structural features are represented approximately to scale relative to the primary sequence.

a new family of human multi-C2 domain proteins were found to have a similar structural layout they were classified as extendedsynaptotagmins. Three extended-synaptotagmins (E-Syt1, 2 and 3) were first identified in human and most obviously differed from the classic synaptotagmins in having three or five tandem C2 domains [1,2]. Subsequent studies have revealed that in fact they are endoplasmic reticulum (ER) proteins and are functionally similar to the tricalbins (Tcbs) of yeast [3,4]. E-Syt1 has been implicated in glucose transporter GLUT4 function [5], E-Syt2 in fibroblast growth factor (FGF) signaling, receptor endocytosis and cytoskeleton regulation [6–8], and most recently all three E-Syts have been implicated in ER-PM tethering and the formation of membrane contact sites, and in lipid transport and Ca2+ signaling [3,4,9,10]. However, despite their evolutionary conservation, as yet very few studies have demonstrated an essential physiological requirement for either the tricalbins or the E-Syts.

2. E-Syt structure-function relationships Members of the E-Syt protein family exist throughout the eukaryotic kingdom. In metazoa, three distinct E-Syt forms, E-Syt1, 2 and 3, can be clearly identified [11]. In contrast, this one to one identity is to a large extent lost between higher and lower eukaryotes, though clear structural parallels can be made even between the mammalian E-Syts [1,2] and the Tcbs of yeast [12]. Both the E-Syts and Tcbs have three or more C2 domains, a predicted N-terminal TM domain and an intervening synaptotagmin-like mitochondrial-lipid binding protein (SMP) domain, (Fig. 1). The in vivo functions of Xenopus E-Syt2 [6,7] and the murine E-Syt2 and 3 [13] have been investigated, and will be discussed below. However, most functional studies of the E-Syts have been performed in cell culture using ectopic expression of the human forms [1–3,6,8,10,14,15].

2.1. Membrane anchoring and E-Syt dimerization In the light microscope, human E-Syt2 and 3 and Xenopus ESyt2 appear to be predominantly localized to the cell periphery, whereas human E-Syt1 displays a more cytoplasmic distribution [1,6]. However, electron microscopy has revealed that the E-Syts and the Tcbs are actually anchored in the ER membrane [3,4]. This contrasts with the synaptotagmins (Syts), which predominantly associate with synaptic vesicles, the PM or the post-Golgi membrane [16,17]. The Syts also traverse these membranes such that N and C-terminal regions lie on either side of the membrane. In contrast, the putative TM domains of all three E-Syts tend to be a little longer than a typical single pass transmembrane sequence and similar in length to other ER proteins such as the reticulons that are anchored in the ER membrane, but do not penetrate it [18]. Neither N- nor C-termini of the E-Syts are accessible on the cell-surface, or indeed from within the ER [3,8]. Thus, membrane anchoring of both E-Syts and Tcbs likely occurs through a hairpin fold in the TM domain, such that this domain does not fully traverse the ER membrane or does so only over an extremely short region [3]. Replacing the first 119 a.a. of E-Syt2b, including its TM domain (Fig. 1), by the equivalent region of Syt1 not only redirects it to the PM, but causes it to penetrate the membrane such that its N-terminus is now available on the cell-surface [8]. Thus, the specificity of E-Syt anchoring in the ER membrane is a property intrinsic to the N-terminal TM domain and the immediately surrounding sequences. Interestingly, the Xenopus and human E-Syts homo- and hetero-dimerize, but do not interact with Syt1 [3,6,8]. The minimal dimerization domain of human E-Syt2 lies within a 52 a.a. sequence that includes the 28/29 a.a. TM domain [8]. This suggests that despite their very different membrane anchoring, dimerization of the E-Syts, like that of the Syts, occurs within or immediately adjacent to the anchoring membrane. Though the TM domain appears to represent the major E-Syt dimerization site, other E-Syt domains also have the potential to dimerize. The SMP domain of E-Syt2 was found to crystallize as a dimer [19], see below, and yeast 2-hybrid data suggests that the

50

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

CLUSTAL O(1.2.1) multiple sequence alignment hE-Syt1C2A, hE-Syt2C2A, hE-Syt3C2A, hE-Syt1C2C,

GIIRIHLLAARGLSSKDKYVKGLIEGKSDPYALVRLGTQTFCSRVIDEELNPQWGETYEVM GVLRIHFIEAQDLQGKDTYLKGLVKGKSDPYGIIRVGNQIFQSRVIKENLSPKWNEVYEAL GVIRVHLLEAEQLAQKDNFL-GL-RGKSDPYAKVSIGLQHFRSRTIYRNLNPTWNEVFEFM HVLRIHVLEAQDLIAKDRFLGGLVKGKSDPYVKLKLAGRSFRSHVVREDLNPRWNEVFEVI *. * ** :: ** .****** : :. : * *:.: .:*.* * *.:* : D D

Ca2+ coord. hE-Syt1C2A, hE-Syt2C2A, hE-Syt3C2A, hE-Syt1C2C,

VHEVPGQEIEVEVFDKDPDKDDFLGRMKLDVGKVLQASVLDDWFPLQG-GQGQVHLRLEWL VYEHPGQELEIELFDEDPDKDDFLGSLMIDLIEVEKERLLDEWFTLDEVPKGKLHLRLEWL VYEVPGQDLEVDLYDEDTDRDDFLGSLQICLGDVMTNRVVDEWFVLNDTTSGRLHLRLEWVTSVPGQELEVEVFDKDLDKDDFLGRCKVRLTTVLNSGFLDEWLTLEDVPSGRLHLRLERL * . ***::*::::*:* *:***** : : * .:*:*: *: .*::***** ED D DD

Ca2+ coord.

CLUSTAL O(1.2.1) multiple sequence alignment hE-Syt1C2B, hE-Syt2C2B, hE-Syt3C2B, hE-Syt1C2D,

GVSSRPDPPSAAILVVYLDRAQDLPL-KKGNKEPNPMVQLSIQDVTQESKAVYSTNCPVWE DKDQANDGLSSALLILYLDSARNLPSGKKISSNPNPVVQMSVGHKAQESKIRYKTNEPVWE SACNLPRNPFDYLNGEYRAKKLSRFARNKVSKDPSSYVKLSVGKKTHTSKTCPHNKDPVWS IQTQKSAELAAALLSIYMERAEDLPL-RKGTKHLSPYATLTVGDSSHKTKTISQTSAPVWD . : * . .* ... . . ::: . :: :* .. ***. N N N N

Potential Ca2+ coord. hE-Syt1C2B, hE-Syt2C2B, hE-Syt3C2B, hE-Syt1C2D,

EAFRFFLQDPQSQELDVQVKDDSRALTLGALTLPLARLLTAPELILDQWFQLSSSGPNSR ENFTFFIHNPKRQDLEVEVRDEQHQCSLGNLKVPLSQLLTSEDMTVSQRFQLSNSGPNST QVFSFFVHNVATERLHLKVLDDDQECALGMLEVPLCQILPYADLTLEQRFQLDHSGLDSL ESASFLIRKPHTESLELQVRGEGT-GVLGSLSLPLSELLVADQLCLDRWFTLSSGQGQVL : *:::. : *.::* : ** * :**..:* :: :.: * *. . DE N N

Potential Ca2+ coord.

