Download Lipid transfer and metabolism across the endolysosomal

Biochimica et Biophysica Acta 1861 (2016) 880–894
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbalip
Lipid transfer and metabolism across the endolysosomal–
mitochondrial boundary☆
Tiziana Daniele a,b, Maria Vittoria Schiaffino c,⁎
a
b
c
Experimental Imaging Center, IRCCS San Raffaele Scientific Institute, Milan, Italy
Department of Experimental Medicine, University of Genoa, Genoa, Italy
Division of Genetics and Cell Biology, IRCCS San Raffaele Scientific Institute, Milan, Italy
a r t i c l e
i n f o
Article history:
Received 14 November 2015
Received in revised form 30 January 2016
Accepted 3 February 2016
Available online 4 February 2016
Keywords:
Lysosomes
Melanosomes
Mitochondria
Autophagosomes
Lipid droplets
Membrane contact sites
a b s t r a c t
Lysosomes and mitochondria occupy a central stage in the maintenance of cellular homeostasis, by playing complementary roles in nutrient sensing and energy metabolism. Specifically, these organelles function as signaling hubs
that integrate environmental and endogenous stimuli with specific metabolic responses. In particular, they control
various lipid biosynthetic and degradative pipelines, either directly or indirectly, by regulating major cellular metabolic pathways, and by physical and functional connections established with each other and with other organelles.
Membrane contact sites allow the exchange of ions and molecules between organelles, even without membrane fusion, and are privileged routes for lipid transfer among different membrane compartments. These inter-organellar
connections typically involve the endoplasmic reticulum. Direct membrane contacts have now been described
also between lysosomes, autophagosomes, lipid droplets, and mitochondria. This review focuses on these recently
identified membrane contact sites, and on their role in lipid biosynthesis, exchange, turnover and catabolism. This
article is part of a Special Issue entitled: The cellular lipid landscape edited by Tim P. Levine and Anant K. Menon.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
Cellular metabolism is tightly regulated and compartmentalized in
space and time within distinct subcellular organelles. Lysosomes and
mitochondria occupy a central stage in these processes. Indeed, they
play overlapping and complementary roles in nutrient sensing and energy metabolism, and operate as dynamic signaling hubs, which
Abbreviations: AMPK, AMP-activated protein kinase; ATG, autophagy-related; CL,
cardiolipin; EM, electron microscopy; EMC, ER membrane protein complex; ER, endoplasmic reticulum; ERMES, ER-mitochondria encounter structure; FFAT, two phenylalanines in
an acid tract; HOPS, homotypic fusion and vacuole protein sorting; LD, lipid droplet; LDL,
low-density lipoprotein; LE, late endosome; LRO, lysosome-related organelle; LTP, lipid
transfer protein; MAM, mitochondria-associated membrane; MCS, membrane contact
site; Mfn2, Mitofusin 2; NVJ, nuclear-vacuolar junction; ORD/ORP, oxysterol-binding protein (OSBP)-related domain/protein; Osh, OSBP homolog (in yeast); PA, phosphatidic acid;
PAS, pre-autophagosomal structure; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PGC1α, peroxisome proliferator-activated receptor-γ co-activator 1α; PH,
pleckstrin homology; PI, phosphatidylinositol; PI4P, phosphatidylinositol-4-phosphate;
PMN, piecemeal microautophagy of the nucleus; PS, phosphatidylserine; SMP,
synaptotagmin-like mitochondrial lipid-binding protein; START, steroidogenic acute regulatory protein (StAR)-related lipid transfer (START); TORC1, target of rapamycin complex
1; VAP, vesicle-associated membrane protein (VAMP)-associated protein; VAST, VAD1
Analog of StAR-related lipid transfer; vCLAMP, vacuole and mitochondria patch.
☆ This article is part of a Special Issue entitled:The cellular lipid landscape edited by
Tim P. Levine and Anant K. Menon.
⁎ Corresponding author at: IRCCS San Raffaele Scientific Institute, Via Olgettina 58,
20132 Milan, Italy.
E-mail address: schiaffi[email protected] (M.V. Schiaffino).
http://dx.doi.org/10.1016/j.bbalip.2016.02.001
1388-1981/© 2016 Elsevier B.V. All rights reserved.
generate coordinated metabolic responses to environmental and endogenous cues (Fig. 1A). Lysosomes act as sensors of amino acid, glucose
and growth factors, by directly controlling the activity of the nutrientsensing mammalian target of rapamycin complex 1 (mTORC1) [1].
They also impact energy metabolism, thanks to their general degradative
functions, which include key direct and indirect roles in multiple lipid catabolic pathways. Indeed, lysosomes execute the hydrolysis of neutral
lipids and lipophagy, and stimulate the β-oxidation of fatty acids in mitochondria and peroxisomes, by way of the transcription factor TFEB and its
target genes [2]. Mitochondria act as sensors of Ca2+ signaling, oxidative
stress and nutrient availability. In addition, they impact energy metabolism by the progressive oxidation of substrates – chiefly fatty acids during
starvation – with ensuing generation of ATP, and in turn direct control of
the energy-sensing AMP-activated protein kinase (AMPK) [3]. These integrated processes require reciprocal physical and functional interaction
among lysosomes, mitochondria, and additional organelles implicated in
lipid biosynthesis, recycling and catabolism, including endoplasmic reticulum (ER), peroxisomes, autophagosomes and lipid droplets (LDs).
The synthesis of cellular lipids takes place primarily in the ER. Specific biosynthetic reactions, however, occur in other organelles, such as the
Golgi apparatus and mitochondria [4,5]. Thus, to appropriately distribute and exchange lipids among the various subcellular compartments,
both vesicular and non-vesicular transport mechanisms are necessary.
Non-vesicular transport is essential for organelles that are not directly
connected to the secretory pathway, such as mitochondria. Furthermore, thanks to its fast mode of action, it can play a corrective role on
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
881
Fig. 1. Major metabolic interactions between lysosomes and mitochondria. (A) Lysosomes and mitochondria share overlapping and complementary roles in cell metabolism, including
nutrient sensing, lipid degradation, energy metabolism and inverse regulation of mTORC1 and AMPK signaling in response to nutrient and energy availability. A more specific list of
organelle functions is provided below. (B) Schematic representation of the inverse regulation of mTORC1 and AMPK. mTORC1 is a primary sensor of nutrient and growth factor
availability, promoting anabolic programs and cell proliferation. AMPK is a primary energy sensor, promoting cell catabolism in response to increased AMP/ATP ratio. These two
kinases are inversely regulated by nutrient availability and ATP levels. They also regulate each other by multiple mechanisms involving lysosomes and mitochondria. (1) In fed
conditions, the Ragulator–Rag complex recruits mTORC1 to the lysosomal surface, where it is activated by Rheb [21]. (2) The lysosomal localization and obligate dimerization of
mTORC1, required for its activation, are also controlled by the ATP-dependent TTT-RUVBL1/2 complex, independently on AMPK [48]. (3) During energy stress conditions, AMPK is
allosterically stimulated by AMP. (4) AMPK activation also requires interaction with the v-ATPase–Ragulator complex on the lysosomal membrane [22]. In this system, Ragulator
appears to work as an alternative adaptor for either mTORC1 or AMPK on the lysosomal membranes, and hence as switch between anabolism and catabolism. Activated AMPK in turn
inhibits mTORC1 by phosphorylating TSC2 (tuberous sclerosis complex 2) [46] and Raptor (regulatory-associated protein of mTOR, complex 1) [47]. Thus, ATP production by
mitochondria impacts on mTORC1 function by multiple AMPK-dependent and independent mechanisms. Here, we specifically focused on the mammalian system. Several aspects of
TORC1 regulation and signaling, including lysosomal (vacuole) localization, sensitivity to amino acids and cross-talk with AMPK, are also conserved in yeast.
the bulk transport operated by vesicular traffic [6]. In recent years, a growing number of membrane contact sites (MCSs) between different organelles has been identified, revealing novel important mechanisms of
inter-organelle communication and exchange [7,8]. MCSs are regions
where the membranes of two organelles become closely juxtaposed
(10–30 nm) without fusing. These structures are formed by a variety
of multiprotein tethering complexes, which consist of both integral
and membrane associated proteins, as well as specific lipids such as
phosphoinositides. Physical inter-organellar connections are involved in
multiple fundamental processes, including Ca2+ signaling, organelle dynamics and maturation, and lipid transfer and metabolism. MCSs between
ER and Golgi, ER and mitochondria, and ER and plasma membrane have
been unambiguously established as a critical way to transfer lipids between different membrane compartments, by means of lipid transfer proteins (LTPs). Similar structural and functional connections are now
beginning to emerge among lysosomes, mitochondria, and functionally
related organelles. In most cases, however, the detailed molecular mechanisms are still unknown. This review focuses on the recently identified
inter-organelle connections, implicated in lipid exchange, turnover and
catabolism, between lysosomes, autophagosomes, lipid droplets, and mitochondria (Fig. 2). For lipid exchange at ER–endosome and lysosome–
peroxisome MCSs, the reader is invited to refer to recent comprehensive
reviews (see this issue and Refs. [9,10]).
For the sake of clarity and completeness, we will start with a general
overview of lysosomal and mitochondrial functions, with particular attention to their roles in metabolic homeostasis. Moreover, we have included four explanatory Boxes, dedicated to LTP families operating at
MCSs discussed in this review, containing conserved ORD (Box 1),
START (Box 2A), START-like/VASt (Box 2B), and SMP (Box 3) lipidbinding domains. In fact, these protein families are emerging as recurrent tethers and/or lipid sensors or transporters at multiple interorganelle connections, thus representing potential candidates for membrane contacts that at the moment are little characterized. The ORD,
START and SMP families share similar domain architectures. In addition
to the core lipid-binding domain, they contain other lipid-binding modules, allowing them to interact with proteins or lipids on opposing
membranes of MCSs. These modules include (i) transmembrane domains, typically responsible for ER-anchoring; (ii) the FFAT motif (two
phenylalanines in an acid tract), which interacts with ER-resident
vesicle-associated membrane protein (VAMP)-associated proteins
(VAPs); (iii) pleckstrin homology (PH) or C2 domains, which interact
with phosphoinositides. Other protein families involved in lipid transport at contact sites, not specifically discussed in this review, include
the phosphatidylinositol transfer proteins (PITP) and the glycolipid
transfer proteins (GLTP). For information about these LTP families, the
reader is invited to refer to recent reviews [11,12]. Finally, we will
882
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
Fig. 2. Lipid flux among lysosomes, mitochondria, endoplasmic reticulum, lipid droplets
and phagophores. Main biosynthetic, metabolic and degradative lipid routes involving
membrane contact sites and other mechanisms are schematically illustrated. (1) In
yeast, lysosomes/vacuoles and mitochondria exchange lipids, by means of a MCS
(vCLAMP) that can partly compensate for ERMES. In mammals, a lipid transfer role may
be played by the MCSs between LROs (melanosomes) and mitochondria. (2) Lysosomes
receive lipids from the ER via the secretory pathway, and may exchange lipids with the
ER directly by MCSs, or indirectly via intermediate steps (peroxisomes? cytosolic LTPs?).
(3) Mitochondria bidirectionally exchange lipids with the ER by MCSs. The ER also
supplies lipids (4) to phagophores at MAMs, and (5) to LDs by membrane continuity.
(6) LDs receive lipids derived from LDLs or autophagy from lysosomes, as free fatty
acids or by means of cytosolic LTPs. In turn, LDs supply lipids (7) for mitochondrial
respiration and (8) for phagophore biogenesis, presumably through MCSs. Finally,
during their formation, the phagophores receive lipids (4) from ER and mitochondria at
MAMs, and (8) possibly directly from LDs. (9) Upon autophagosome maturation,
phagophores convey their content, including other engulfed organelles, to lysosomes by
membrane fusion. Except for vCLAMP (1), the lipid fluxes described in the scheme
illustrated here occur in the mammalian system. In most cases, however, similar
pipelines, including those mediated by ER–mitochondria or ER–LD connections, have
been described in yeast.
consider that late endosomes (LEs) and lysosomes, distinguished by
their physical properties and ultrastructure, share functional features
and protein constituents, including commonly used markers such as
Lamp-1 or Rab7 [13]. For the purpose of this review, unless otherwise
specified, we will jointly refer to them as LEs/lysosomes, or simply
lysosomes.
