Diss. ETH No. 14766
Caveolae-mediated endocytosis of Simian Virus 40
A dissertation submitted to the
Swiss Federal Institute of Technology Zürich
for the degree of
Doctor of Natural Sciences
presented by
Lucas Lodewijk Pelkmans
Drs. University of Utrecht
born on September 2nd, 1975
citizen of The Netherlands
accepted on the recommendation of
Prof. Dr. Ari Helenius, examiner
Prof. Dr. Urs Greber, co-examiner
Prof. Dr. Ulrike Kutay, co-examiner
Prof. Dr. Gaudenz Danuser, co-examiner
2002
Cover:
Fluorescent Simian Virus 40 particles in membranous tubules moving along
microtubule tracks (see page 56)
Contents
iii
Contents
Zusammenfassung
vi
Summary
vii
1. Introduction: The many faces of endocytosis
1
1.1. Multiple types of endocytosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
1.2. Phagocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
1.2.1. Receptors and their activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
1.2.2. Transduction of the phagocytic signal. . . . . . . . . . . . . . . . . . . . . . . .5
1.2.3. Phosphoinositides in phagocytosis . . . . . . . . . . . . . . . . . . . . . . . . . .6
1.2.4. Small GTPases of the Rho family in phagocytosis. . . . . . . . . . . . . .9
1.2.5. Actin dynamics in phagocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . .10
1.2.6. Sealing the phagosome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
1.2.7. Scaffolds in phagocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14
1.3. Clathrin-mediated endocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.3.1. Formation and internalization of clathrin-coated vesicles. . . . . .16
1.3.2. Regulation of clathrin-coated vesicle formation. . . . . . . . . . . . . . 21
1.3.3. Clathrin-coated vesicle formation and the actin cytoskeleton. . .23
1.4. Destinations after internalization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.4.1. Sorting in the endocytic system. . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.4.2. Rab GTPases in the endocytic pathway. . . . . . . . . . . . . . . . . . . . . .26
1.4.3. Linking endocytic and biosynthetic pathways. . . . . . . . . . . . . . . .28
1.5. Clathrin-independent endocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
1.5.1. Macropinocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.5.2. Evidence for caveolae-mediated endocytosis. . . . . . . . . . . . . . . . .31
1.5.3. Caveolae as sites of Simian Virus 40 entry? . . . . . . . . . . . . . . . . . .32
1.6. Outline of this thesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2. Caveolar endocytosis of Simian Virus 40 reveals a new two-step
vesicular transport pathway to the ER
37
2.1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.2. Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.2.1. Texas Red-labeled SV40 and caveolin-1-GFP behave
iv
Caveolae-mediated endocytosis of SV40
normally. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41
2.2.2. After binding to the cell surface , SV40 moves into
stationary caveolae in the membrane and is internalized. . . . . . 45
2.2.3. Entry into caveosomes – intermediate organelles in
caveolar endocytosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46
2.2.4. SV40 is sorted from caveosomes. . . . . . . . . . . . . . . . . . . . . . . . . . . .50
2.2.5. Entry of SV40 is a two-step process. . . . . . . . . . . . . . . . . . . . . . . . .53
2.2.6. After sorting, SV40-containing carriers travel along
microtubules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.2.7. After sorting, SV40 rapidly accumulates in the smooth ER . . . . 55
2.3. Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59
2.4. Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62
2.5. Movies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68
3. Local actin polymerization and dynamin recruitment in
SV40-induced internalization of caveolae
73
3.1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.2. Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.2.1. Internalization of SV40 depends on active kinases,
an intact actin cytoskeleton and dynamin2. . . . . . . . . . . . . . . . . . .77
3.2.2. Caveolae-sequestered SV40 particles are entrapped
in the membrane by the actin cytoskeleton . . . . . . . . . . . . . . . . . . 81
3.2.3. Caveolae-sequestered SV40 induces the formation of
actin tails. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .82
3.2.4. SV40-induced actin tail formation depends on
functional caveolae and active tyrosine kinases. . . . . . . . . . . . . . .85