CLUSTAL O(1.2.1) multiple sequence alignment hE-Syt1C2E, hE-Syt2C2C, hE-Syt3C2C,

PLGQVKLTLWYYSEERKLVSIVHGCRSLRQNGRDPPDPYVSLLLLPDKNRGTKRRTSQKKRT PLGQIQLTIRHSSQRNKLIVVVHACRNLIAFSEDGSDPYVRMYLLPDKRRSGRRKTHVSKKT QLGEIQLTVRYVCLRRCLSVLINGCRNLTPCTSSGADPYVRVYLLPERKWACRKKTSVKRKT **:::**: : . .. * ::..**.* . **** : ***::. . :::* .::* ED D -basic groove-

Potential Ca2+ coord. hE-Syt1C2E, hE-Syt2C2C, hE-Syt3C2C,

LSPEFNERFEWELPLDEAQRRKLDVSVKSNSSFMSRERELLGKVQLDLAETDLSQGVARWYDLLNPVFDQSFDFSVSLPEVQRRTLDVAVKNSGGFLSKDKGLLGKVLVALASEELAKGWTQWYDLT LEPLFDETFEFFVPMEEVKKRSLDVAVKNSRPLGSHRRKELGKVLIDLSKEDLIKGFSQWYELT *.* *:: *:: : : *.::*.***:**.. : *: : **** : *:. :* :* ::**:* D

Potential Ca2+ coord.

Fig. 2. Sequence alignments of the human E-Syt C2 domains generated by CLUSTAL [64]. The sequences for E-Syt1 are from Acc. No. NP 056107, E-Syt2a/b; DQ993201/NM 020728, E-Syt3; NM 031913. The identity of the Ca2+ coordinating residues in the C2A domain of E-Syt2 are taken from the C2A-C2B structure [29]. Two structures are presently available for the C2B domain of E-Syt2 [19,29], but the potential Ca2+ coordinating residues do not appear in either structure and are therefore indicated as potential coordinators based on a generic C2 domain structure. The identity of potential Ca2+ coordinating residues in the C2C domain of ESyt2 derives from the unpublished solution structure of this domain (Acc. No. 2DMG A), taking account of their position in this structure.

Plasma membrane

~20nm

E-Syt2/3

+Ca2+

+Ca2+

E-Syt1

E-Syt2/3

~10nm

+Ca2+

ER membrane

E-Syt1

Fig. 3. Coordination between C2 domains regulates the tightness of the ER-PM contact sites. E-Syt2/3 (in mauve) binds the PM through its C2C domain in a semi-constitutive manner. An increase in cytosolic Ca2+ would then tighten binding by enhancing the C2A/C2B interaction with PM phospholipids. E-Syt1 (in orange) does not significantly interact with PM phospholipids at low cytosolic Ca2+ , but could also bind in a progressively tighter manner at enhanced cytosolic Ca2+ .

C-termini of the Tcbs beyond the last C2 domain may also direct homo- and hetero-dimerization [12]. As will be seen below, the interaction of E-Syt1 with E-Syt2/3 is further modulated by receptor signaling and by changes in subcellular localization.

2.2. The synaptotagmin-like mitochondrial-lipid binding protein (SMP) domain All three E-Syts harbor an SMP domain starting 43–45 a.a. Cterminal to the TM domain (Fig. 1). SMP domains are ∼ 180a.a. folds

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

PIP5K

DAG

PI(4,5)P

PLC IP3

DGK

PI

PA

PA

CDS

CDP-DG

Plasma membrane

PS ORP 5/8

Nir2/3

Ca2+

IP3R

PI4KA

PI4P

51

PIS

PI

Sac1

PI4P

PS ER membrane

Ca2+ Fig. 4. Lipid transport between the endoplasmic reticulum and plasma membrane. Phosphatidylinositol 4,5-bisphosphate (PI(4,5)P) is cleaved to diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3) by PLC, which leads to the release of Ca2+ from the ER, enhancing the cytosolic Ca2+ concentration. DAG is converted to phosphatidic acid (PA) by DGK, and is transported to the ER membrane by the Nir2/3 phosphatidylinositol (PI) transfer proteins localized at ER-PM junctions. PA is converted by CDS and PIS to PI in the ER membrane in a two-step reaction via the intermediate cytidine diphosphate-diacylglycerol (CDP-DG). Nir2/3 transports this PI to the PM using the counter-transport of PA. PI is subsequently converted to phosphatidylinositol 4-phosphate (PI4P) by PI4KA and phosphorylated by PIP5K to replenish PI(4,5)P in the PM. PI4P levels in the PM are regulated by its transfer to the ER membrane by ORP5/8, using the counter-transport of phosphatidylserine (PS) to the PM. The PI4P in the ER membrane is then converted to PI by Sac1 phosphatase. Lipids and enzymes are shown in bold text respectively in black and blue. Abbreviations: IP3R (IP3 receptor), DGK (DAG kinase), CDS (CDP-DG synthase or phosphatidate cytidylyltransferase), PIS (PI synthase), PI4KA (PI 4 kinase alpha), PIP5K (PI4P 5 kinase).

belonging to the tubular-lipid-binding TULIP superfamily of lipid binding domains and are typically found in proteins implicated in the non-vesicular transfer of lipids between membranes [20,21]. The recent crystal structure of the SMP-C2A-C2B segment of E-Syt2 revealed that the SMP domain forms a ␤-barrel fold typical of the TULIP domains [19]. The hydrophobic channel of the crystallized SMP domain contained lipids derived from the membrane of the bacterial expression host. When the same construct was expressed in human cells, it was found to co-purify with a range of phospholipids (phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI), and phosphatidylserine (PS)) suggesting broad phospholipid binding properties. Though it is clear from these data that the SMP domains of the E-Syts have the potential to bind and transport lipids, whether this occurs in a physiological setting is as yet unknown. In the crystal structure, the E-Syt2 SMP domain formed an end-to-end dimer, suggesting that, in addition to dimerization via their TM domains, the E-Syts could also dimerize via their SMP domains. Interestingly, depletion of E-Syt2 reduced E-Syt1 binding to the PM at low, but not at high Ca2+ , while deletion of the SMP of E-Syt1 also affected its recruitment to the PM at low but not high Ca2+ [15]. This was interpreted as suggesting that the SMP could promote E-Syt1/2 heterodimerization. However, as yet it has not been demonstrated that the SMP domains of E-Syt1 and 2 can in fact interact. Further, deletion of the SMP domain of E-Syt2 had no effect on its capacity to dimerize in human cells [8]. Thus, E-Syt dimerization in vivo is unlikely to be driven by an SMP–SMP domain interaction, though this could provide additional stability.