2. Lysosomes: functions and membrane contacts with the ER
Lysosomes (vacuoles in yeast) represent the most important catabolic compartment of eukaryotic cells, containing a wide set of acidic
hydrolases able to digest not only macromolecules (such as proteins,
lipids, sugars, nucleic acids), but also entire organelles and pathogenic
organisms [14,15]. A continuous flux of membrane and protein constituents reaches the lysosomes through the secretory and endocytic
pathways, while extracellular and intracellular substrates destined to
degradation and reprocessing are delivered to the organelles mainly
by endocytosis and autophagy, respectively. Lysosomes are also
implicated in intracellular signaling, since they are considered acidic,
non-canonical-calcium stores. Upon different stimuli, lysosomes have
been shown to release Ca2 + by means of Ca2 + permeable channels,
such as mucolipins (TRPML) and two-pore channels (TPC) [16]. This
in turn can evoke and/or amplify ER-dependent Ca2+ release at MCSs
between the two organelles, modulating Ca2+ fluxes and/or signaling
[17,18]. Furthermore, not only secretory lysosomes, specialized organelles of hematopoietic cells, but also conventional lysosomes in many
cell types are sensitive to elevation of intracellular Ca2+, and can be secreted to promote repair of plasma membrane wounds [19,20].
Recently, lysosomes have been recognized as key signaling hubs
necessary on the one hand for mTORC1 activation by nutrients and
growth factors [1,21], and on the other for AMPK activation in response
to glucose starvation. The mechanism responsible for the inverse regulation of these two kinases appears to be their reciprocally exclusive
binding to the same adaptor/activator complex (Ragulator) on the lysosomal membranes [22] (Fig. 1B). Furthermore, lysosomes were shown
to regulate their own biogenesis and function, autophagy and lipid catabolism by β-oxidation in mitochondria and peroxisomes, by means
of an autoregulatory loop generated by the transcription factor TFEB.
In normal fed conditions, mTORC1 activation on lysosomes phosphorylates and inactivates TFEB [23]. Vice versa upon starvation, lysosomal
Ca2 + release through mucolipin 1 (TRPML1) generates an elevated
Ca2+ microdomain, responsible for the activation of the protein phosphatase calcineurin, which in turn dephosphorylates TFEB and promotes its nuclear translocation [24]. Once in nucleus, TFEB stimulates
its own transcription, as well as the expression of target genes, including
the peroxisome proliferator-activated receptor-α (PPARα) and the
PPARγ co-activator 1α (PGC1α) [2]. It is therefore not surprising that,
in addition to classical lysosomal storage diseases, disfuction of lysosomal–autophagic pathways has been associated to a variety of other
disease conditions, including metabolic, infectious, immune and common neurodegenerative disorders (i.e. Parkinson, Alzheimer, and Huntington disease) [25].
The ER forms MCSs with the majority of membrane compartments,
including LEs and lysosomes [8]. In mammals, contacts between ER
and LEs/lysosomes are implicated in the regulation of signaling processes [26], in organelle maturation and dynamics [27,28] and possibly also
in lipid exchange. In fact, LEs/lysosomes represent the main source of
extracellular-derived cholesterol (mainly in the form of low-density lipoproteins, LDLs), which needs to be transferred to other compartments
for recycling, or storage. Consistently, lipid-binding proteins of the ORD
and START families (see Boxes 1 and 2A) have been associated to MCSs
between ER and LEs/lysosomes [28–31]. However, the roles of these
LTPs remain controversial, since they appear to function as lipid sensors
and molecular tethers, rather than lipid transporters, coordinating
membrane lipid content with organelle dynamics. The Rab7 effector
and cholesterol sensor ORPL1 functions on LEs to promote dynein activity. On the contrary, in low cholesterol conditions, it undergoes a conformational change that induces ER–LE MCSs, with consequent
displacement of the dynein–dynactin complex and peripheral distribution of the organelles [28,32]. The two sterol-binding proteins STARD3
and STARD3NL are known to participate in the MCS between LEs and
the ER, and to regulate endosome morphology and dynamics. Thanks
to cooperation between their MENTAL and START domains, they may
also possibly contribute to cholesterol export from late endosomal compartments towards the ER [30,33,34]. Finally, the ER-anchored protein
ORP5 was shown to interact with NPC1 and to be involved in cholesterol
exit from endosomes/lysosomes, since its knock-down caused cholesterol accumulation in the lysosomal membrane [29]. However, the
latter phenotype might represent an indirect effect, since recent
findings indicated that ORP5 (and its yeast homolog Osh6) does not
bind cholesterol in appreciable amounts and it is rather implicated in
phosphatidylserine (PS)-phosphatidylinositol-4-phosphate (PI4P)
counter-transport at the ER–plasma membrane boundary [35,36].
In yeast, the connection between the nuclear ER and the vacuole
(nuclear–vacuolar junction, NVJ) probably represents the simplest
MCS, formed by only two tethering proteins, Nvj1 on the nuclear membrane and Vac8 on the vacuole membrane [37]. The function of this contact mainly concerns the piecemeal microautophagy of the nucleus
(PMN) during nutrient starvation, although it is also implicated in
sphingolipid biosynthesis and may represent a “backup” mechanism,
an alternative route for lipid transport and homeostasis (see below). Interestingly, among the proteins recruited at NVJ and involved in the
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
883
Box 1
Protein families involved in lipid transport at membrane contacts: architecture of ORD containing proteins.
The founder member of this family is the Oxysterol-binding protein (OSBP). OSBP and OSBP-related proteins (ORPs) in mammals, designated
OSBP homolog (Osh) proteins in yeast, share a structural feature, namely the OSBP-related domain (ORD) at their C-terminus, encompassing
about 400 amino acids. This domain is built around a central 19-strand antiparallel β-sheet, which folds into a β barrel-like structure with a lipid-accommodating pocket, covered by a flexible α-helical lid, which shields the bound sterol from the aqueous phase [124,125]. ORPs are widespread among eukaryotic species and have been shown to sense, bind and actively counter-transport lipids across organelles at MCSs, in
particular sterols or PS with PI4P, thus controlling membrane lipid composition and metabolism [35,36,126–129].
Proteins belonging to the ORP family contain multiple domains for membrane targeting, including transmembrane (TM), PH and FFAT domains,
and can be subdivided into different subgroups according to their architecture, as follows.
1) Osh1, Osh2 and ORP1L comprise ankyrin repeats at their N-termini, which have been reported to mediate protein–protein interactions.
2) Osh3, OSBP, ORP3a, ORP4a, ORP6a, ORP7, and ORP9f contain both a PH domain, responsible for the interaction with phosphoinositides
(such as PI4P at the Golgi or plasma membrane), and a FFAT motif, which is recognized by VAPs (ER integral membrane proteins). Osh3
also contains a Golgi localization domain (GOLD).
3) ORP5a and ORP8a host both a PH domain for PI4P binding at the plasma membrane, and an ER-anchoring hydrophobic tail sequence at
their C-terminus.
4) ORP10 and ORP11 display only a PH domain, for phosphoinositide binding.
5) ORP2, and ORP9a contain a FFAT motif, which is recognized by VAPs (ER integral membrane proteins).
6) Osh4, Osh5, Osh6, Osh7, ORP1S and ORP4g display no additional motifs to the ORD.
Osh1 and ORP1L localize at ER-lysosome/vacuole MCSs, Osh4 and OSBP localize at ER-Golgi MCSs, and Osh3, Osh6, ORP5a and ORP8a localize
at ER-plasma membrane MCSs. Amino acid length of individual proteins is given in brackets. Since many isoforms, resulting from different
transcripts variants, are known for some of these proteins, when applicable the name of the specific isoforms (such as “S” or “L”, “a”, “g”, or
“f”) is indicated.
PMN, there are lipid modifying enzymes and putative LTPs. These include the enoyl-CoA reductase Tsc13, required for the biosynthesis of
very-long-chain fatty acids [38], together with the ORP Osh1 [39] and
the SMP-containing Nvj2 [40] (see Box 3), presumably implicated in
lipid exchange. Most recently, novel conserved inter-organelle tethering proteins were identified, possibly implicated in lipid metabolism
at the ER–vacuole interface. Mdm1 localizes at the NVJ and contains
an uncharacterized PXA domain [41], while Lam6/Ltc1 localizes to multiple MCSs, including the NVJ via interaction with Vac8, and contains a
START-like/VASt domain (see below and Box 2B) [42]. Whether all of
these mammalian and yeast LTPs actually mediate lipid transfer between ER and lysosomes is still unknown.
3. Mitochondria: functions and membrane contacts with the ER
Mitochondria are endosymbiotic organelles, containing their own
genome and protein synthesis apparatus, and acting as the primary
site of energy production in eukaryotic cells. In addition, they are required for a plethora of other essential functions, including biosynthesis
and degradation of lipids, Ca2 + buffering and signaling, regulation of
884
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
Box 2A
Protein families involved in lipid transport at membrane contacts: architecture of START containing proteins.
The founder member of this family is the steroidogenic acute regulatory protein (StAR). The StAR-related lipid transfer (START) domain, approximately 210 amino acid long, adopts a ‘helix-grip’ fold, in which a central nine-strands antiparallel β-sheet is gripped by N-terminal and C-terminal α-helices. The structure forms an amphiphilic inner cavity with a narrow opening, so that it becomes accessible to the ligand, most likely by a
major structural rearrangement of the domain, possibly involving the C-terminal α helix acting as a lid. The canonical START domain is conserved in plants and animals, but not in budding yeast. In mammals it is common to a family of 15 proteins, designated STARD1-15, which
are found in a variety of subcellular localizations, in particular at MCSs, and bind glycerolipids, sphingolipids and sterols, functioning as lipid sensors or transporters [34].
Proteins belonging to the STARD family differ considerably in their domain architecture, as follows.
1) STARD1/StAR has a mitochondrial targeting signal at its N-terminus.
2) STARD3/MLN64 hosts MENTAL (MLN64 N-terminal domain), a four transmembrane-spanning domain, responsible for late endosomal
targeting, homo and hetero-oligomerization and cholesterol-binding. The MENTAL domain also holds a FFAT-like motif at its C-terminus,
for VAP binding on the ER.
3) STARD2, 4, 5, 6, 7, and 10 only show the START domain.
4) STARD9 is a unique, very large (500 kDa) member of the family, with an N-terminal kinesin motor domain and an FHA phosphoprotein
binding domain.
5) STARD11/CERT contains both a PH domain, responsible for the interaction with phosphoinositides on the Golgi complex, and a FFAT
motif, which is recognized by VAPs on the ER.
6) STARD8, 12 and 13 display a RhoGAP domain, which stimulates the GTPase activity of small GTPases belonging to the Rho family. They
also comprise a SAM (sterile α motif) domain that is a putative protein–protein interaction module.
7) STARD14 and 15 comprise two acyl–CoA hydrolase domains.
Amino acid length of individual proteins is given in brackets. Since many isoforms, resulting from different transcripts variants, are known for
some of these proteins, only the amino acid length of the main isoform (typically named “a” or “1”) is reported.
apoptosis and autophagy. This multiplicity of roles, highly relevant for
cell survival and activity, are accomplished by mitochondria thanks to
their ability to appropriately trigger and process cellular signals, including those induced by nutrient availability. Indeed, mitochondria can
modify their dynamics upon starvation, by elongating in order to
avoid degradation by autophagy [43,44], and by adapting their motility
to the variable extracellular glucose concentration [45]. Moreover, mitochondria control their own biogenesis and function, thanks to an
autoregulatory feedback loop mediated by the AMPK signaling cascade.
Upon energy deprivation, AMPK activates PGC1α, a co-activator that
enhances the activity of nuclear transcription factors promoting the expression of mitochondrial genes. In turn, this mechanism induces both
the oxidation of glucose and fatty acids, and mitochondrial biogenesis
[3]. Finally, the function of mitochondria impacts profoundly on the
overall cellular metabolism. In fact, in addition to AMPK, it also controls
the mTORC1 pathway, in both AMPK-dependent and independent ways
[46–48] (Fig. 1B).