3.2.5. Depolymerization/polymerization of actin and transient
recruitment of dynamin2 during SV40 internalization. . . . . . . . .86
3.3. Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .91
3.4. Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93
3.5. Movies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97
4. Endocytosis via caveolae
101
4.1. Caveolae. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Contents
v
4.2. Caveolar entry of Simian Virus 40. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .106
4.2.1. Sequestration and internalization. . . . . . . . . . . . . . . . . . . . . . . . . .107
4.2.2. Transport to caveosomes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .109
4.2.3. Molecular sorting and transport to the ER. . . . . . . . . . . . . . . . . . 111
4.3. Caveolar endocytosis of other ligands and membrane
constituents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
4.4. How does it work? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
4.5. Perspectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5. General conclusions
119
Abbreviations
125
Bibliography
127
Acknowledgements
149
Curriculum Vitae
150
List of publications
151
vi
Caveolae-mediated endocytosis of SV40
Zusammenfassung
Endozytose ist ein essentieller zellulärer Prozess, der die Erhaltung der zellulären
Homeostase, die Aufnahme von Nährstoffen und die Abwehr von Mikroorganismen
ermöglicht. Von den verschiedenen bekannten Formen der Endozytose sind die
Prozesse der Phagozytose und der Clathrin-vermittelten Endozytose auf molekularer
Ebene detailliert charakterisiert. Im Gegensatz dazu weiss man sehr wenig über
andere Arten der Endozytose. Da viele Krankheitserreger, unter ihnen auch Viren,
die verschiedenen Endozytosewege für ihren Eintritt in die Zelle ausnutzen, stellen
sie exzellente Hilfsmittel zur Erforschung dieser Prozesse dar.
Simian Virus 40, ein doppelsträngiges DNA Virus der Papova Virusfamilie, dessen
nur ansatzweise beschriebener Aufnahmeweg von Caveolae an der Zelloberfläche
zum glatten endoplasmatischen Retikulum führt, wurde hier benutzt, um diese Form
der Endozytose detaillierter zu charakterisieren. Die Analyse des Viruseintritts in
lebenden Zellen zeigte, dass internalisierte Caveolae sich zuerst in einem neuen,
intrazellulären Kompartiment ansammeln, welches Caveosom genannt wurde.
Caveosomen konnten als prä-existierende Organellen definiert werden, die reich an
Caveolin-1 und einigen ,,lipid-raft’’-Bestandteilen waren und einen neutralen pH
aufwiesen, während sie frei waren von Markern des biosynthetischen und des
klassischen Endozytosewegs. Von Caveosomen aus wurden die Viruspartikel in
tubuläre Strukturen sortiert, die kein Caveolin-1 mehr aufwiesen, und entlang von
Mikrotubuli zum glatten endoplasmatischen Retikulum transportiert.
Untersuchungen der frühen Aufnahmestadien zeigten, dass die Viruspartikel nach
ihrer Bindung in Caveolae eine vorübergehende Depolymerisierung von Aktin
Stressfasern induzierten. Aktin wurde an virusbeladene Caveolae rekrutiert, wo es
zu kleinen Plattformen organisiert wurde, an denen dann Aktin „Schweife“ gebildet
wurden. Dynamin2 wurde ebenfalls vorübergehend an virusbeladene Caveolae
rekrutiert. Diese virus-induzierten Ereignisse waren abhängig von Cholesterin, sowie
von der Aktivierung von Tyrosinkinasen zur Phosphorylierung von CaveolaeProteinen. Nur so war die Abschnürung endozytotischer Vesikel aus Caveolae und
deren Transport zu Caveosomen, sowie eine virale Infektion möglich.
Zusammengefasst ist Endozytose von Caveolae ein Ligand-vermittelter Prozess,
welcher eine weitreichende Reorganisation des Aktin Zytoskeletts beinhaltet und
einen neuartigen zweistufigen Transportweg von der Plasmamembran über
Caveosomen zum endoplasmatischen Retikulum beschreibt. Der Aufnahmeweg
umgeht Endosomen und Golgi-Apparat und ist Teil der infektiösen Route von SV40.
Summary
vii
Summary
Endocytosis is an essential cellular process that allows the maintenance of cellular
homeostasis, the uptake of nutrients and other substances, the transduction of intraand intercellular signals and an effective defense against microorganisms.
Endocytosis occurs in many different forms. Of these, the processes of phagocytosis
and clathrin-mediated endocytosis have been characterized in molecular detail.
Little is known about other types of endocytosis. Since many pathogens, including
animal viruses, have evolved ways to exploit different endocytic pathways in order
to gain entry into the host cell, they provide excellent tools to study endocytosis.