2.3. The C2 domains and Ca2+ coordination The C2 domain was first identified as Conserved Domain 2 of the PKC kinase family, hence its name [22], and shown to be distinct from the C1 domain responsible for diacylglycerol and phorbol ester activation of PKC [23]. Subsequently, the structure of the synaptic vesicle protein p65, later known as synaptotagmin 1, revealed two C2 domain homologies responsible for Ca2+ dependent phospholipid binding and suggested that this may be a common function for C2 domains [24]. Since these early discoveries, C2 domains have been identified in well over two hundred different proteins, making them the second most common lipid binding domain [25].

C2 domains are “all-beta” structures often having three Ca2+ -binding loops that coordinate two or three Ca2+ ions via aspartic/asparagine residues. In combination with an adjacent lysine-rich polybasic cluster in a “␤-groove”, this allows binding to a range of phospholipids [26]. However, many C2 domains lack some or all of the residues required to coordinate Ca2+ , lack the polybasic ␤-groove or lack both. How these structural variations relate to the specificity of phospholipid binding is as yet unclear. In multi-C2 domain proteins, domains lacking Ca2+ coordination residues are often still able to bind phospholipids or to participate in the phospholipid interactions of adjacent domains. Examples of this are the C2B domain of E-Syt2, which enhances the Ca2+ dependent phospholipid binding of the C2A domain [1], and the C2C and C2E domains respectively of E-Syt2 and E-Syt1, which are essential for association of these proteins to the PM [1,3,6,15] (Figs. 1 and 2). The C2C domain of E-Syt2 has also been shown to participate in protein–protein interactions by recruiting the p21-GTPase Activated Kinase (PAK1), a cytoskeletal regulator, making it a dual function C2 domain [7]. Consistent with this, other C2 domains have been shown to mediate protein–protein interactions, for example, those of retinal ciliary proteins that are unable to coordinate Ca2+ or bind phospholipids [27]. The C2 domains of the E-Syts, like those of the Syts, all belong to the PKC-related S-variant or type I family, a categorization based on the topology of their ␤-strands [26]. The three E-Syts are clearly closely related in evolution and their C2 domains display an evident phylogeny, such that the C2A domains display significant homology to each other and the same is true for the C2B domains (Figs. 1 and 2). The C2C and C2D domains of mammalian E-Syt1 display strong similarity respectively with the C2A and C2B domains, suggesting that they result from an internal duplication [11]. The most C-terminal C2 domains of the three E-Syts, C2C in E-Syt2 and −3 and C2E in E-Syt1, are also clearly related and are distinct from the other C2 domains. To date only three C2 domain structures, all from E-Syt2 (Fig. 1), are available; a solution structure for the C2C domain [28], a crystal structure of the segment enclosing the C2A and C2B domains [29], and a crystal structure of the segment from the SMP domain through to the C2B domain [19]. As predicted, the C2A domain is shown to coordinate three Ca2+ ions and clearly identifies the residues involved [29] (Fig. 2). The related C2A domains of E-Syt1 and 3 and C2C domain of E-Syt1 all retain the residues shown to coordinate Ca2+ in this E-Syt2 structure [29]. The

52

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

FGFR PI(4,5)P

DAG

PI PI(4,5)P

Plasma membrane

FGF/heparin

E-Syt1

PLC IP3

Nir2/3 E-Syt2/3

VAP

Ca2+

ER membrane E-Syt1

IP3R Ca2+

Fig. 5. Model of E-Syt function in FGF and Ca2+ signaling. Activation of FGFR by FGF and heparin results in dimerization and autophosphorylation of its C-terminal kinase domain in the cytoplasm (dark green lobes). Phosphorylation of tyrosine 766 (pY766) of FGFR recruits PLC leading to its activation. PLC cleavage of PI(4,5)P releases IP3 leading to calcium release by the IP3 receptor and increase in cytosolic Ca2+ concentration. This in turn leads to tightening of ER-PM junctions, replenishment of PI(4,5)P on the PM through the exchange of PI by the Nir2/3 proteins and prolonged FGF signaling.

C2C domain of E-Syt1 was also suggested to harbor a polybasic ␤groove [3], though to a lesser degree this could also be true of the C2A domains of all three E-Syts (Fig. 2). Neither of the available E-Syt2C2B domain structures includes the potential Ca2+ coordination loops, suggesting that they are flexible and hence probably do not coordinate Ca2+ . The E-Syt2C2C domain solution structure cannot provide direct evidence for or against of Ca2+ coordination, however the residues of the potential coordination loops are visible and suggest that this domain would be unlikely to bind Ca2+ . The C2C domain does, however, have a very evident polybasic ␤-groove that is visible in the solution structure (Fig. 2). 2.4. Lipid-binding properties of the E-Syt C2 domains The first evidence that these domains interact with phospholipids showed that the C2A-C2B segment of E-Syt2 was retained on artificial liposomes of negative or neutral phospholipids in a Ca2+ -dependent manner, while the separated domains showed little retention [1]. This suggested cooperative phospholipid binding driven by the Ca2+ -dependence of C2A. Analysis of the phospholipid binding of the individual C2 domains of Xenopus E-Syt2, using a standardized overlay assay on immobilized lipids, also observed no binding of C2A to any of the 14 lipid species tested either in the presence or absence of Ca2+ [6]. But these authors did detect Ca2+ -independent binding of C2B mainly to PI 3-phosphate (PI3P) and more strongly of C2C to a range of PIs, especially PI3P, PI4 P and PI5P. The C2C domain was further shown to be essential for localization of both human and Xenopus E-Syt2 to the PM, suggesting that this was mediated by C2C binding to PM phospholipids [1,6]. Mutation of the polybasic ␤-groove of C2C was also sufficient to prevent E-Syt2 binding to the PM, as was depletion of PI 4,5-bisphosphate (PI(4,5)P) [3]. This suggests that E-Syt2 is constitutively targeted to PI(4,5)P in the PM by its C2C domain. The close homologies between the C2 domains of E-Syt2 and E-Syt3 (Fig. 2) suggest that these findings are in large part valid for both these E-Syts. In contrast, E-Syt1 association with the PM depends on an elevation in cytoplasmic Ca2+ concentration. This is mediated at least in part by the combined actions of its Ca2+ -dependent C2C and Ca2+ -independent C2E domains [15]. This suggests a functional analogy with the C2A-C2B domains of E-Syt2, Ca2+ -dependent and independent domains acting together as a calcium-dependent phospholipid binding motif [1].