Essential to mitochondrial functions and quality control is the dynamic shape and distribution of the mitochondrial network, constantly
fusing and dividing, thanks to pro-fusion and pro-fission machineries
[49]. These morphology and positioning features are adapted in a precise space and time-controlled fashion, and with respect to other cellular structures or organelles. For example, mitochondrial distribution to
neuronal and immunological synapses is essential for the proper development and function of neurons and lymphocytes, respectively [50–53].
Furthermore, mitochondria are physically and functionally connected to
the ER by well-known juxtapositions of major physiopathological relevance (for extensive recent reviews see [54,55]). MCSs between the
ER and mitochondria are also defined as mitochondria-associated membranes (MAMs) and consist of ER membrane areas that are reversibly
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
885
Box 2B
Protein families involved in lipid transport at membrane contacts: architecture of START-like/VASt containing proteins.
Although budding yeast has no canonical START proteins, a conserved family of eukaryotic proteins containing a distantly related, START-like
domain was recently identified [42,76,130]. This domain, also called VASt domain (VAD1 Analog of StAR-related lipid transfer), is poorly conserved in terms of primary sequence, but is predicted to form a hydrophobic pocket for lipid binding similar to the canonical START domain
[130]. The START-like/VASt domain is present in six yeast proteins, which bind sterols and are implicated in lipid transport at MCSs
[42,75,76], and in three human proteins of uncharacterized function.
Proteins belonging to the START-like/VASt family also share GRAM domains, belonging to the PH domain superfamily, and transmembrane domains (typically for ER-anchoring). Yeast members of the family contain additional domains, as follows.
1) Ysp1 and Sip3 present, at their N-terminus, a BAR (Bin–Amphiphysin–Rvs) domain, which is a protein dimerization motif able to sense or
induce membrane curvature and thus regulating membrane binding.
2) Ysp2/Ltc4 and Lam4/Ltc3 host coiled-coil motifs.
Ysp1 and Sip3, and Ysp2 and Lam4 localize at ER–plasma membrane contact sites, while Lam5 and Lam6 show multiple localizations (ER–vacuole, ER–mitochondria, vacuole–mitochondria contact sites). Amino acid length of individual proteins is given in brackets.
tethered to the mitochondria. These structures are implicated in several
functions, including regulation of Ca2+ buffering and signaling, lipid
biosynthesis and transfer, mitochondrial dynamics, and cell survival.
In fact, MAMs are fundamental to sense and regulate Ca2 + release
from the ER, since they form cytosolic microdomains where sufficiently
high Ca2+ concentrations are achieved to activate the low affinity mitochondrial Ca2+ uniporter [56,57].
MAMs are also involved in the transfer of lipids between ER and mitochondria, a process necessary for the maintenance of mitochondrial
membrane composition and for proper phospholipid biosynthesis in the
whole cell. In fact, mitochondria are capable to perform only a subset of
phospholipid synthesis reactions and need to import from the ER numerous phospholipids: phosphatidic acid (PA), phosphatidylserine (PS),
phosphatidylcholine (PC), phosphatidylinositol (PI), and sphingolipids
[5,58]. They also import from the ER the small amount of cholesterol
present in their outer membrane, needed, in specialized steroidogenic
cells, for the synthesis of steroid hormones [59]. On the other hand, mitochondria synthesize their specific phospholipid cardiolipin (CL) and,
most importantly, the majority of phosphatidylethanolamine (PE) required by cell membranes, by converting ER-derived PA and PS, respectively. Part of PE is then transported back to the ER, where it is used to
generate PC. Therefore, a large, bidirectional lipid flux exists between
ER and mitochondria [5].
Lipid-synthesizing enzymes are enriched at MAMs, suggesting
that high local concentration of lipids could favor LTP-mediated
transfer between membranes and subsequent enzymatic reactions.
Despite the strong evidence for lipid exchange at MAMs, most mechanistic details are still discussed. A number of ER–mitochondria
established or postulated tethering complexes have been identified
in yeast and metazoans. In mammalian cells the following protein
complexes have been described:
(1) the inositol triphosphate receptor (IP3R)–cytosolic glucose regulated protein 75 (GRP75)–mitochondrial voltage dependent
anion channel (VDAC) complex [60];
(2) the Mitofusin 2 (Mfn2) homo or heterotypic complexes, although at the moment it is unclear whether Mfn2 acts as a molecular tether [61], or a tethering antagonist [62,63];
(3) the B cell receptor-associated protein 31 (Bap31)–mitochondrial
Fission 1 homolog (Fis1) complex [64];
(4) the ER vesicle-associated membrane protein-associated protein B (VAPB)–mitochondrial protein tyrosine phosphatase
interacting protein 51 (PTPIP51) complex [65,66].
Several different protein complexes have been identified in yeast
cells as well:
(1) the ER–mitochondria encounter structure (ERMES) [67];
(2) the mitochondrial myosin 2 receptor 1 (Mmr1) containing complex [68];
(3) the ER membrane protein complex (EMC)–mitochondrial
translocase of the outer membrane (TOM) complex [69];
(4) the ER-associated sterol transporter Lam6/Ltc1–Tom70/71 complex [42].
Some of these tethering complexes, necessary for normal phospholipid biosynthesis and composition of mitochondrial membranes, have
886
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
Box 3
Protein families involved in lipid transport at membrane contacts: architecture of SMP containing proteins.
The SMP (synaptotagmin-like mitochondrial lipid-binding protein) domain derives its name from the heterogeneous features of the group of proteins that contain it. It comprises a β helix followed by four β strands and another α helix, with an overall length of about 200 amino acids [131].
More recently, the SMP family has been structurally correlated to the TULIP (tubular lipid-binding proteins) superfamily of proteins, displaying
the ability to bind lipids and/or other hydrophobic ligands. Members of this superfamily show little sequence identity, but share a common tubular fold, forming a central cavity resembling a hydrophobic tunnel, which would facilitate the transfer of lipids between adjacent membranes
[132]. Indeed, some SMP domains, namely those of Mmm1 and Mdm12, and of E-Syt2, were demonstrated to bind phospholipids [70,133].
The SMP domain is common to a family of membrane-associated proteins, widespread among eukaryotic species and characterized by their
exclusive localization at MCSs [40]. Proteins containing SMP domains mediate inter-organelle tethering at MCSs and are likely implicated in lipid
exchange and metabolism [40,67,70,133–135].
Proteins belonging to the SMP family also contain one or more transmembrane (TM) or hairpin domains for membrane anchoring (typically ER)
and can be subdivided into four categories according to their domain architecture, as follows [131].
1) Tricalbins (Tcb1-3) and extended-synaptotagmins (E-Syt1-3) are synaptotagmin-like proteins, containing C2 domains, which bind anionic lipids (PI4,5P2 at the plasma membrane) in a Ca2+ dependent manner. They hold an N-terminal hairpin region that anchors them to the
cytoplasmic leaflet of the ER without spanning the bilayer.
2) Nvj2 and HT008/TEX (testis-expressed protein 2) contain a PH domain, responsible for the interaction with phosphoinositides.
3) PDZK8 hosts both a PDZ domain, which is a protein–protein interaction domain, and a C1 domain, which binds diacylglycerol.
4) Mmm1, Mdm12 and Mdm34 are three (out of four) subunits of the ERMES complex, responsible for the best known ER–mitochondria
contact site in yeast.
Amino acid length of individual proteins is given in brackets.
been directly implicated in lipid exchange [42,67,69]. Among them,
however, only the ERMES complex contains subunits carrying putative
phospholipid binding domains. Indeed, out of four ERMES subunits,
three contain SMP domains (see Box 3), and two of them (Mdm12
and Mmm1) have been demonstrated by computational modeling and
electron microscopy (EM) analysis to assemble in a tubular structure
traversed by a hydrophobic tunnel [70]. These two SMP domains bind
phospholipids, particularly PC, supporting the direct involvement of
ERMES in phospholipid exchange at MAMs [70]. Remarkably, ERMES
is not conserved in metazoans.
4. Membrane contacts between mitochondria and
lysosomes/vacuoles
4.1. Identification of the vacuole and mitochondria patch (vCLAMP) in yeast
cells
While the presence of MCSs between either LEs/lysosomes, or mitochondria, and the ER is documented by extensive evidence, direct
interaction of lysosomes/vacuoles (or lysosome-related organelles,
LROs) with mitochondria emerged only recently in yeast and mammals
[71–73]. In yeast, two papers reported the identification of a novel physical contact, named vCLAMP, between the lysosome-like vacuole and
mitochondria, dynamically regulated by nutrient stimuli [71,72]. Different
approaches by the two groups led to the identification of the same implicated protein, Vps39 (also known as Vam6), a component of homotypic
fusion and vacuole protein sorting (HOPS) tethering complex [74]. Optical
and EM analyses revealed that Vps39, possibly acting as a tether, localizes
to the newly identified vacuole–mitochondria contacts, and that the protein, together with its vacuolar binding partner, the Rab7-like GTPase
Ypt7, is the only HOPS component necessary and sufficient for vCLAMP
formation. In fact, Vps39 over-expression leads to a massive increase of
vacuole–mitochondria contacts, which otherwise are rarely observed,
while Vps39 mutations cause loss of vCLAMPs [71,72].
The initial drive of these studies was the surprising observation that,
in yeast, the best known candidate for lipid exchange between ER and
mitochondria, i.e. the ERMES complex, can be mutated without a dramatic phenotype. Indeed, although ERMES mutants cannot grow in
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
media requiring respiration, they are viable in regular fermentable
conditions [67]. The mutation of ERMES clearly leads to abnormalities
in the rate of phospholipid biosynthesis, but at steady-state only the
mitochondria-specific phospholipid CL shows a significant reduction,
suggesting the presence of compensatory mechanisms [67]. Interestingly, ERMES and vCLAMP appear coregulated, possibly carrying partially
redundant functions, one compensating for the other, both upon genetic
and metabolic perturbations of the system (Fig. 3) [71,72]. Indeed,
887
Vps39 mutations cause both a loss of vCLAMPs and an increase of
ERMES-mediated contact sites. Vice versa, mutation of ERMES causes a
dramatic increase of vCLAMPs, and Vps39 over-expression is sufficient
to almost completely rescue the defective phenotype of ERMES mutants, allowing them to grow even in respiratory media. Moreover,
coregulation of the two mitochondrial membrane contacts is evident
under different growth conditions, either promoting or repressing mitochondrial respiration. In cells grown in glycerol as the only carbon
source (promoting respiration), the number of ERMES sites is strongly
increased (concomitantly with expansion of the mitochondrial network), while vCLAMPs are reduced via Vps39 phosphorylation. Instead,
opposite effects are observed when the cells are grown in the presence
of glucose (promoting fermentation) [71]. Finally, although mutation of
both complexes is lethal, a conditional knockout strategy revealed that
at least one of the functions of this novel contact site concerns phospholipid transport. Combined mutation of ERMES and vCLAMP severely affects the exchange and synthesis of phospholipids, in both the ER and
mitochondria. While PS and PI accumulate, the amounts of PE, PC and
CL decrease dramatically, suggesting that both these MCSs facilitate
phospholipid exchange with mitochondria [72].
4.2. Regulatory players of the ERMES–vCLAMP cross-talk in yeast cells
The findings described above underline the concept that contact
sites are dynamic and can change in size in response to genetic perturbations or environmental stimuli. How this is regulated, however, remains unclear. Subsequent work by the Nunnari and Schuldiner
groups addressed this issue by searching for physical interactors of the
ERMES complex, which could justify how the cross-talk of ERMES and
vCLAMP is accomplished in a dynamic and reciprocal way [42,75].
Both research groups identified Lam6/Ltc1, an ER integral membrane
protein belonging to a conserved protein family, containing a STARTlike/VASt domain, and capable to bind sterols [42,76]. Lam6/Ltc1 associates to multiple MCSs, including the NVJ (and possibly other non-NVJ
ER–vacuole contacts), and the ER–mitochondria juxtaposition, likely
via binding partners such as the vacuolar protein Vac8 and the mitochondrial import receptors Tom70/71, but it is also localized to vCLAMP
[42,75,76]. While Lam6/Ltc1 is not required for contact formation, it is
necessary for the reciprocal coordination and compensatory role of
ERMES and vCLAMP (Fig. 4). Indeed, Lam6/Ltc1 is sufficient to increase
the size of ERMES, vCLAMP and NVJ upon over-expression, and it is
needed for the expansion of ERMES in the absence of vCLAMP, or vice
versa for the expansion of vCLAMP under conditions of compromised
ERMES [75]. In contrast, Lam6/Ltc1 mutation leads to a sick/lethal phenotype in the presence of ERMES loss of function [42,75]. Which is the
molecular function of Lam6/Ltc1? This protein was shown to selectively
exchange cholesterol between liposomes in vitro, suggesting that it
could operate as a sterol transfer or sensor in vivo [42]. Its ability to participate in the functional compensation by ERMES for vCLAMP, and vice
versa, may imply alternative or additional roles, either as a tether, or as a
phospholipid transporter. At variance with ERMES, Lam6/Ltc1 belongs
to a family of highly conserved proteins, from yeast to mammals.