Simian Virus 40 (SV40), a double stranded DNA virus of the papova virus family,
utilizes a poorly characterized uptake pathway via cell surface caveolae to the
smooth endoplasmic reticulum (ER). It was used in this thesis as a tool to
characterize this clathrin-independent endocytic pathway in more detail.
Following the entry of virus particles in living cells, it was found that the virus
induces the internalization of caveolae, and that the internalized caveolae deliver
their cargo to a previously unknown intracellular organelle that was termed the
caveosome. Caveosomes were found to be pre-existing, enriched in caveolin-1 and
several other lipid-raft constituents and to contain a neutral pH. They were devoid of
markers from the classical endocytic pathway or biosynthetic compartments. From
caveosomes, virus particles were sorted into tubular carriers lacking caveolin-1 that
were transported along microtubules to the smooth endoplasmic reticulum.
Studies of the early stages of uptake showed that after binding to caveolae, the virus
particles induced a transient breakdown of actin stress fibers. Actin was recruited to
the virus-loaded caveolae as small actin patches that served as sites for the formation
of actin ‘tails’. Dynamin2 was also transiently recruited to virus-loaded caveolae.
These virus-triggered events depended on the presence of cholesterol and on the
activation of tyrosine kinases that phosphorylated proteins in caveolae. They were
necessary for formation of caveolae-derived endocytic vesicles and their transport to
caveosomes, as well as for virus infection.
Taken together, caveolar endocytosis is a ligand-triggered event that involves
extensive rearrangement of the actin cytoskeleton. It defines a new, two-step
transport pathway from plasma membrane caveolae, via caveosomes to the
endoplasmic reticulum. The pathway by-passes endosomes and the Golgi complex,
and is part of the productive infectious route used by SV40.
Chapter 1
Introduction: The many faces of endocytosis
Lucas Pelkmans
Institute of Biochemistry, Swiss Federal Institute of Technology (ETH) Zürich
2
Caveolae-mediated endocytosis of SV40
1.1. Multiple types of endocytosis
Cells have evolved a variety of mechanisms to internalize extracellular
particles and solutes by invagination of the plasma membrane and
detachment of vesicles and vacuoles. These mechanisms are collectively
known as endocytosis, and the intracellular pathways taken by the incoming
membrane and cargo as endocytic pathways.
Endocytosis is a major activity of all eukaryotic cells: The amount of
membrane internalized by animal cells such as macrophages can reach more
than twice the surface area per hour. Besides the maintenance of cellular
homeostasis by regulating the composition of the plasma membrane though
selective internalization, the major function of endocytosis is the uptake of
nutrients. It occurs via the internalization of nutrient-bound receptors that are
in general rapidly recycled back to the plasma membrane. Furthermore,
endocytosis is necessary for the transmission of neuronal, metabolic and
proliferative signals, for the regulation of intercellular communication, and
for an effective defense against invading microorganisms. Paradoxically,
many pathogens have evolved ways to exploit endocytosis to gain entry into
their host cells and to exert their infectious or toxic activity.
Endocytic processes have historically been divided into two general types: the
internalization of large (>0.5 µm) particles was termed ‘phagocytosis’ and the
uptake of fluid and small particles and molecules was termed ‘pinocytosis’.
The first evidence that cells have the capacity to engulf and internalize
particles came from experiments performed by Metchnikoff in the 1880s with
amoeboid cells and blue litmus particles. He coined the term ‘phagocytosis’
(Metchnikoff, 1887). The major form of pinocytosis was discovered in the
1960s by Roth and Porter using the electron microscope. They described the
uptake of yolk proteins into mosquito oocytes via 80 nm ‘bristle-coated’
membrane vesicles (Roth and Porter, 1964). After purification of similarlooking vesicles from brain extracts, they were named clathrin-coated vesicles,
according to the major protein constituent clathrin that forms the coat (Pearse,
1975). During the same period, other types of pinocytosis were observed that
Chapter 1. Introduction: The many faces of endocytosis
3
did not appear to use clathrin-coated vesicles. Therefore, pinocytosis was
further divided into a clathrin-mediated and clathrin-independent subtype.
The term clathrin-independent endocytosis stands for pinocytic activity that
occurs independently of clathrin, and clathrin-mediated endocytosis stands
for pinocytic activity that is mediated by clathrin. Both types of endocytosis
internalize extracellular fluid and both perform receptor-mediated
endocytosis, namely the selective uptake of membrane-associated receptors
and their potential cargo by concentrating them in the internalizing
structures. Whereas clathrin-mediated uptake occurs constitutively in all cells,
this is not always the case for clathrin-independent uptake mechanisms.