2.5. PM association via combined C2 action The data suggest a regulation of E-Syt association with the PM in which the multiple C2 domains determine the tightness, dynamic and/or topology of association. E-Syt2 and 3 use their C2C domains to bind semi-constitutively to the PM, dependent on local phospholipid composition and abundance. The C2A and C2B domains then regulate the tightness and/or topology of this binding in response to cytosolic Ca2+ concentration, essentially causing a Ca2+ -dependent “zipping-up” (Fig. 3). In the case of E-Syt1, significant recruitment to the PM requires elevated Ca2+ levels and the actions of the Ca2+ dependent C2C domain in combination with the C2E domain. By analogy with C2A/C2B domains of E-Syt2 and 3, Ca2+ -dependent PM association of E-Syt1 would be expected to be aided by the C2A and C2B domains, again allowing a dynamic tightening or “zippingup” of its PM association dependent on free Ca2+ . The C-terminal C2 domain of E-Syt1 is unable of itself to direct PM association, despite being analogous to the C-terminal C2 domains of E-Syt2 and 3. This is presumably due to a weaker phospholipid interaction. Post-translational modifications and association with other proteins are also likely to be involved in modulating E-Syt functions, but these have yet to be studied in detail. Interestingly in this context, mass-spectrometry has identified a number of phosphorylation sites on the E-Syts and it has been shown that the association of E-Syt1 with E-Syt2 is strongly promoted by stimulation of cells with FGF [8,30]. 3. E-Syts and cell signaling 3.1. E-Syt functions in receptor signaling The first indication of a physiological role for the E-Syts resulted from a study of their function in the induction of mesoderm in early Xenopus embryos [6]. A yeast 2-hybrid screen, using the actin cytoskeleton regulator and Rac/Cdc42 target PAK1 as bait, identified an interaction with the last C2 domain of Xenopus E-Syt2. (It was later demonstrated that E-Syt2 recruitment of PAK1 modulated cell adhesion and wound healing in Xenopus embryos and that E-Syt2 overexpression led to disruption of cortical actin and suppression of stress fiber formation in human cells [7]). Depletion of E-Syt2 from Xenopus blastulae suppressed the induction of early mesodermal genes, and this was found to be due to an inhi-

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

bition of ERK MAP-Kinase activation by fibroblast growth factor (FGF) [6]. E-Syt2 depletion delayed the rapid phase of clathrindependent endocytosis of the FGF receptor (FGFR) that occurred immediately (∼5 min) after receptor activation. This in turn prevented the ERK activation essential for induction of the early mesodermal genes. E-Syt2 was observed to interact not only with FGFR, but also with EGFR and other RTKs, as well as with the clathrin adapter complex Adaptin-2 (AP-2). It was therefore classified as an endocytic adapter for the clathrin-dependent pathway. Human E-Syt2 was subsequently observed to transiently co-localize and associate selectively with activated FGFR, as well as to associate with very early FGFR positive endosomes following FGF stimulation [8]. Thus, both Xenopus E-Syt2 and human E-Syt2 and 3, but not E-Syt1, specifically interacted with activated FGFR during the very early stage of its endocytosis. The interaction of E-Syt2 with FGFR was found to be mediated by a direct contact of a sequence between the TM and SMP domains of E-Syt2 and a site in the upper N-terminal lobe of the FGFR kinase domain, and thus not to involve any of the previously identified E-Syt domains. E-Syt access to its binding site on FGFR was possible only in the open activated receptor conformation, explaining the specificity of the interaction. However, during FGFR endocytosis E-Syt2 remained closely associated with the PM and was never observed in later endosomes, suggesting that it remained tethered in the ER throughout [6,8]. E-Syt1 has also been implicated in the cellular response to insulin, which induces an interaction with the glucose transporter GLUT4 in the PM, but the physiological relevance of this remains unknown [5]. 3.2. E-Syts in the formation of ER-PM contact sites Most recently, the E-Syts and the Tcbs have been implicated in the formation of ER to PM contact sites, in non-vesicular lipid transport and the cellular response to store operated calcium entry (SOCE) [3,4,10]. These are all functions regulated by extracellular signaling, and hence are probably directly or indirectly related to E-Syt functions in receptor signaling. The endoplasmic reticulum (ER) is a complex network involved in vesicle-mediated anterograde and retrograde transport, protein synthesis, Ca2+ storage and lipid metabolism [31]. In addition to fusion between the ER and vesicles from the Golgi complex and endosomes involved in retrograde transport, the ER can also contact though not fuse with other cell membranes. These membrane contact sites (MCSs) can occur between the ER and mitochondria, the ER and the Golgi or between the ER and the PM [32,33]. Several tethering proteins have been implicated in the formation of ER-PM contact sites or junctions. These include the junctophilins, TMEM16 (Ist2 in yeast), VAMP-Associated Proteins (VAPs) (Scs2 and 22 in yeast) and most recently the E-Syts and Tcbs. These proteins are all anchored in the ER and have domains allowing them to bind lipids on the PM in trans [33–35]. But here the similarity ends and each appears to play a distinctly different role. The junctophilins are necessary for sarcoplasmic reticulum (SR)PM junctions in muscle cells. TMEM16/Ist2 is a member of the anoctamin family of calcium-activated chloride channels and lipid scramblases, whereas VAPs are involved in numerous interactions at ER-PM contact sites, notably with oxysterol-binding (OSBP)related protein (ORP) family members active in the regulation of PI exchange. The fourth group, the E-Syts and Tcbs, appear to function to some extent redundantly with the other tethering proteins. The first indications of this came from a study of the Tcbs [4]. Combined deletion of the six potential yeast ER-PM tethering proteins, Ist2/TMEM16, the VAP homologs Scs2/22, and all three Tcbs was necessary to eliminate ER-PM junctions. The Tcbs had previously been implicated in lipid interactions and response to Ca2+ , but their loss had no physiological consequence apart from a hypersensi-