Fig. 3. ERMES and vCLAMP membrane contacts are dynamically coordinated in yeast
cells.(A) Schematic representation of MCSs, connecting nuclear membrane (nuclear ER),
ER, vacuole and mitochondria: the NVJ, composed by Nvj1 (on the nuclear ER) and Vac8
(on the vacuole); ERMES, comprising Mmm1 (on the ER), and Mdm10, Mdm12 and
Mdm34 (on mitochondria); vCLAMP, connecting the ER and the vacuole, and comprising
the cytosolic protein Vps39 (also part of the HOPS complex, not shown). Upon genetic or
environmental perturbations, ERMES and vCLAMP are dynamically coregulated. (B) In
Vps39 mutants, the ERMES complex expands to compensate. (C) In ERMES mutants,
vCLAMP expands to compensate. In addition, Vps39 over-expression rescues almost
completely the ERMES-mutant phenotype, while combined mutation of both ERMES
and Vps39 is lethal (similarly to ERMES and EMC mutants; not shown). Finally, when wild
type cells are grown in non-fermentable medium (glycerol, promoting respiration) rather
than in regular medium (glucose), vCLAMP is repressed and ERMES expands, similarly to
the situation shown in panel B. Consistently, ERMES mutants do not grow in non-fermentable media.
888
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
Therefore, its identification opens promising perspectives for the understanding of general molecular mechanisms operating at contact sites.
An additional regulatory player of the ERMES–vCLAMP cross-talk is
the vacuolar protein sorting 13 (Vps13), a large protein of unknown
molecular function peripherally associated to endosomes, very well
conserved up to metazoans [77]. Similarly to Vps39, point mutations
of Vps13, presumably leading to gain-of-function, are sufficient to rescue all consequences of ERMES deficiency, while combined loss of
Vps13 and ERMES is lethal. Vps13 localizes to vacuoles at contacts
with mitochondria and the nucleus, suggesting that it is part of vCLAMP
and NVJ, depending on the growth conditions. In particular, when the
formation of vCLAMP is repressed by growing the cells in nonfermentable carbon sources (glycerol), Vps13 is dynamically and reversibly relocalized to the NVJ (Fig. 5). The ability of this protein to compensate for ERMES deficiency and redistribute to vCLAMPs or NVJs
further suggests that these contacts function coordinately in redundant
pathways [77]. Altogether, these studies indicate that the function of
ERMES can be at least partly bypassed by the function of other MCSs, including vCLAMPs, NVJ and possibly EMC, which establish a redundant
and coordinated organelle network, and ensure alternative routes for
lipid transport to mitochondria. In this scenario, both Lam6/Ltc1 and
Vps13 might play a key regulatory role, by redistributing to distinct
MCSs and promoting their physical and functional expansion, depending on genetic alterations and/or growth conditions.
The sensitivity to nutrients in the growth medium raises the question whether this MCS network functionally communicates with one
of the main nutrient sensitive pathway, i.e. TORC1. In addition to
being a component of the HOPS complex and vCLAMP contact, Vps39
acts as a nucleotide exchange factor (GEF) for the Rag GTPase Gtr1, a
subunit of the vacuolar-membrane-associated EGO complex, which in
yeast links nutrient sensing with TORC1 activity [78,79]. Despite a few
differences between the two systems, the EGO complex represents the
yeast functional counterpart of the Ragulator–Rags machinery, which
in mammals recruits and activates mTORC1 on the lysosomal membrane. This process, which occurs in response to amino acids, ultimately
stimulates cell growth [21]. Interestingly, the proteomic analysis of
Lam6/Ltc1 immunoprecipitates identified the TORC1 subunit Kog1
and the down-stream TORC1 target Npr1 as putative vacuolar
interactors [42]. It is conceivable that, in addition or connected to their
role in contact formation at the vacuole/mitochondrial boundary,
Vps39 and Lam6/Ltc1 also cooperate in the regulation of TORC1 activity,
to integrate energy signals from mitochondria to nutrient signals from
the vacuole.
5. Membrane contacts between mitochondria and lysosome-related
organelles
In mammalian cells, the presence of physical connections between
mitochondria and the endolysosomal system has also been observed,
in particular with LROs [73,80]. LROs comprise various cell typespecific organelles, including melanosomes in melanocytes, platelet
dense granules, and secretory lysosomes in hematopoietic cells. These
organelles share features of conventional lysosomes, yet are characterized by unique morphological and biochemical characteristics, consistent with their specialized functions [81]. The biogenesis of LRO is a
sophisticated process requiring the intervention of dedicated pathways
and effectors. Melanosomes are devoted to the synthesis, distribution
and transfer of melanin pigments. They originate from endosomal precursors and subsequently undergo a series of maturation stages, characterized by a precise timing of protein delivery, progressive melanin
deposition and organelle growth, and maturation-dependent transport
towards the cell periphery [82,83]. These events require an intense biosynthetic effort and the availability of energy at precise time points and
subcellular locations.
Electron tomography studies in mouse melanocytes showed that
during their biogenesis melanosomes establish tight physical contacts
Fig. 4. Lam6/Ltc1 regulates the cross-talk among membrane contacts in yeast cells.
(A) Schematic representation of MCSs, connecting nuclear membrane (nuclear ER), ER,
vacuole and mitochondria (NVJ, ERMES, vCLAMP), and of their regulatory partner Lam6/
Ltc1. Lam6/Ltc1 is an integral membrane protein, which binds ERMES, Vac8 (on the
vacuole), and TOM70/71 (on the mitochondria). At steady state it is mainly localized at
ERMES contacts, but also at NVJ and vCLAMP. (B, C) Lam6/Ltc1 is necessary for the
cross-talk between ERMES and vCLAMP, since expansion of ERMES in Vps39 mutants, or
expansion of vCLAMP upon ERMES down-regulation, require an intact Lam6/Ltc1
function. Lam6/Ltc1 over-expression does not rescue the ERMES-mutant phenotype, but
combined mutation of both ERMES and Lam6/Ltc1 is lethal (while mutation of Lam6/
Ltc1 and Vps39 is not; not shown).
with mitochondria. These connections do not evolve into continuities
of the melanosomal lumen and mitochondrial inner compartments.
Rather, the two membranes remain connected by electron-dense fibrillar bridges, 20–30 nm long, resembling the protein tethers linking
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
889
The information summarized here suggests that melanosome–mitochondrion contacts, which display a dynamic, organelle maturationdependent behavior, play a role in the melanogenesis process, in particular in the proper maturation of the organelles. The molecular tethers
and function of these inter-organellar connections remain unknown.
The only available information concerns Mfn2. Immunogold analysis
in melanocytes showed that this protein specifically localizes not only
to mitochondrial outer membranes and ER–mitochondria contacts, as
expected, but also to contacts between melanosomes and mitochondria.
In addition, the melanosome–mitochondrion juxtaposition is significantly reduced upon Mfn2 knock-down by siRNAs, suggesting the involvement of the protein in the formation of these connections [73].
Nevertheless, while the role of Mfn2 as a profusion mitochondrial protein is established, its precise role in both the ER–mitochondria and
the melanosome–mitochondrion contacts remains to be clarified [61–
63,73]. Another potential tethering candidate, the OA1 receptor, is suggested by the increased frequency of contacts established by OA1positive melanosomes compared to OA1-negative organelles. Moreover, in OA1-overexpressing cells, 20% of the contacts showed a long,
unusual extension, with striking images of melanosomes strictly adherent or “embraced” by mitochondria [73] (Fig. 6). However, the length of
melanosome–mitochondrion contacts does not correlate with the OA1
labeling density. Therefore, it is unlikely that OA1 functions as a true
tether. Rather the effects of OA1 could depend on its ability to stimulate
melanogenesis, further supporting the idea that the inter-organellar
connections are relevant during the organellogenesis process.
The function of melanosome–mitochondrion contacts may be related to several aspects of organelle biogenesis. Since the pharmacological
inhibition of mitochondrial ATP synthesis severely reduced contact formation and impaired melanosome biogenesis, by affecting in particular
the developing organelles, it is possible that close mitochondria are necessary to provide melanosomes with the ATP necessary for their maturation and/or transport. Given that melanosomes contain Ca2+ [87–89],
release of the ion might be the signal to activate the mitochondrial metabolism, with ensuing increased ATP production in the vicinity of maturing organelles. Maturation is also characterized by a progressive
increase in the size of melanosomes. Although the new membrane required for melanosome growth is probably delivered by vesicular traffic,
together with the melanogenic enzymes [83], a possible role of lipid
transfer at melanosome–mitochondrion contacts cannot be excluded
[80].
Fig. 5. Vps13 regulates the cross-talk among membrane contacts in yeast cells.
(A) Schematic representation of MCSs, connecting nuclear membrane (nuclear ER), ER,
vacuole and mitochondria (NVJ, ERMES, vCLAMP), and of their regulatory partner
Vps13. Vps13 is a peripheral membrane protein of endosomes, which binds Vps39. At
steady state in regular medium (glucose), it is mainly localized at vCLAMP. (B) During
growth in non-fermentable medium (glycerol), vCLAMP is repressed and Vps13
translocates to the NVJ. Gain-of-function mutations of Vps13 completely rescue the
ERMES-mutant phenotype, while combined mutation of both ERMES and Vps13 is lethal
(not shown). During growth in non-fermentable medium, respiration is promoted and
the mitochondrial network is expanded.
mitochondria to the ER [73]. Quantitative EM analysis showed that melanosome–mitochondrion contacts are relatively rare in resting cells.
Nevertheless, they are substantially more frequent with early organelles, particularly when melanogenesis is actively stimulated, involving
up to 17–28% of immature melanosomes. In fact, melanosome–mitochondrion contacts are significantly more abundant around the nucleus,
the area where new melanosomes are generated. Moreover, their frequency and length increase when melanosome biogenesis is stimulated
by exogenous expression of OA1, a G protein-coupled receptor implicated in ocular albinism and organellogenesis, in Oa1-KO melanocytes [84–
86]. By contrast, contacts are reduced in conditions of abnormal melanosome biogenesis, due to a primary melanosomal defect (Oa1-KO melanocytes), or to a primary mitochondrial dysfunction (Mfn2 silencing
or inhibition of ATP synthase) [73].
6. Membrane contacts between mitochondria and autophagosomes
During starvation, a highly conserved process termed macroautophagy (or simply autophagy, literally “self eating”) is activated,
allowing the cells to retrieve nutrients via degradation of their own intracellular components. Cytosolic macromolecules and entire organelles are
first engulfed within double-membrane vacuoles, measuring 0.5–1.5 μm
in diameter and named autophagosomes, and then are delivered to lysosomes for degradation. The process proceeds by three main steps, the formation of the so called “isolation membrane” (or phagophore), the
closure of the isolation membrane around the cytosolic content to generate the autophagosome, and the fusion of the autophagosome with the lysosome/vacuole to produce the autolysosome. Autophagy can be nonselective, when induced by nutrient deprivation and/or TORC1 inhibition,
or can be substrate-specific, when single damaged organelles or cytosolic
components are involved. In the latter case, the targets of the autophagic
process are labeled by specific signals (such as ubiquitination) recognized
by cytosolic receptors. However, the subsequent steps are in common
with bulk autophagy [90,91].
6.1. Properties of the autophagosomal membranes
Compared to other organelles, the peculiarity of autophagosomes is
that upon induction of autophagy they form rapidly (in a few minutes)
890
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
(PAS), where most early ATG proteins are transiently localized [92,93].