Over the past 40 years, phagocytosis and clathrin-mediated endocytosis have
been well characterized and the machinery appears to be distinct from each
other. The apparent mechanistic difference may, however, be caused by the
fact that phagocytosis has often been approached from a cell-signaling point
of view, while clathrin-mediated uptake has been studied from a more
biomechanical angel. As will be clear, recent studies indicate that the
mechanisms may in fact overlap to a significant extent, especially where the
actin cytoskeleton is involved.
The events happening at the plasma membrane of most mammalian cells,
including the different kinds of endocytosis, are affected by the dense, plasma
membrane-associated cortical actin cytoskeleton. Since this is understood in
detail for the mechanism of phagocytosis, it provides many important clues
for other uptake mechansisms that are still poorly characterized. Furthermore,
several important factors of the phagocytosis machinery are now emerging as
key molecules in the regulation of clathrin-mediated endocytosis as well.
Therefore, this introduction starts with an extensive discussion of
phagocytosis.
4
Caveolae-mediated endocytosis of SV40
1.2. Phagocytosis
Phagocytosis has four defining features: 1) It describes the internalization of
large particles. 2) It is generally associated with specialized cells such as
phagocytic protozoa (Dictyostelium, Acanthamoeba) and phagocytic leukocytes
of the mammalian immune system (macrophages, neutrophils). 3) It is
induced by the binding of the particle to cell surface receptors, which
transduces a phagocytic stimulus to the cytoplasm. 4) It involves the actin
cytoskeleton to engulf the bound particle into an intracellular phagosome
(Greenberg et al., 1990; Greenberg et al., 1991).
Beyond these common features, the detailed mechanisms are variable: Each
particle type induces its own specific process. The phagocytic signal is
transduced by a variety of different receptors, that induce slightly different
signal transduction cascades (Underhill and Ozinsky, 2002). In an attempt to
categorize different forms of phagocytosis, early classifications used the
morphological appearance of particle engulfment. According to this criterion,
phagocytosis can be divided into three main categories: 1) The classical zipper
model, in which pseudopod extensions tightly encapsulate the particle via
increasing numbers of particle-cell surface receptor interactions. One example
of this is Fcγ-receptor-mediated phagocytosis in macrophages and
polymorphonuclear leukocytes. 2) The sinking model, in which the particle
sinks into the phagocytic cell, and no pseudopods are formed. This is the case
in complement-receptor-mediated (e.g. integrin-mediated) phagocytosis. 3)
The ruffling model, where large membrane ruffles engulf extensive amounts
of extracellular material including the particle (Aderem and Underhill, 1999;
May and Machesky, 2001; Underhill and Ozinsky, 2002). This process is
similar to macropinocytosis, and is induced by intracellular parasites such as
Salmonella (Amer and Swanson, 2002).
Below, the mechanism of phagocytosis is summarized mainly according to
the best-understood type of phagocytosis, namely Fcγ-receptor-mediated
phagocytosis.
Chapter 1. Introduction: The many faces of endocytosis
5
1.2.1. Receptors and their activation during phagocytosis
Upon binding of opsonized (IgG or complement) particles to receptors in the
membrane of a macrophage a phagocytic signal is generated. Receptors
capable of generating such a signal upon binding of a particle include Fcγreceptors, complement receptors and other integrins, scavenger receptors,
lectins and Toll-like receptors (Underhill and Ozinsky, 2002). Immediately
after binding of a particle, Fcγ-receptors undergo a conformational change
that exposes immunoreceptor tyrosine-based activation motifs (ITAM) on the
cytosolic site (May and Machesky, 2001) (fig. 1). At the same time, several Src
protein tyrosine kinases (Src PTKs) are activated. The activation of multiple
Src PTKs provides redundancy to the system. Activation is mediated by dephosphorylation of a tyrosine residue in the C-terminal tail of the Src PTK,
which abolishes the intra-molecular interaction between the phosphorylated
tyrosine and a Src Homology 2 (SH2) domain, thus exposing the kinase
domain (Thomas and Brugge, 1997).