53

tivity to cycloheximide [12]. This hypersensitivity was suppressed by expression of RSP5p, a C2-domain containing ubiquitin ligase required for fluid-phase endocytosis. Interestingly, the mammalian homolog of RSP5p is Nedd4, an ubiquitin ligase required for endocytosis of activated FGFR [36], indirectly supporting the implication of E-Syts in receptor endocytosis. Consistent with their anchorage in the ER and their ability to bind phospholipids, it was found that over-expression of the three E-Syts increased the amount of ER contacting the PM, while their knock-down reduced ER-PM contact. Knock-down of the E-Syts also suppressed Ca2+ -induced expansion of ER-PM contacts, and PI(4,5)P depletion reduced ER-PM contacts [3]. Thus, these authors suggested that the E-Syts function to tether the ER to the PM in response to elevated cytosolic Ca2+ via interactions with PM phospholipids, particularly PI(4,5)P. 3.3. E-Syt functions in lipid transport Two interrelated functions of ER-PM junctions have been studied in some detail, lipid transport and store-operated Ca2+ entry (SOCE). The lipid composition of the PM is a factor that is very important for intracellular signaling and is regulated by the cytosolic Ca2+ levels. These are in turn regulated by the activation of cell surface receptors [37]. The oxysterol-binding (OSBP)-related proteins (ORPs) are a family of eukaryotic phospholipid-binding proteins found at ER-PM contact sites [38] that were recently implicated in anterograde transport of phosphatidylserine (PS). PS is implicated in a variety of cell signaling events including the activation of protein kinase C (PKC) [39]. Human ORP5 and 8 and yeast Osh6p and 7p were shown to transport PS from the ER to the PM, and this is driven by the counter-transport of phosphatidylinositol 4-phosphate (PI4P) to the ER [40,41] (Fig. 4). Once transferred to the ER membrane, PI4 P is converted to PI by the action of the Sac1 phosphatase that is under control of yet another ORP family member, Osh3 in yeast [42,43]. In turn, PI is transported back onto the PM by the action of the PI transfer proteins (PITPs) Nir2 and 3, and this was recently shown to require the counter-transport of phosphatidic acid (PA) generated from diacylglycerol (DAG), a product of the phospholipase C (PLC) signaling pathway [44,45]. Once inserted into the PM, PI is rapidly converted to PI4P, a substrate of the ORPs, and to PI(4,5)P, whose cleavage to DAG and inositol 1,4,5-trisphosphate (IP3) by PLC leads to an enhancement of cytoplasmic Ca2+ levels via the store operated Ca2+ entry (SOCE) response. SOCE is the process by which this loss of Ca2+ from the ER store is sensed and calcium homeostasis re-established [46–48]. Cleavage of PI(4,5)P by PLC releases IP3 from the PM, and this binds the IP3 receptor (IP3R) in the ER membrane, opening its Ca2+ channel and releasing stored calcium ions into the cytosol. The concomitant reduction of Ca2+ concentration in the ER activates the SOCE response, allowing a Ca2+ influx from the extracellular space via an interaction between the Stromal Interaction Molecule 1 (STIM1) in the ER and the Orai1 channel in the PM. For both non-vesicular anterograde phospholipid transport and SOCE to function, the ER and PM membranes must be brought to within one or a few tens of nanometres of each other at junction sites [32,49]. Recent data strongly suggests that it is the capacity of the E-Syts to tether the ER to the PM and to regulate this tethering in response to changes in cytosolic Ca2+ that couples SOCE to anterograde PI transfer from the ER to the PM. Briefly, ESyt1 recruitment to ER-PM contact sites on receptor-induced Ca2+ signaling (SOCE) was found to tighten the gap between the two membranes from 20 nm in resting cells, to only 10 nm in the stimulated cells [10] (Fig. 3). This was shown to enhance the ability of the Nir2/3 transporters to transfer PI to the PM, compensating for its loss during intense receptor stimulation (Fig. 5). Thus, ESyt1 enhances the replenishment of PI(4,5)P on the PM lost during

54

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

intense receptor stimulation by tightening the ER-PM contact in response to increased cytosolic Ca2+ , and hence facilitating a return to PI(4,5)P homeostasis. This said, E-Syt2 and 3 levels have been shown to determine the extent of ER-PM junctioning and since they are also responsive to enhanced cytosolic Ca2+ , they probably play an equally important role as E-Syt1 in re-establishing PI(4,5)P homeostasis after receptor activation. 3.4. model for E-Syt function in FGF receptor signaling As mentioned, E-Syt2 was first implicated in FGF signaling and FGFR endocytosis, though this may represent a more general property in RTK signaling [6,7]. On activation, FGFR recruits PLC and stimulates cleavage of PI(4,5)P, releasing IP3 and activating Ca2+ signaling [50–53], thus initiating the process of replenishing PI(4,5)P that is stimulated by tightened E-Syt tethering at ER-PM junctions (Fig. 5). Hence, the E-Syts would naturally enhance and/or prolong signaling through the FGF pathway by maintaining PI(4,5)P homeostasis. The E-Syts display a strong preference to bind PI(4,5)P, predominantly a PM phospholipid but one that is also present on early clathrin-coated endocytic vesicles [54]. Thus, given that ESyt2 and 3 directly interact with activated FGFR independently of their ability to bind PI(4,5)P, they would act to regroup the activated receptors with PI(4,5)P at ER-PM junctions. This would have two major effects; on the one hand it would enhance the cleavage of PI(4,5)P by PLC by bringing them into close proximity, on the other hand it would link the activated receptor to the site of clathrincoated pit formation. Clathrin-coated pits have been shown to form preferentially at ER-PM contact sites [55]. The interaction of E-Syt2 with AP-2, whether direct or indirect, further suggests the validity of this scenario. Thus, we suggest that in addition to its role in phospholipid homeostasis, the combined ability of the E-Syts to interact with activated FGFR and to tether the ER to PM phospholipids would promote the formation of clathrin coated pits, subsequent receptor endocytosis, and perhaps also its recycling, e.g., Ref. [56]. 4. E-Syt functions in vivo Despite the well-defined molecular functions of the E-Syts in signaling and lipid homeostasis, information on the requirements for the E-Syts in cell and organism biology is very limited. The first indications suggested that E-Syt2 was required for the induction of mesoderm in the early Xenopus embryo [6]. Morpholino antisense depletion of E-Syt2 caused severe shortening of the larval trunk and suppressed expression of the early mesodermal marker Xbra, both characteristics of loss of signaling through the FGF pathway, reviewed in [50]. Further, these phenotypes could both be restored by ectopic E-Syt2 expression, and correlated with a failure in a rapid phase of FGFR endocytosis required for activation of the ERK pathway, a prerequisite for mesoderm induction [57,58]. We later observed that depletion of E-Syt2 from Xenopus blastocysts enhanced blastomere migration in wound healing assays, while its depletion suppressed wound healing. This correlated with GTPase Cdc42 activation of PAK1 and changes in cortical actin. Since E-Syt2 was shown to recruit PAK1, it was suggested that E-Syt2 suppressed blastomere migration by reducing activation of PAK1 and, hence, the maintenance of cortical actin structures. The E-Syt2 gene was found to be broadly expressed throughout mouse development and in the adult, while expression of the ESyt3 gene was more tissue specific [13]. Given the Xenopus data, it was therefore surprising to find that deletion of the E-Syt2 gene, and even of both the E-Syt2 and 3 genes, gave no detectable phenotype, mice developing and reproducing to all intents and purposes normally [13]. However, Esyt2−/− and Esyt2/3−/− embryonic fibroblasts (MEFs) were found to be sensitive to oxidative stress and to