In mammalian cells in starvation conditions, the phagophore nucleates
from multiple membrane structures, named omegasomes, similar to the
Greek letter omega in EM images [94–96]. Omegasomes exhibit a cupshaped structure in continuity with the ER, as shown by electron tomography analyses [95,96]. They can be envisaged as specialized PI3Penriched EM subdomains, where the hierarchical assembly of ATG
proteins promotes the formation of the isolation membrane [90]. One of
the most striking features of the omegasomes, compared to the ER
cisternae that surround them, is the very dark appearance of their
membrane upon osmium staining. This property confirms the
unique composition of omegasomes, highly enriched in lipids compared to other cell membranes.
These considerations are in agreement with initial freeze-fracture
EM studies on early autophagosomes, which revealed the poor protein
content of both autophagosomal membranes [97–99]. The freezefracture technique splits biological membranes, visualizing transmembrane proteins as intramembrane particles of size proportional to the
number of membrane-spanning α helixes. Studies performed both on
mouse and yeast cells [97,98], and on isolated rat organelles [99]
uncovered a surprising smoothness of the outer and inner
autophagosomal membranes, with intramembrane particles up to 100
times less numerous than in typical ER, Golgi, plasma membrane, lysosomal, and mitochondrial membranes [99]. These findings indicated
that a unique primary property of early autophagic vacuoles is the almost complete lack of transmembrane proteins. The resulting idea
was that the autophagosomal membranes are specialized just for delivering cytosolic components to degradation. In any instance, the possible
derivation of these membranes from other organelles would require an
intense remodeling, necessary for the observed very low level of transmembrane proteins [97–99].
6.2. Origins of the autophagosomal membranes
Fig. 6. Membrane contacts between melanosomes and mitochondria are enhanced by
OA1. Representative EM images of long contacts between mitochondria and
melanosomes in Oa1-KO melanocytes transiently transfected with a plasmid expressing
wild type OA1. In these conditions melanosome biogenesis is actively stimulated and
results in a significant increase in the frequency and length of inter-organellar contacts.
However, a clear correlation between the length of contacts and the labeling for OA1
(represented by the gold particles) is lacking, suggesting that the protein does not
function as a real tether. (A) Long contacts weakly labeled for OA1 (T.D. and M.V.S.
unpublished results). (B) Long contacts strongly labeled for OA1 (Reprinted from
Current Biology, Vol 24, Pages 393–403; Authors: Daniele T., Hurbain I., Vago R., Casari G.,
Raposo G., Tacchetti C., and Schiaffino M.V.; Title: Mitochondria and melanosomes
establish physical contacts modulated by Mfn2 and involved in organelle biogenesis;
Copyright 2014, with permission from Elsevier). Mito, mitochondria; I–IV, melanosomal
maturation stages I (immature) to IV (mature, fully melanized organelles).
and de novo. Especially during starvation, the number of autophagosomes
increases dramatically. In order to cope with this task, therefore, the cells
must have the capacity to rapidly generate and/or mobilize a substantial
amount of membranes. The discovery in yeast and mammals of the highly
conserved group of autophagy-related (ATG) genes has advanced substantially our understanding of the molecular mechanisms underlying
autophagosome biogenesis. However, the origin and composition of the
autophagosome membranes remains a matter of debate [91]. In yeast
cells subjected to nutrient deprivation, the isolation membrane originates
from a single peri-vacuolar site, named pre-autophagosomal structure
Despite our poor knowledge about the exact lipid/protein composition of the isolation membrane, it is generally agreed that in mammals
it nucleates from the omegasome, and is subsequently expanded, thanks
to lipid contributions from several sources to become an autophagosome.
The lipid sources involved could be, via vesicular transport or nonvesicular membrane flow at MCSs, other pre-existing cytoplasmic
organelles, including Golgi, endosomes, plasma membrane, LDs and mitochondria [100–103]. However, live cell imaging analyses in starved mammalian cells showed that the membranes contributing to the very initial
stages of phagophore formation are the mitochondrial outer membranes.
In fact, both early and late autophagosomal markers, Atg5 and LC3 respectively, appear as transient punctae on mitochondria [101]. In addition, the
transfer to autophagosomes of fluorescently-labeled mitochondrial outer
membrane proteins and lipids (NBD-PS, converted to NBD-PE in mitochondria) reveals the existence of transient membrane continuity between the two organelles. This continuity, however, does not allow the
transfer of proteins spanning the two leaflets of the mitochondrial outer
membrane. These results suggest the existence of a diffusion barrier
that prevents the delivery to autophagosomes of most mitochondrial proteins, but not of mitochondrial lipids [101]. Overall, these findings supported the idea that the membrane source to forming autophagosomes
is provided by mitochondria, rather than the ER.
These contrasting views were eventually combined into a single
model, based on the discovery that, in starved mammalian cells, the origin and initial growth of the isolation membrane occurs at ER–mitochondria contacts [104]. By optical and EM analyses, as well as by
subcellular fractionation, the early autophagosome markers Atg5 and
Atg14L were shown to localize at MAMs. Live cell imaging showed
that Atg5 localization remained at both organelles throughout
autophagosome formation, but with different dynamics. Atg5 displays
stable association with the ER, consistent with the role of this organelle
as the platform of autophagosome formation. In contrast, the association
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
with mitochondria exhibits oscillations, suggesting that the role of the
latter consists in the systematic supply of further components [104].
The physical and functional integrity of MAMs is necessary for
autophagosome biogenesis. In fact, knock-out or knock-down of either
Mnf2 or phosphofurin acidic cluster sorting protein-2 (PACS-2), a cytosolic protein implicated in ER–mitochondria communication [105], severely affects the overall process [101,104]. Finally, upon starvation the
ER integral membrane protein syntaxin 17 is relocated to MAMs, behaving as an up-stream signal for the specific re-localization of the early autophagy marker Atg14. In the absence of syntaxin 17, the maturation of
phagophores to form autophagosomes and subsequently autolysosomes
is severely impaired, suggesting a critical role of MAM localization for
proper phagophore maturation [104].
Results analogous to those obtained in mammals were obtained in
yeast, in relation to a selective type of autophagy, namely the selective
removal of mitochondria (mitophagy) [106]. When yeast cells are cultured in nutrient limiting conditions (or in the presence of the TORC1
inhibitor rapamycin), multiple autophagic pathways are activated.
First, the isolation membrane devoted to general autophagy nucleates
at the PAS structure, near the vacuole [92]. Concomitantly, the NVJ
grows progressively and eventually invaginates into the vacuole, leading to the special type of autophagy PMN, whereby bits of the nucleus
become digested. In these conditions, mitophagy is also activated to
eliminate damaged or unnecessary mitochondria, by promoting their
recruitment to the PAS, by means of the Atg32 receptor and the Atg11
scaffold [107,108].
Several mitophagy-defective strains were identified, including mutants for all of the ERMES subunits, by the use of a special screening
strategy for genes implicated in the induction of mitophagy upon starvation [106]. The results showed that the phenotype of ERMES mutants
is limited to mitophagy and does not depend on respiratory deficiency,
nor on abnormal mitochondrial morphology, but can be complemented
by artificial ER–mitochondrial tethering (except for the ERMES subunit
Mdm12). ERMES localizes to sites of autophagosome formation and
PAS, but appears specifically restricted to the border of the forming
phagophore, opposite to the vacuolar membrane. Fluorescence complementation assays showed that ERMES interacts with Atg8 (homologous
to the LC3 family in mammals), but not with mitophagy specific effectors Atg32/11. Moreover, the morphological analysis of ATG proteins
in ERMES mutants showed alterations concerning only the late autophagy markers. These results support the concept that ERMES is required
at a relatively late step of mitophagy, specifically at the expansion of
the isolation membrane, possibly thanks to its role in promoting lipid
exchange between membranes, or in maintaining the integrity of the
ER–mitochondria connection [106]. At variance with mammals, in
yeast mitophagy is induced by nutrient starvation concomitantly with
non-selective autophagy and most of the known molecular processes
of mitophagy are the same of non-selective autophagy. Thus, the results
of these findings could possibly be of relevance for general autophagy as
well [106,109].
Overall, the current evidence supports a model in which the isolation membrane emerges from MAMs, where ER and mitochondria likely
cooperate to supply the necessary amounts of lipids to the growing
phagophore. Subsequently, other organelles contribute to the maturation of the autophagosome, mainly by vesicular traffic mechanisms
[90]. Once initiated, the expansion of the phagophore membrane proceeds at a rapid rate and with peculiar protein-poor features, different
from both the ER and mitochondrial membranes, as revealed by
freeze-fracture EM [97–99]. How are these characteristics achieved? It
is likely that de novo lipid synthesis and exchange are needed. MAMs
are enriched in lipid synthesizing enzymes, and while the ER is the
main site for lipid biosynthesis, mitochondria are required to generate
PE, a phospholipid present throughout cellular membranes and
enriched in autophagosomes. Although the mechanisms and LTPs mediating lipid exchange at MAMs are not yet understood, it is tempting to
speculate that the isolation membrane might be (at least partly)
891
generated by direct lipid insertion, which could justify the loss of integral proteins from the contributing membranes.
7. Membrane contacts between mitochondria and lipid droplets
LDs are present in all cells and play critical roles in lipid metabolism,
representing the main cellular storage for neutral lipids, primarily triacylglycerols and steryl esters. At variance with all other compartments,
they are surrounded by a lipid monolayer, composed of phospholipids
and free cholesterol, facing the cytosol with their polar heads. The surface includes also several proteins, with structural (perilipin), enzymatic (lipases, lipid synthases) or membrane traffic functions [110]. In order
to accumulate and release lipids, LDs are physically and/or functionally
interconnected to several organelles, including ER, lysosomes, mitochondria, peroxisomes and other LDs [111]. De novo LD biogenesis occurs by budding from the ER, thanks to a conserved family of fat
storage-inducing transmembrane (FIT) proteins [112]. Subsequently,
the two compartments remain in contact, with the ER supplying LDs
with newly synthetized molecules, most likely by membrane conduits
that connect the two membranes and ensure both lipid and protein
transfer [113,114]. Other LD connections are less characterized. Close
association between LDs and mitochondria in mammals (or LDs and
peroxisomes in yeast) is a common finding in many cell types, especially
in highly oxidative tissues [115,116] and during nutrient starvation
[117]. In fact, in order to support oxidative respiration and produce
ATP, the cells adapt their metabolism from glycolysis to β-oxidation of
fatty acids in mitochondria. To this end, fatty acids are mobilized and released from LDs, either by lipolysis or lipophagy, and transferred to mitochondria. Although Perilipin 5 has been proposed as a possible
mediator of LD–mitochondria contacts [116], the responsible machinery has not been established.
The actual route by which fatty acids, which are toxic to cells, travel
from LDs to mitochondria has been recently investigated in living mammalian cells by using a fluorescent fatty acid probe (BODIPY-tagged 12carbon saturated fatty acid) [117]. Rambold et al. showed that, under
serum and amino acid starvation, lipolysis represents the main mechanism of fatty acid transfer to mitochondria. To operate, this transfer requires inter-organelle proximity and mitochondrial fusion. Downregulation or inhibition of cytoplasmic neutral lipases is sufficient to
completely block fatty acid transfer to mitochondria and to reduce mitochondrial respiration, causing enlarged LDs, during both short and long
times of starvation. Lipophagy and in general autophagy are critical to
replenish LDs with new fatty acids, which then can be transferred to mitochondria upon lipolysis. Autophagy, however, is relevant only upon
prolonged starvation. In fact, at short time points, inhibition of autophagy does not interfere with fatty acid redistribution and oxygen consumption by mitochondria. By contrast, upon prolonged starvation,
the inhibition of autophagy depresses mitochondrial respiration. In
these conditions, independently on lipophagy, the general autophagic
process, involving digestion of other organelles, plays the most relevant
role, since it is responsible for the maintenance and replenishment of LD
stores. In addition, during starvation mitochondria appear highly fused,
a morphological property that is required to homogeneously distribute
the fatty acids transferred from LDs, and to maintain mitochondrial respiration over time [117]. Indeed, defective mitochondrial fusion (and βoxidation) leads to re-export of fatty acids or to their accumulation into
LDs, which increase in size and number. Eventually, the excess of fatty
acids can be discharged in the extracellular space by a lipase-mediated
mechanism. Overall, this coordinated three-organelle mechanism, involving autophagosomes, LDs and mitochondria, ensures a rapid availability of fatty acids to mitochondria, without a concomitant excess of
free fatty acids in the cytosol. In other words, at least during amino
acid and serum starvation, LDs appear to function as conduits of fatty
acids destined to mitochondria.