display a defect in their migratory response to basic FGF stimulation. Thus, though loss of both E-Syt2 and 3 is compatible with mouse survival under laboratory conditions, subtle physiological deficiencies may yet become apparent. Even more surprising is that our ongoing studies suggest Esyt1/2/3−/− mice may also be viable (Herdman and Tremblay, unpublished data). Despite the enormous difference in organism complexity, the data from mouse parallel those for the Tcbs from yeast [4,12] and suggest very significant functional redundancy among the ER-PM tethering complexes in both higher and lower eukaryotes. However, this begs the question of why did E-Syt2-loss generate such a definitive phenotype in Xenopus. In part, the answer to this may lie in the composition [59] and the rapid expansion [60] of the cell membranes in the developing amphibian embryo, as well as the very large size of the embryonic cells [61], all of which might be expected to restrict the efficiency of receptor signaling and endocytosis. It is interesting to note that E-Syt1 has been identified as a target of the oncogenic fusion kinase CD74-ROS, commonly found in nonsmall cell lung cancer (NSCLC) [62]. Active CD74-ROS is oncogenic and causes invasiveness in NSCLC cell lines and this occurs simultaneously with E-Syt1 phosphorylation on tyrosine 993 (Y993). In E-Syt1 depleted cells, the CD74-ROS kinase was still able to cause transformation, however the invasive capacity of the cells was lost. Y993 is in the C2E domain of E-Syt1 and this site is also conserved in the C2C domains of E-Syt2 and E-Syt3. The same E-Syt1 site was found phosphorylated in numerous other cancer cell lines and NSCLC human biopsies, e.g., Ref. [63]. Thus, the post-translational regulation of the E-Syts may have significant relevance to disease development. 5. Conclusion Despite the rapidly growing understanding of the molecular and cellular functions of the E-Syts, the understanding of their in vivo functions has lagged far behind. In vivo studies have been hampered by the near complete absence of a phenotype associated with deletion of these genes. Functional links have been made with FGF and Ca2+ signaling, with the formation of ER-PM contacts, with endocytosis, and with cell migration. However, apart from the data from Xenopus and a possible role in cancer progression, no in vivo requirement for either the Tcbs or the E-Syts has as yet been discovered. This said, the evolutionary conservation of these proteins strongly suggests that they are indeed needed for organism survival. Clearly, much more work is needed before we can understand why. Acknowledgements This work was supported by operating grants from the Cancer Research Society/S. Cohen Fund and from the Natural Sciences and Engineering Research Council (NSERC) of Canada. C.H. is the recipient of a Frederick Banting and Charles Best Canada Graduate Scholarship from the Canadian Institutes of Health Research. The Research Centre of the Quebec University Hospital Centre (CR-CHU de Québec) is supported by a grant from the Fonds de recherche du Québec—Santé(FRQS). References [1] S.W. Min, W.P. Chang, T.C. Sudhof, E-Syts, a family of membranous Ca2+ -sensor proteins with multiple C2 domains, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 3823–3828. [2] N.J. Morris, S.A. Ross, J.M. Neveu, W.S. Lane, G.E. Lienhard, Cloning and preliminary characterization of a 121 k Da protein with multiple predicted C2 domains, Biochim. Biophys. Acta 1431 (1999) 525–530. [3] F. Giordano, Y. Saheki, O. Idevall-Hagren, S.F. Colombo, M. Pirruccello, I. Milosevic, E.O. Gracheva, S.N. Bagriantsev, N. Borgese, P. De Camilli, PI(4,5)P2-dependent and CaPI(4,5)P2(2+) -regulated ER-PM interactions

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16] [17]

[18]

[19]

[20]

[21] [22]

[23] [24]

[25] [26]

[27]

[28]

mediated by the extended synaptotagmins, Cell 153 (2013) 1494–1509, http://dx.doi.org/10.1016/j.cell.2013.05.026. A.G. Manford, C.J. Stefan, H.L. Yuan, J.A. Macgurn, S.D. Emr, ER-to-plasma membrane tethering proteins regulate cell signaling and ER morphology, Dev. Cell 23 (2012) 1129–1140, http://dx.doi.org/10.1016/j.devcel.2012.11.004. V. Lalioti, G. Muruais, A. Dinarina, J. van Damme, J. Vandekerckhove, I.V. Sandoval, The atypical kinase Cdk5 is activated by insulin, regulates the association between GLUT4 and E-Syt1, and modulates glucose transport in 3T3-L1 adipocytes, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 4249–4253, http:// dx.doi.org/10.1073/pnas.0900218106. S. Jean, A. Mikryukov, M.G. Tremblay, J. Baril, F. Guillou, S. Bellenfant, T. Moss, Extended-synaptotagmin-2 mediates FGF receptor endocytosis and ERK activation in vivo, Dev. Cell (2010), http://dx.doi.org/10.1016/j.devcel.2010. 08.007, S1534-5807(10) 00382-5 [pii]. S. Jean, M.G. Tremblay, C. Herdman, F. Guillou, T. Moss, The endocytic adapter E-Syt2 recruits the p21 GTPase activated kinase PAK1 to mediate actin dynamics and FGF signalling, Biol. Open 1 (2012) 731–738, http://dx.doi.org/ 10.1242/bio.2012968. M.G. Tremblay, C. Herdman, F. Guillou, P.K. Mishra, J. Baril, S. Bellenfant, T. Moss, Extended synaptotagmin interaction with the fibroblast growth factor receptor depends on receptor conformation, not catalytic activity, J. Biol. Chem. 290 (2015) 16142–16156, http://dx.doi.org/10.1074/jbc.M115.656918. J. Liou, C.L. Chang, Unveiling physiological functions of extended synaptotagmins, Cell Cycle 14 (2015) 799–800, http://dx.doi.org/10.1080/ 15384101.2015.1010951. C.L. Chang, T.S. Hsieh, T.T. Yang, K.G. Rothberg, D.B. Azizoglu, E. Volk, J.C. Liao, J. Liou, Feedback regulation of receptor-induced Ca(2+) signaling mediated by E-Syt1 and Nir2 at endoplasmic reticulum–plasma membrane junctions, Cell Rep. 5 (2013) 813–825, http://dx.doi.org/10.1016/j.celrep.2013.09.038. M. Craxton, A manual collection of syt, esyt, rph3a, rph3al, doc2, and dblc2 genes from 46 metazoan genomes—an open access resource for neuroscience and evolutionary biology, BMC Genomics 11 (2010) 37, http://dx.doi.org/10. 1186/1471-2164-11-37, 1471-2164-11-37 [pii]. C.E. Creutz, S.L. Snyder, T.A. Schulz, Characterization of the yeast tricalbins: membrane-bound multi-c2-domain proteins that form complexes involved in membrane trafficking, Cell. Mol. Life Sci. 61 (2004) 1208–1220. C. Herdman, M.G. Tremblay, P.K. Mishra, T. Moss, Loss of extended synaptotagmins esyt2 and esyt3 does not affect mouse development or viability, but in vitro cell migration and survival under stress are affected, Cell Cycle 13 (2014) 2616–2625, http://dx.doi.org/10.4161/15384101.2014. 943573. R. Fernandez-Busnadiego, Y. Saheki, P. De Camilli, Three-dimensional architecture of extended synaptotagmin-mediated endoplasmic reticulum-plasma membrane contact sites, in: Proc, Natl Acad Sci U. S. A. (2015), http://dx.doi.org/10.1073/pnas.1503191112. O. Idevall-Hagren, A. Lu, B. Xie, P. De Camilli, Triggered Ca2+ influx is required for extended synaptotagmin 1-induced ER-plasma membrane tethering, EMBO J. 34 (2015) 2291–2305, http://dx.doi.org/10.15252/embj.201591565. T.C. Südhof, Synaptotagmins: why so many? J. Biol. Chem. 277 (2002) 7629–7632, http://dx.doi.org/10.1074/jbc.R100052200. E.R. Chapman, How does synaptotagmin trigger neurotransmitter release? Annu. Rev. Biochem 77 (2008) 615–641, http://dx.doi.org/10.1146/annurev. biochem.77.062005.101135. G.K. Voeltz, W.A. Prinz, Y. Shibata, J.M. Rist, T.A. Rapoport, A class of membrane proteins shaping the tubular endoplasmic reticulum, Cell 124 (2006) 573–586, http://dx.doi.org/10.1016/j.cell.2005.11.047. C.M. Schauder, X. Wu, Y. Saheki, P. Narayanaswamy, F. Torta, M.R. Wenk, P. De Camilli, K.M. Reinisch, Structure of a lipid-bound extended synaptotagmin indicates a role in lipid transfer, Nature 510 (2014) 552–555, http://dx.doi. org/10.1038/nature13269. K.O. Kopec, V. Alva, A.N. Lupas, Homology of SMP domains to the TULIP superfamily of lipid-binding proteins provides a structural basis for lipid exchange between ER and mitochondria, Bioinformatics 26 (2010) 1927–1931, http://dx.doi.org/10.1093/bioinformatics/btq326. I. Lee, W. Hong, Diverse membrane-associated proteins contain a novel SMP domain, FASEB J. 20 (2006) 202–206. U. Kikkawa, A. Kishimoto, Y. Nishizuka, The protein kinase C family: heterogeneity and its implications, Annu. Rev. Biochem. 58 (1989) 31–44, http://dx.doi.org/10.1146/annurev.bi.58.070189.000335. D.J. Burns, R.M. Bell, Protein kinase C contains two phorbol ester binding domains, J. Biol. Chem. 266 (1991) 18330–18338. M.S. Perin, V.A. Fried, G.A. Mignery, R. Jahn, T.C. Sudhof, Phospholipid binding by a synaptic vesicle protein homologous to the regulatory region of protein kinase C, Nature 345 (1990) 260–263, http://dx.doi.org/10.1038/345260a0. R.V. Stahelin, Lipid binding domains: more than simple lipid effectors, J. Lipid Res. 50 (2009) S299-–S304, http://dx.doi.org/10.1194/jlr.R800078-JLR200. S. Corbalan-Garcia, J.C. Gomez-Fernandez, Signaling through C2 domains: more than one lipid target, Biochim. Biophys. Acta 1838 (2014) 1536–1547, http://dx.doi.org/10.1016/j.bbamem.2014.01.008. K. Remans, M. Burger, I.R. Vetter, A. Wittinghofer, C2 domains as protein–protein interaction modules in the ciliary transition zone, Cell Rep. 8 (2014) 1–9, http://dx.doi.org/10.1016/j.celrep.2014.05.049. T. Nagashima, F. Hayashi, S. Yokoyama, Solution structure of the third C2 domain of KIAA1228 protein, 2009.