On the other side, the cross-talk between mitochondria, LDs and
autophagosomes probably depends on starvation/stress conditions
892
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
and cell type. In starved mammalian cells, LDs have been also shown to
provide lipid precursors to nascent autophagosomes, thus contributing
to autophagic membrane formation [103]. These findings were further
supported by studies in yeast, where the biogenesis of LDs, the mobilization of neutral lipids by lipolysis, and MCSs between ER and LDs
were all shown to be important for the autophagic process [118].
These findings support the possibility that under nitrogen starvation,
yeast cells use lipids of LDs, derived from lipolysis, to provide the ER
with precursors for phospholipid synthesis, necessary to properly accomplish the formation and elongation of the phagophore. Due to
their characteristics, LDs can donate lipids to other organelles for various processes, including membrane assembly and energy production,
and are strictly connected to the ER and mitochondria. While all three
organelles clearly contribute to the autophagic process, the detailed
clarification of the role of LDs and the route followed by LD-derived
lipids will require further investigation.
pathogenesis of metabolic syndrome and other diseases, with significant
pathogenetic and possibly therapeutical implications [54,55,122,123].
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
This work was supported in part by the Vision of Children
Foundation-USA (to M.V.S.). Work by T.D. was supported in part by
the Fondazione San Paolo, Italy. We are indebted to Dr. J. Meldolesi for
precious advices and critical reading of the manuscript, and we thank
Drs. G. Casari and S. Schiaffino for helpful suggestions. We apologize to
colleagues whose relevant work could not be cited due to space
limitations.
References
8. Conclusions
In recent years, we have witnessed the identification of several new
MCSs, bridging together different organelles in a physical and functional
manner. These inter-organellar connections are critical for lipid biosynthesis, transport and metabolism, generating intracellular routes for
lipids that are equivalent to (or more extensive than) the routes
established by the secretory pathway for proteins. Compared to proteins, however, the study of lipids requires different approaches and
techniques that make possible their labeling and biochemical manipulations. At the moment, these techniques are still under development. Despite these difficulties, we begin to appreciate novel concepts and the
general scenario, related to lipid and organelle homeostasis. First, organelles are most often juxtaposed by means of multiple and dynamic
tethering/transport complexes, as observed not only between ER and
mitochondria, but also between ER and plasma membrane [119].
These different MCSs may carry out unique functions, and in addition
they are capable to partially compensate for each other, providing supplementary pathways that allow cell adaptation under various genetic
or environmental perturbations.
Furthermore, lipid transport, together with Ca2+ signaling, can involve several membrane compartments, which cooperate with each
other to accomplish coordinated functions. For example, the formation
of the phagofore, which requires contributions from both the ER and
mitochondrial membranes, might also involve LDs [90,118]. Likewise,
mitochondrial lipid biosynthesis in yeast involves, in a partially redundant way, the ERMES, EMC, vCLAMP and NVJ contacts [67,69,71,72],
and cholesterol export from lysosomes towards the plasma membrane
or the ER involves peroxisomes [120]. Therefore, it is likely that different
subcellular organelles integrate their “independent”, compartmentalized activities in programs much more coordinated than previously
thought, to generate networks of greater structural and functional complexity. An outstanding example of how different subcellular organelles
can cooperate to generate qualitatively superior functions has been recently provided by the characterization of the “ocelloid” of monocellular
dinoflagellates. This eye-like structure, so complex that it was initially
mistaken for a multicellular eye, is in fact built from pre-existing organelles, including mitochondria, plastids and LDs, which assemble in a
subcellular “organ” analogous to the cornea, lens, iris and retina found
in animal eyes [121].
Finally, MCSs are highly relevant for biomedical research and human
pathology, since they are revealing various roles or connections
previously attributed to single organelles. For instance, lysosome–peroxisome membrane contacts have been shown to play a key role in cholesterol transport, and are necessary to avoid cholesterol accumulation
within lysosomes. This mechanism could account for the clinical phenotype observed in genetic peroxisomal diseases [120]. Similarly, abnormal function or expansion of MAMs has been implicated in the
[1] S.M. Ferguson, Beyond indigestion: emerging roles for lysosome-based signaling in
human disease, Curr. Opin. Cell Biol. 35 (2015) 59–68.
[2] C. Settembre, A. Fraldi, D.L. Medina, A. Ballabio, Signals from the lysosome: a control centre for cellular clearance and energy metabolism, Nat. Rev. Mol. Cell Biol. 14
(2013) 283–296.
[3] D.G. Hardie, F.A. Ross, S.A. Hawley, AMPK: a nutrient and energy sensor that maintains energy homeostasis, Nat. Rev. Mol. Cell Biol. 13 (2012) 251–262.
[4] G. Drin, Topological regulation of lipid balance in cells, Annu. Rev. Biochem. (2014).
[5] T. Tatsuta, M. Scharwey, T. Langer, Mitochondrial lipid trafficking, Trends Cell Biol.
24 (2014) 44–52.
[6] J.C. Holthuis, A.K. Menon, Lipid landscapes and pipelines in membrane homeostasis, Nature 510 (2014) 48–57.
[7] 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.
[8] M.J. Phillips, G.K. Voeltz, Structure and function of ER membrane contact sites with
other organelles, Nat. Rev. Mol. Cell Biol. 17 (2016) 69–82.
[9] N. Shai, M. Schuldiner, E. Zalckvar, No peroxisome is an island — peroxisome contact sites, Biochim. Biophys. Acta (2015).
[10] R.L. Farr, C. Lismont, S.R. Terlecky, M. Fransen, Peroxisome biogenesis in mammalian cells: the impact of genes and environment, Biochim. Biophys. Acta (2015).
[11] A. Grabon, D. Khan, V.A. Bankaitis, Phosphatidylinositol transfer proteins and instructive regulation of lipid kinase biology, Biochim. Biophys. Acta 1851 (2015)
724–735.
[12] P. Mattjus, Specificity of the mammalian glycolipid transfer proteins, Chem. Phys.
Lipids (2015).
[13] C.C. Scott, F. Vacca, J. Gruenberg, Endosome maturation, transport and functions,
Semin. Cell Dev. Biol. 31 (2014) 2–10.
[14] P. Saftig, J. Klumperman, Lysosome biogenesis and lysosomal membrane proteins:
trafficking meets function, Nat. Rev. Mol. Cell Biol. 10 (2009) 623–635.
[15] J.P. Luzio, P.R. Pryor, N.A. Bright, Lysosomes: fusion and function, Nat. Rev. Mol. Cell
Biol. 8 (2007) 622–632.
[16] S. Patel, R. Docampo, Acidic calcium stores open for business: expanding the potential for intracellular Ca2+ signaling, Trends Cell Biol. 20 (2010) 277–286.
[17] B.S. Kilpatrick, E.R. Eden, A.H. Schapira, C.E. Futter, S. Patel, Direct mobilisation of
lysosomal Ca2+ triggers complex Ca2+ signals, J. Cell Sci. 126 (2013) 60–66.
[18] A.J. Morgan, L.C. Davis, S.K. Wagner, A.M. Lewis, J. Parrington, G.C. Churchill, A.
Galione, Bidirectional Ca(2)(+) signaling occurs between the endoplasmic reticulum and acidic organelles, J. Cell Biol. 200 (2013) 789–805.
[19] A. Reddy, E.V. Caler, N.W. Andrews, Plasma membrane repair is mediated by
Ca(2+)-regulated exocytosis of lysosomes, Cell 106 (2001) 157–169.
[20] N.W. Andrews, P.E. Almeida, M. Corrotte, Damage control: cellular mechanisms of
plasma membrane repair, Trends Cell Biol. 24 (2014) 734–742.
[21] L. Bar-Peled, D.M. Sabatini, Regulation of mTORC1 by amino acids, Trends Cell Biol.
24 (2014) 400–406.
[22] C.S. Zhang, B. Jiang, M. Li, M. Zhu, Y. Peng, Y.L. Zhang, Y.Q. Wu, T.Y. Li, Y. Liang, Z. Lu,
et al., The lysosomal v-ATPase-regulator complex is a common activator for AMPK
and mTORC1, acting as a switch between catabolism and anabolism, Cell Metab. 20
(2014) 526–540.
[23] C. Settembre, R. Zoncu, D.L. Medina, F. Vetrini, S. Erdin, T. Huynh, M. Ferron, G.
Karsenty, M.C. Vellard, V. Facchinetti, et al., A lysosome-to-nucleus signalling
mechanism senses and regulates the lysosome via mTOR and TFEB, Embo J. 31
(2012) 1095–1108.
[24] D.L. Medina, S. Di Paola, I. Peluso, A. Armani, D. De Stefani, R. Venditti, S.
Montefusco, A. Scotto-Rosato, C. Prezioso, A. Forrester, et al., Lysosomal calcium
signalling regulates autophagy through calcineurin and TFEB, Nat. Cell Biol. 17
(2015) 288–299.
[25] H. Harris, D.C. Rubinsztein, Control of autophagy as a therapy for neurodegenerative disease, Nat. Rev. Neurol. 8 (2012) 108–117.
[26] 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.
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
[27] J.R. Friedman, J.R. Dibenedetto, M. West, A.A. Rowland, G.K. Voeltz, Endoplasmic reticulum–endosome contact increases as endosomes traffic and mature, Mol. Biol.
Cell 24 (2013) 1030–1040.
[28] N. Rocha, C. Kuijl, R. van der Kant, L. Janssen, D. Houben, H. Janssen, W. Zwart, J.
Neefjes, Cholesterol sensor ORP1L contacts the ER protein VAP to control Rab7RILP-p150 glued and late endosome positioning, J. Cell Biol. 185 (2009) 1209–1225.
[29] X. Du, J. Kumar, C. Ferguson, T.A. Schulz, Y.S. Ong, W. Hong, W.A. Prinz, R.G. Parton,
A.J. Brown, H. Yang, A role for oxysterol-binding protein-related protein 5 in
endosomal cholesterol trafficking, J. Cell Biol. 192 (2011) 121–135.
[30] F. Alpy, A. Rousseau, Y. Schwab, F. Legueux, I. Stoll, C. Wendling, C. Spiegelhalter, P.
Kessler, C. Mathelin, M.C. Rio, et al., STARD3 or STARD3NL and VAP form a novel molecular tether between late endosomes and the ER, J. Cell Sci. 126 (2013) 5500–5512.
[31] N.L. Cianciola, D.J. Greene, R.E. Morton, C.R. Carlin, Adenovirus RIDalpha uncovers a
novel pathway requiring ORP1L for lipid droplet formation independent of NPC1,
Mol. Biol. Cell 24 (2013) 3309–3325.
[32] M. Johansson, N. Rocha, W. Zwart, I. Jordens, L. Janssen, C. Kuijl, V.M. Olkkonen, J.
Neefjes, Activation of endosomal dynein motors by stepwise assembly of Rab7RILP-p150Glued, ORP1L, and the receptor betalll spectrin, J. Cell Biol. 176 (2007)
459–471.
[33] F. Alpy, V.K. Latchumanan, V. Kedinger, A. Janoshazi, C. Thiele, C. Wendling, M.C.
Rio, C. Tomasetto, Functional characterization of the MENTAL domain, J. Biol.
Chem. 280 (2005) 17945–17952.
[34] F. Alpy, C. Tomasetto, START ships lipids across interorganelle space, Biochimie 96
(2014) 85–95.
[35] J. Chung, F. Torta, K. Masai, L. Lucast, H. Czapla, L.B. Tanner, P. Narayanaswamy, M.R.
Wenk, F. Nakatsu, P. De Camilli, INTRACELLULAR TRANSPORT. PI4P/
phosphatidylserine countertransport at ORP5- and ORP8-mediated ER-plasma
membrane contacts, Science 349 (2015) 428–432.
[36] 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 349 (2015) 432–436.
[37] X. Pan, P. Roberts, Y. Chen, E. Kvam, N. Shulga, K. Huang, S. Lemmon, D.S. Goldfarb,
Nucleus–vacuole junctions in Saccharomyces cerevisiae are formed through the direct interaction of Vac8p with Nvj1p, Mol. Biol. Cell 11 (2000) 2445–2457.
[38] S.D. Kohlwein, S. Eder, C.S. Oh, C.E. Martin, K. Gable, D. Bacikova, T. Dunn, Tsc13p is
required for fatty acid elongation and localizes to a novel structure at the nuclear–
vacuolar interface in Saccharomyces cerevisiae, Mol. Cell. Biol. 21 (2001) 109–125.
[39] T.P. Levine, S. Munro, Dual targeting of Osh1p, a yeast homologue of oxysterolbinding protein, to both the Golgi and the nucleus-vacuole junction, Mol. Biol.
Cell 12 (2001) 1633–1644.
[40] A. Toulmay, W.A. Prinz, A conserved membrane-binding domain targets proteins
to organelle contact sites, J. Cell Sci. 125 (2012) 49–58.
[41] W.M. Henne, L. Zhu, Z. Balogi, C. Stefan, J.A. Pleiss, S.D. Emr, Mdm1/Snx13 is a novel
ER–endolysosomal interorganelle tethering protein, J. Cell Biol. 210 (2015)
541–551.
[42] A. Murley, R.D. Sarsam, A. Toulmay, J. Yamada, W.A. Prinz, J. Nunnari, Ltc1 is an ERlocalized sterol transporter and a component of ER–mitochondria and ER–vacuole
contacts, J. Cell Biol. 209 (2015) 539–548.
[43] L.C. Gomes, G. Di Benedetto, L. Scorrano, During autophagy mitochondria elongate,
are spared from degradation and sustain cell viability, Nat. Cell Biol. 13 (2011)
589–598.
[44] A.S. Rambold, B. Kostelecky, N. Elia, J. Lippincott-Schwartz, Tubular network formation protects mitochondria from autophagosomal degradation during nutrient
starvation, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 10190–10195.
[45] G. Pekkurnaz, J.C. Trinidad, X. Wang, D. Kong, T.L. Schwarz, Glucose regulates mitochondrial motility via Milton modification by O-GlcNAc transferase, Cell 158
(2014) 54–68.
[46] K. Inoki, H. Ouyang, T. Zhu, C. Lindvall, Y. Wang, X. Zhang, Q. Yang, C. Bennett, Y.
Harada, K. Stankunas, et al., TSC2 integrates Wnt and energy signals via a coordinated phosphorylation by AMPK and GSK3 to regulate cell growth, Cell 126
(2006) 955–968.
[47] D.M. Gwinn, D.B. Shackelford, D.F. Egan, M.M. Mihaylova, A. Mery, D.S. Vasquez,
B.E. Turk, R.J. Shaw, AMPK phosphorylation of raptor mediates a metabolic checkpoint, Mol. Cell 30 (2008) 214–226.
[48] S.G. Kim, G.R. Hoffman, G. Poulogiannis, G.R. Buel, Y.J. Jang, K.W. Lee, B.Y. Kim, R.L.
Erikson, L.C. Cantley, A.Y. Choo, et al., Metabolic stress controls mTORC1 lysosomal
localization and dimerization by regulating the TTT-RUVBL1/2 complex, Mol. Cell
49 (2013) 172–185.
[49] B. Westermann, Mitochondrial fusion and fission in cell life and death, Nat. Rev.
Mol. Cell Biol. 11 (2010) 872–884.
[50] A.F. MacAskill, T.A. Atkin, J.T. Kittler, Mitochondrial trafficking and the provision of
energy and calcium buffering at excitatory synapses, Eur. J. Neurosci. 32 (2010)
231–240.
[51] A. Quintana, C. Schwindling, A.S. Wenning, U. Becherer, J. Rettig, E.C. Schwarz, M.
Hoth, T cell activation requires mitochondrial translocation to the immunological
synapse, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 14418–14423.
[52] E. Abarca-Rojano, S. Muniz-Hernandez, M.M. Moreno-Altamirano, R. MondragonFlores, F. Enriquez-Rincon, F.J. Sanchez-Garcia, Re-organization of mitochondria
at the NK cell immune synapse, Immunol. Lett. 122 (2009) 18–25.
[53] Z. Li, K. Okamoto, Y. Hayashi, M. Sheng, The importance of dendritic mitochondria
in the morphogenesis and plasticity of spines and synapses, Cell 119 (2004)
873–887.
[54] C. Lopez-Crisosto, R. Bravo-Sagua, M. Rodriguez-Pena, C. Mera, P.F. Castro, A.F.
Quest, B.A. Rothermel, M. Cifuentes, S. Lavandero, ER-to-mitochondria miscommunication and metabolic diseases, Biochim. Biophys. Acta 1852 (2015)
2096–2105.
893
[55] C. Giorgi, S. Missiroli, S. Patergnani, J. Duszynski, M.R. Wieckowski, P. Pinton,
Mitochondria-associated membranes: composition, molecular mechanisms,
and physiopathological implications, Antioxid. Redox Signal. 22 (2015)
995–1019.
[56] D. De Stefani, A. Raffaello, E. Teardo, I. Szabo, R. Rizzuto, A forty-kilodalton protein
of the inner membrane is the mitochondrial calcium uniporter, Nature 476 (2011)
336–340.
[57] J.M. Baughman, F. Perocchi, H.S. Girgis, M. Plovanich, C.A. Belcher-Timme, Y. Sancak,
X.R. Bao, L. Strittmatter, O. Goldberger, R.L. Bogorad, et al., Integrative genomics identifies MCU as an essential component of the mitochondrial calcium uniporter, Nature
476 (2011) 341–345.
[58] J.E. Vance, MAM (mitochondria-associated membranes) in mammalian cells: lipids
and beyond, Biochim. Biophys. Acta 1841 (2013) 595–609.
[59] W.L. Miller, Steroid hormone synthesis in mitochondria, Mol. Cell. Endocrinol. 379
(2013) 62–73.
[60] G. Szabadkai, K. Bianchi, P. Varnai, D. De Stefani, M.R. Wieckowski, D. Cavagna, A.I.
Nagy, T. Balla, R. Rizzuto, Chaperone-mediated coupling of endoplasmic reticulum
and mitochondrial Ca2+ channels, J. Cell Biol. 175 (2006) 901–911.
[61] O.M. de Brito, L. Scorrano, Mitofusin 2 tethers endoplasmic reticulum to mitochondria, Nature 456 (2008) 605–610.
[62] P. Cosson, A. Marchetti, M. Ravazzola, L. Orci, Mitofusin-2 independent juxtaposition
of endoplasmic reticulum and mitochondria: an ultrastructural study, PLoS One 7
(2012), e46293.
[63] R. Filadi, E. Greotti, G. Turacchio, A. Luini, T. Pozzan, P. Pizzo, Mitofusin 2 ablation increases endoplasmic reticulum–mitochondria coupling, Proc. Natl. Acad. Sci. U. S. A.
112 (2015) E2174–E2181.
[64] R. Iwasawa, A.L. Mahul-Mellier, C. Datler, E. Pazarentzos, S. Grimm, Fis1 and Bap31
bridge the mitochondria–ER interface to establish a platform for apoptosis induction, Embo J. 30 (2011) 556–568.
[65] R. Stoica, K.J. De Vos, S. Paillusson, S. Mueller, R.M. Sancho, K.F. Lau, G. VizcayBarrena, W.L. Lin, Y.F. Xu, J. Lewis, et al., ER-mitochondria associations are regulated by the VAPB-PTPIP51 interaction and are disrupted by ALS/FTD-associated TDP43, Nat. Commun. 5 (2014) 3996.
[66] K.J. De Vos, G.M. Morotz, R. Stoica, E.L. Tudor, K.F. Lau, S. Ackerley, A. Warley, C.E.
Shaw, C.C. Miller, VAPB interacts with the mitochondrial protein PTPIP51 to regulate
calcium homeostasis, Hum. Mol. Genet. 21 (2012) 1299–1311.
[67] B. Kornmann, E. Currie, S.R. Collins, M. Schuldiner, J. Nunnari, J.S. Weissman, P.
Walter, An ER–mitochondria tethering complex revealed by a synthetic biology
screen, Science 325 (2009) 477–481.
[68] T.C. Swayne, C. Zhou, I.R. Boldogh, J.K. Charalel, J.R. McFaline-Figueroa, S. Thoms, C.
Yang, G. Leung, J. McInnes, R. Erdmann, et al., Role for cER and Mmr1p in Anchorage of mitochondria at sites of polarized surface growth in budding yeast, Curr.
Biol. 21 (2011) 1994–1999.
[69] S. Lahiri, J.T. Chao, S. Tavassoli, A.K. Wong, V. Choudhary, B.P. Young, C.J. Loewen,
W.A. Prinz, A conserved endoplasmic reticulum membrane protein complex
(EMC) facilitates phospholipid transfer from the ER to mitochondria, PLoS Biol.
12 (2014), e1001969.
[70] A.P. AhYoung, J. Jiang, J. Zhang, X. Khoi Dang, J.A. Loo, Z.H. Zhou, P.F. Egea, Conserved SMP domains of the ERMES complex bind phospholipids and mediate tether assembly, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) E3179–E3188.
[71] C. Honscher, M. Mari, K. Auffarth, M. Bohnert, J. Griffith, W. Geerts, M. van der Laan,
M. Cabrera, F. Reggiori, C. Ungermann, Cellular metabolism regulates contact sites
between vacuoles and mitochondria, Dev. Cell 30 (2014) 86–94.
[72] Y. Elbaz-Alon, E. Rosenfeld-Gur, V. Shinder, A.H. Futerman, T. Geiger, M. Schuldiner, A
dynamic interface between vacuoles and mitochondria in yeast, Dev. Cell 30 (2014)
95–102.
[73] T. Daniele, I. Hurbain, R. Vago, G. Casari, G. Raposo, C. Tacchetti, M.V. Schiaffino, Mitochondria and melanosomes establish physical contacts modulated by mfn2 and
involved in organelle biogenesis, Curr. Biol. 24 (2014) 393–403.
[74] C. Stroupe, K.M. Collins, R.A. Fratti, W. Wickner, Purification of active HOPS complex reveals its affinities for phosphoinositides and the SNARE Vam7p, Embo J.
25 (2006) 1579–1589.
[75] Y. Elbaz-Alon, M. Eisenberg-Bord, V. Shinder, S.B. Stiller, E. Shimoni, N. Wiedemann, T.
Geiger, M. Schuldiner, Lam6 regulates the extent of contacts between organelles, Cell
Rep. 12 (2015) 7–14.
[76] A.T. Gatta, L.H. Wong, Y.Y. Sere, D.M. Calderon-Norena, S. Cockcroft, A.K. Menon,
T.P. Levine, A new family of StART domain proteins at membrane contact sites
has a role in ER–PM sterol transport, Elife 4 (2015).
[77] A.B. Lang, A.T. John Peter, P. Walter, B. Kornmann, ER-mitochondrial junctions can
be bypassed by dominant mutations in the endosomal protein Vps13, J. Cell Biol.
210 (2015) 883–890.
[78] M. Binda, M.P. Peli-Gulli, G. Bonfils, N. Panchaud, J. Urban, T.W. Sturgill, R. Loewith,
C. De Virgilio, The Vam6 GEF controls TORC1 by activating the EGO complex, Mol.
Cell 35 (2009) 563–573.
[79] F. Dubouloz, O. Deloche, V. Wanke, E. Cameroni, C. De Virgilio, The TOR and EGO
protein complexes orchestrate microautophagy in yeast, Mol. Cell 19 (2005)
15–26.
[80] T. Daniele, M.V. Schiaffino, Organelle biogenesis and interorganellar connections:
better in contact than in isolation, Commun. Integr. Biol. 7 (2014), e29587.
[81] M.S. Marks, H.F. Heijnen, G. Raposo, Lysosome-related organelles: unusual compartments become mainstream, Curr. Opin. Cell Biol. 25 (2013) 495–505.
[82] G. Raposo, D. Tenza, D.M. Murphy, J.F. Berson, M.S. Marks, Distinct protein sorting
and localization to premelanosomes, melanosomes, and lysosomes in pigmented
melanocytic cells, J. Cell Biol. 152 (2001) 809–824.