55

[29] J. Xu, T. Bacaj, A. Zhou, D.R. Tomchick, T.C. Sudhof, J. Rizo, Structure and Ca(2+) -binding properties of the tandem C(2) domains of E-Syt2, Structure 22 (2014) 269–280, http://dx.doi.org/10.1016/j.str.2013.11.011. [30] P.V. Hornbeck, J.M. Kornhauser, S. Tkachev, B. Zhang, E. Skrzypek, B. Murray, V. Latham, M. Sullivan, Phosphositeplus: a comprehensive resource for investigating the structure and function of experimentally determined post-translational modifications in man and mouse, Nucleic Acids Res. 40 (2012) D261–D270, http://dx.doi.org/10.1093/nar/gkr1122. [31] A.R. English, G.K. Voeltz, Endoplasmic reticulum structure and interconnections with other organelles, Cold Spring Harb. Perspect. Biol. 5 (2013) a013227, http://dx.doi.org/10.1101/cshperspect.a013227. [32] S. Lahiri, A. Toulmay, W.A. Prinz, Membrane contact sites, gateways for lipid homeostasis, Curr. Opin. Cell Biol. 33 (2015) 82–87, http://dx.doi.org/10.1016/ j.ceb.2014.12.004. [33] W.M. Henne, J. Liou, S.D. Emr, Molecular mechanisms of inter-organelle ER-PM contact sites, Curr. Opin. Cell Biol. 35 (2015) 123–130, http://dx.doi. org/10.1016/j.ceb.2015.05.001. [34] C.J. Stefan, A.G. Manford, S.D. Emr, ER-PM connections: sites of information transfer and inter-organelle communication, Curr. Opin. Cell Biol. 25 (2013) 434–442, http://dx.doi.org/10.1016/j.ceb.2013.02.020. [35] K. Kunzelmann, TMEM16, LRRC8A, bestrophin: chloride channels controlled by Ca(2+) and cell volume, Trends Biochem. Sci. 40 (2015) 535–543, http://dx. doi.org/10.1016/j.tibs.2015.07.005. [36] A. Persaud, P. Alberts, M. Hayes, S. Guettler, I. Clarke, F. Sicheri, P. Dirks, B. Ciruna, D. Rotin, Nedd4-1 binds and ubiquitylates activated FGFR1 to control its endocytosis and function, EMBO J. 30 (2011) 3259–3273, http://dx.doi.org/ 10.1038/emboj.2011.234. [37] R.H. Michell, Inositol lipids in cellular signalling mechanisms, Trends Biochem. Sci. 17 (1992) 274–276. [38] V.M. Olkkonen, S. Li, Oxysterol-binding proteins: sterol and phosphoinositide sensors coordinating transport, signaling and metabolism, Prog. Lipid Res. 52 (2013) 529–538, http://dx.doi.org/10.1016/j.plipres.2013.06.004. [39] J.E. Vance, G. Tasseva, Formation and function of phosphatidylserine and phosphatidylethanolamine in mammalian cells, Biochim. Biophys. Acta (BBA): Mol. Cell Biol. Lipids 1831 (2013) 543–554, http://dx.doi.org/10.1016/j. bbalip.2012.08.016. [40] J. Chung, F. Torta, K. Masai, L. Lucast, H. Czapla, L.B. Tanner, P. Narayanaswamy, M.R. Wenk, F. Nakatsu, P. De Camilli, PI4P/phosphatidylserine countertransport at ORP5- and ORP8-mediated ER-plasma membrane contacts, Science (New York, NY) 349 (2015) 428–432, http://dx.doi.org/10.1126/science.aab1370. [41] J. Moser von Filseck, A. Copic, V. Delfosse, S. Vanni, C.L. Jackson, W. Bourguet, G. Drin, Intracellular transport. Phosphatidylserine transport by ORP/Osh proteins is driven by phosphatidylinositol 4-phosphate, Science (2015) 432–436, http://dx.doi.org/10.1126/science.aab1346. [42] A. Roy, T.P. Levine, Multiple pools of phosphatidylinositol 4-phosphate detected using the pleckstrin homology domain of Osh2p, J. Biol. Chem. 279 (2004) 44683–44689, http://dx.doi.org/10.1074/jbc.M401583200. [43] C.J. Stefan, A.G. Manford, D. Baird, J. Yamada-Hanff, Y. Mao, S.D. Emr, Osh proteins regulate phosphoinositide metabolism at ER-plasma membrane contact sites, Cell 144 (2011) 389–401, http://dx.doi.org/10.1016/j.cell.2010. 12.034. [44] Y.J. Kim, M.L. Guzman-Hernandez, E. Wisniewski, T. Balla, Phosphatidylinositol–phosphatidic acid exchange by Nir2 at ER-PM contact sites maintains phosphoinositide signaling competence, Dev. Cell 33 (2015) 549–561, http://dx.doi.org/10.1016/j.devcel.2015.04.028. [45] C.L. Chang, J. Liou, Phosphatidylinositol 4,5-bisphosphate homeostasis regulated by Nir2 and Nir3 proteins at endoplasmic reticulum-plasma membrane junctions, J. Biol. Chem. 290 (2015) 14289–14301, http://dx.doi. org/10.1074/jbc.M114.621375. [46] J.W. Putney, Capacitative calcium entry: from concept to molecules, Immunol. Rev. 231 (2009) 10–22, http://dx.doi.org/10.1111/j.1600-065X.2009.00810.x. [47] S. Feske, Calcium signalling in lymphocyte activation and disease, Nat. Rev. Immunol. 7 (2007) 690–702 http://www.nature.com/nri/journal/v7/n9/ suppinfo/nri2152 S1.html. [48] O. Brandman, J. Liou, W.S. Park, T. Meyer, STIM2 is a feedback regulator that stabilizes basal cytosolic and endoplasmic reticulum Ca2+ levels, Cell 131 (2007) 1327–1339, http://dx.doi.org/10.1016/j.cell.2007.11.039. [49] S.C. Helle, G. Kanfer, K. Kolar, A. Lang, A.H. Michel, B. Kornmann, Organization and function of membrane contact sites, Biochim. Biophys. Acta 1833 (2013) 2526–2541, http://dx.doi.org/10.1016/j.bbamcr.2013.01.028. [50] M.E. Pownall, H.V. Isaacs, FGF signalling in vertebrate development, San Rafael (CA) 2010. [51] D.M. Ornitz, N. Itoh, The fibroblast growth factor signaling pathway, Wiley Interdiscip. Rev.: Dev. Biol. 4 (2015) 215–266, http://dx.doi.org/10.1002/ wdev.176. [52] M.A. Lemmon, J. Schlessinger, Cell signaling by receptor-tyrosine kinases, Cell 141 (2010) 1117–1134, http://dx.doi.org/10.1016/j.cell.2010.06.011. [53] B. Thisse, C. Thisse, Functions and regulations of fibroblast growth factor signaling during embryonic development, Dev. Biol. 287 (2005) 390–402. [54] Y. Posor, M. Eichhorn-Grunig, V. Haucke, Phosphoinositides in endocytosis, Biochim. Biophys. Acta 1851 (2015) 794–804, http://dx.doi.org/10.1016/j. bbalip.2014.09.014. [55] P.D. Fox, C.J. Haberkorn, A.V. Weigel, J.L. Higgins, E.J. Akin, M.J. Kennedy, D. Krapf, M.M. Tamkun, Plasma membrane domains enriched in cortical

56

[56]

[57]

[58]

[59] [60]

C. Herdman, T. Moss / Pharmacological Research 107 (2016) 48–56 endoplasmic reticulum function as membrane protein trafficking hubs, Mol. Biol. Cell 24 (2013) 2703–2713, http://dx.doi.org/10.1091/mbc.E12-12-0895. E.R. Eden, I.J. White, A. Tsapara, C.E. Futter, Membrane contacts between endosomes and er provide sites for PTP1B-epidermal growth factor receptor interaction, Nat. Cell Biol. 12 (2010) 267–272, http://dx.doi.org/10.1038/ ncb2026. E. Delaune, P. Lemaire, L. Kodjabachian, Neural induction in Xenopus requires early FGF signalling in addition to BMP inhibition, Development 132 (2005) 299–310, http://dx.doi.org/10.1242/dev.01582, dev.01582 [pii]. R.B. Fletcher, R.M. Harland, The role of FGF signaling in the establishment and maintenance of mesodermal gene expression in Xenopus, Dev. Dyn. 237 (2008) 1243–1254, http://dx.doi.org/10.1002/dvdy.21517. M. Mes-Hartree, J.B. Armstrong, Lipid composition of developing Xenopus laevis embryos, Can. J. Biochem. 54 (1976) 578–582. M.V. Danilchik, E.E. Brown, Membrane dynamics of cleavage furrow closure in Xenopus laevis, Dev. Dyn. 237 (2008) 565–579, http://dx.doi.org/10.1002/ dvdy.21442.

[61] P. Hausen, M. Riebesell, The early development of Xenopus laevis: an atlas of the histology, Verlag Z. Naturforschung (1991). [62] H.J. Jun, H. Johnson, R.T. Bronson, S. de Feraudy, F. White, A. Charest, The oncogenic lung cancer fusion kinase CD74-ROS activates a novel invasiveness pathway through E-Syt1 phosphorylation, Cancer Res. 72 (2012) 3764–3774, http://dx.doi.org/10.1158/0008-5472.can-11-3990. [63] M.L. Knowlton, L.M. Selfors, C.N. Wrobel, T.L. Gu, B.A. Ballif, S.P. Gygi, R. Polakiewicz, J.S. Brugge, Profiling Y561-dependent and -independent substrates of CSF-1R in epithelial cells, PLoS One 5 (2010) e13587, http://dx. doi.org/10.1371/journal.pone.0013587. [64] F. Sievers, A. Wilm, D. Dineen, T.J. Gibson, K. Karplus, W. Li, R. Lopez, H. McWilliam, M. Remmert, J. Soding, J.D. Thompson, D.G. Higgins, Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega, Mol. Syst. Biol. 7 (2011) 539, http://dx.doi.org/10.1038/msb.2011.75.