[83] A. Sitaram, M.S. Marks, Mechanisms of protein delivery to melanosomes in pigment cells, Physiology (Bethesda) 27 (2012) 85–99.
894
T. Daniele, M.V. Schiaffino / Biochimica et Biophysica Acta 1861 (2016) 880–894
[84] G. Innamorati, R. Piccirillo, P. Bagnato, I. Palmisano, M.V. Schiaffino, The
melanosomal/lysosomal protein OA1 has properties of a G protein-coupled receptor, Pigment Cell Res. 19 (2006) 125–135.
[85] I. Palmisano, P. Bagnato, A. Palmigiano, G. Innamorati, G. Rotondo, D. Altimare, C.
Venturi, E.V. Sviderskaya, R. Piccirillo, M. Coppola, et al., The ocular albinism type
1 protein, an intracellular G protein-coupled receptor, regulates melanosome
transport in pigment cells, Hum. Mol. Genet. 17 (2008) 3487–3501.
[86] M.V. Schiaffino, Signaling pathways in melanosome biogenesis and pathology, Int.
J. Biochem. Cell Biol. 42 (2010) 1094–1104.
[87] M.J. Hoogduijn, N.P. Smit, A. van der Laarse, A.F. van Nieuwpoort, J.M. Wood, A.J.
Thody, Melanin has a role in Ca2+ homeostasis in human melanocytes, Pigment
Cell Res. 16 (2003) 127–132.
[88] R. Salceda, G. Sanchez-Chavez, Calcium uptake, release and ryanodine binding in
melanosomes from retinal pigment epithelium, Cell Calcium 27 (2000) 223–229.
[89] W.D. Bush, J.D. Simon, Quantification of Ca(2+) binding to melanin supports the
hypothesis that melanosomes serve a functional role in regulating calcium homeostasis, Pigment Cell Res. 20 (2007) 134–139.
[90] C.A. Lamb, T. Yoshimori, S.A. Tooze, The autophagosome: origins unknown, biogenesis complex, Nat. Rev. Mol. Cell Biol. 14 (2013) 759–774.
[91] S.T. Shibutani, T. Yoshimori, A current perspective of autophagosome biogenesis,
Cell Res. 24 (2014) 58–68.
[92] K. Suzuki, Y. Ohsumi, Current knowledge of the pre-autophagosomal structure
(PAS), FEBS Lett. 584 (2010) 1280–1286.
[93] K. Suzuki, M. Akioka, C. Kondo-Kakuta, H. Yamamoto, Y. Ohsumi, Fine mapping of
autophagy-related proteins during autophagosome formation in Saccharomyces
cerevisiae, J. Cell Sci. 126 (2013) 2534–2544.
[94] E.L. Axe, S.A. Walker, M. Manifava, P. Chandra, H.L. Roderick, A. Habermann, G.
Griffiths, N.T. Ktistakis, Autophagosome formation from membrane compartments
enriched in phosphatidylinositol 3-phosphate and dynamically connected to the
endoplasmic reticulum, J. Cell Biol. 182 (2008) 685–701.
[95] M. Hayashi-Nishino, N. Fujita, T. Noda, A. Yamaguchi, T. Yoshimori, A. Yamamoto, A
subdomain of the endoplasmic reticulum forms a cradle for autophagosome formation, Nat. Cell Biol. 11 (2009) 1433–1437.
[96] P. Yla-Anttila, H. Vihinen, E. Jokitalo, E.L. Eskelinen, 3D tomography reveals connections between the phagophore and endoplasmic reticulum, Autophagy 5 (2009)
1180–1185.
[97] G. Rez, J. Meldolesi, Freeze-fracture of drug-induced autophagocytosis in the
mouse exocrine pancreas, Lab. Investig. 43 (1980) 269–277.
[98] M. Baba, M. Osumi, Y. Ohsumi, Analysis of the membrane structures involved in
autophagy in yeast by freeze-replica method, Cell Struct. Funct. 20 (1995)
465–471.
[99] M. Fengsrud, E.S. Erichsen, T.O. Berg, C. Raiborg, P.O. Seglen, Ultrastructural characterization of the delimiting membranes of isolated autophagosomes and
amphisomes by freeze-fracture electron microscopy, Eur. J. Cell Biol. 79 (2000)
871–882.
[100] B. Ravikumar, K. Moreau, L. Jahreiss, C. Puri, D.C. Rubinsztein, Plasma membrane
contributes to the formation of pre-autophagosomal structures, Nat. Cell Biol. 12
(2010) 747–757.
[101] D.W. Hailey, A.S. Rambold, P. Satpute-Krishnan, K. Mitra, R. Sougrat, P.K. Kim, J.
Lippincott-Schwartz, Mitochondria supply membranes for autophagosome biogenesis during starvation, Cell 141 (2010) 656–667.
[102] J. Biazik, P. Yla-Anttila, H. Vihinen, E. Jokitalo, E.L. Eskelinen, Ultrastructural relationship of the phagophore with surrounding organelles, Autophagy 11 (2015)
439–451.
[103] N. Dupont, S. Chauhan, J. Arko-Mensah, E.F. Castillo, A. Masedunskas, R. Weigert, H.
Robenek, T. Proikas-Cezanne, V. Deretic, Neutral lipid stores and lipase PNPLA5
contribute to autophagosome biogenesis, Curr. Biol. 24 (2014) 609–620.
[104] M. Hamasaki, N. Furuta, A. Matsuda, A. Nezu, A. Yamamoto, N. Fujita, H. Oomori, T.
Noda, T. Haraguchi, Y. Hiraoka, et al., Autophagosomes form at ER-mitochondria
contact sites, Nature 495 (2013) 389–393.
[105] T. Simmen, J.E. Aslan, A.D. Blagoveshchenskaya, L. Thomas, L. Wan, Y. Xiang, S.F.
Feliciangeli, C.H. Hung, C.M. Crump, G. Thomas, PACS-2 controls endoplasmic reticulum–mitochondria communication and Bid-mediated apoptosis, Embo J. 24
(2005) 717–729.
[106] S. Bockler, B. Westermann, Mitochondrial ER contacts are crucial for mitophagy in
yeast, Dev. Cell 28 (2014) 450–458.
[107] T. Kanki, K. Wang, Y. Cao, M. Baba, D.J. Klionsky, Atg32 is a mitochondrial protein
that confers selectivity during mitophagy, Dev. Cell 17 (2009) 98–109.
[108] K. Okamoto, N. Kondo-Okamoto, Y. Ohsumi, Mitochondria-anchored receptor
Atg32 mediates degradation of mitochondria via selective autophagy, Dev. Cell
17 (2009) 87–97.
[109] F. Reggiori, D.J. Klionsky, Autophagic processes in yeast: mechanism, machinery
and regulation, Genetics 194 (2013) 341–361.
[110] Y. Guo, K.R. Cordes, R.V. Farese Jr., T.C. Walther, Lipid droplets at a glance, J. Cell Sci.
122 (2009) 749–752.
[111] A.D. Barbosa, D.B. Savage, S. Siniossoglou, Lipid droplet–organelle interactions:
emerging roles in lipid metabolism, Curr. Opin. Cell Biol. 35 (2015) 91–97.
[112] V. Choudhary, N. Ojha, A. Golden, W.A. Prinz, A conserved family of proteins facilitates nascent lipid droplet budding from the ER, J. Cell Biol. 211 (2015) 261–271.
[113] F. Wilfling, H. Wang, J.T. Haas, N. Krahmer, T.J. Gould, A. Uchida, J.X. Cheng, M.
Graham, R. Christiano, F. Frohlich, et al., Triacylglycerol synthesis enzymes mediate
lipid droplet growth by relocalizing from the ER to lipid droplets, Dev. Cell 24
(2013) 384–399.
[114] A. Kassan, A. Herms, A. Fernandez-Vidal, M. Bosch, N.L. Schieber, B.J. Reddy, A.
Fajardo, M. Gelabert-Baldrich, F. Tebar, C. Enrich, et al., Acyl–CoA synthetase 3 promotes lipid droplet biogenesis in ER microdomains, J. Cell Biol. 203 (2013)
985–1001.
[115] C.S. Shaw, D.A. Jones, A.J. Wagenmakers, Network distribution of mitochondria and
lipid droplets in human muscle fibres, Histochem. Cell Biol. 129 (2008) 65–72.
[116] H. Wang, U. Sreenivasan, H. Hu, A. Saladino, B.M. Polster, L.M. Lund, D.W. Gong,
W.C. Stanley, C. Sztalryd, Perilipin 5, a lipid droplet-associated protein, provides
physical and metabolic linkage to mitochondria, J. Lipid Res. 52 (2011) 2159–2168.
[117] A.S. Rambold, S. Cohen, J. Lippincott-Schwartz, Fatty acid trafficking in starved
cells: regulation by lipid droplet lipolysis, autophagy, and mitochondrial fusion dynamics, Dev. Cell 32 (2015) 678–692.
[118] T. Shpilka, E. Welter, N. Borovsky, N. Amar, M. Mari, F. Reggiori, Z. Elazar, Lipid
droplets and their component triglycerides and steryl esters regulate
autophagosome biogenesis, Embo J. 34 (2015) 2117–2131.
[119] 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.
[120] B.B. Chu, Y.C. Liao, W. Qi, C. Xie, X. Du, J. Wang, H. Yang, H.H. Miao, B.L. Li, B.L. Song,
Cholesterol transport through lysosome-peroxisome membrane contacts, Cell 161
(2015) 291–306.
[121] G.S. Gavelis, S. Hayakawa, R.A. White 3rd, T. Gojobori, C.A. Suttle, P.J. Keeling, B.S.
Leander, Eye-like ocelloids are built from different endosymbiotically acquired
components, Nature 523 (2015) 204–207.
[122] A.P. Arruda, B.M. Pers, G. Parlakgul, E. Guney, K. Inouye, G.S. Hotamisligil, Chronic
enrichment of hepatic endoplasmic reticulum–mitochondria contact leads to mitochondrial dysfunction in obesity, Nat. Med. 20 (2014) 1427–1435.
[123] E. Tubbs, P. Theurey, G. Vial, N. Bendridi, A. Bravard, M.A. Chauvin, J. Ji-Cao, F.
Zoulim, B. Bartosch, M. Ovize, et al., Mitochondria-associated endoplasmic reticulum membrane (MAM) integrity is required for insulin signaling and is implicated
in hepatic insulin resistance, Diabetes 63 (2014) 3279–3294.
[124] S. Raychaudhuri, W.A. Prinz, The diverse functions of oxysterol-binding proteins,
Annu. Rev. Cell Dev. Biol. 26 (2010) 157–177.
[125] T. Vihervaara, M. Jansen, R.L. Uronen, Y. Ohsaki, E. Ikonen, V.M. Olkkonen, Cytoplasmic oxysterol-binding proteins: sterol sensors or transporters? Chem. Phys. Lipids
164 (2011) 443–450.
[126] V.M. Olkkonen, S. Li, Oxysterol-binding proteins: sterol and phosphoinositide sensors coordinating transport, signaling and metabolism, Prog. Lipid Res. 52 (2013)
529–538.
[127] K. Maeda, K. Anand, A. Chiapparino, A. Kumar, M. Poletto, M. Kaksonen, A.C. Gavin,
Interactome map uncovers phosphatidylserine transport by oxysterol-binding proteins, Nature 501 (2013) 257–261.
[128] 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.
[129] B. Mesmin, J. Bigay, J. Moser von Filseck, S. Lacas-Gervais, G. Drin, B. Antonny, A
four-step cycle driven by PI(4)P hydrolysis directs sterol/PI(4)P exchange by the
ER-Golgi tether OSBP, Cell 155 (2013) 830–843.
[130] M. Khafif, L. Cottret, C. Balague, S. Raffaele, Identification and phylogenetic analyses
of VASt, an uncharacterized protein domain associated with lipid-binding domains
in eukaryotes, BMC Bioinf. 15 (2014) 222.
[131] I. Lee, W. Hong, Diverse membrane-associated proteins contain a novel SMP domain, FASEB J. 20 (2006) 202–206.
[132] 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.
[133] 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 (2014).
[134] 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)P(2)-dependent
and Ca(2+)-regulated ER-PM interactions mediated by the extended
synaptotagmins, Cell 153 (2013) 1494–1509.
[135] 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.