How phagocytic receptor-binding leads to Src PTK-activation, is not
understood, but it involves the activity of the tyrosine phosphatase CD45
(Adamczewski et al., 1995; Thomas and Brugge, 1997). Upon activation, Src
PTKs rapidly phosphorylate the exposed tyrosines in the ITAM domains of
the Fcγ-receptors, which now provide high-affinity binding sites for the
exposed SH2 domains of Src PTKs (Thomas and Brugge, 1997). Thus, upon
binding of an opsonized particle to phagocytic receptors, active ITAM-Src
PTK domains are instantly formed on the cytosolic site of the plasma
membrane. As will become clear, these serve as launching pads for the
transduction of the phagocytic signal.
1.2.2. Transduction of the phagocytic signal
Besides the Src PTKs, several other SH2-domain containing proteins are now
recruited to the phosphorylated tyrosines in the ITAM motifs, where they are
rapidly activated by the Src PTKs. The protein tyrosine kinase Syk plays a
particularly important role (Darby et al., 1994; Greenberg et al., 1994;
Greenberg et al., 1996). Activated Syk attenuates the phagocytic signal by
further phosphorylating Fcγ-receptors (Kiefer et al., 1998) and is, together
with Src PTKs, the principal transducer of the phagocytic signal (fig. 1).
6
Caveolae-mediated endocytosis of SV40
Transduction is achieved by phosphorylation of a broad range of recruited
substrates, including PI3-kinases (PI3-K), Phospholipase C (PLC), Protein
Kinase C (PKC), regulatory enzymes of the Rho family of GTPases, adaptor
proteins (Shc, Cbl, DOK), focal adhesion proteins (Focal Adhesion Kinase or
FAK, paxillin, talin, tensin) and other actin cytoskeletal proteins (cortactin,
ezrin) (Aderem and Underhill, 1999; Thomas and Brugge, 1997; Underhill and
Ozinsky, 2002). The resulting cascade of protein activations, protein-lipid and
protein-protein interactions leads to the formation of a complex and highly
controlled phagocytosis machinery (Underhill and Ozinsky, 2002). The
existence of multiple feed-forward and feedback mechanisms makes it
difficult to describe the exact chronological sequence of events. Therefore, the
progression of phagocytosis is here presented along the sequential action of
three main groups of molecules: 1) phosphoinositides, 2) small GTPases of the
Rho and ARF family, and 3) molecules involved in actin dynamics.
Fig. 1. Overview of the early steps in the activation of the phagocytic signal
After binding of an opsonized particle to Fcγ-Rs, ITAM motifs on the cytosolic side of the Fcγ-R are
rapidly exposed and Src-PTKs are activated by dephosphorylation of a phosphotyrosine in the C-term.
domain, leading to opening of the SrcPTK. The exposed Src kinase domain now rapidly
phosphorylates tyrosines in the ITAM motifs. Phosphorylated ITAM motifs recruit Syk and a plethora
of other SH2-domain containing proteins, including the adaptors Grb2, Nck and Shc. SrcPTKs and Syk
phosphorylate PI-kinases, PKC, phospholipases and RhoGTPase regulators, leading to transduction of
the phagocytic signal. Phosphorylation of focal adhesion proteins and other actin-binding proteins
recruit actin beneath the particle (see fig. 2).
1.2.3. Phosphoinositides in phagocytosis
Phosphoinositides have recently emerged as key molecules in the
coordination of many membrane trafficking events, including phagocytosis
(Martin, 2001; Simonsen et al., 2001). They are phosphorylation variants of the
Chapter 1. Introduction: The many faces of endocytosis
7
phospholipid phosphatidylinositol (PI), located in the cytosolic leaflet of
membranes where they serve as docking and activation sites for a variety of
proteins. Because different protein domains have different affinity for the
various phosphoinositides, phosphoinositide phosphorylation and
dephosphorylation provides an excellent mechanism to specifically recruit
proteins to membrane subdomains. Indeed, PI(4)P5-kinases and PI3-kinases
(Botelho et al., 2000; Araki et al., 1996) and possibly also PI4-kinases are
rapidly recruited and/or activated below the particle upon activation of a
phagocytic signal, where they regulate highly localized changes in the levels
of the different phosphoinositides. Negative regulation of the amount of
phosphoinositides is mainly accomplished by the action of Phospholipase C
(Botelho et al., 2000).
The first phosphoinositide kinase identified to play a crucial role in
phagocytosis was a PI3-kinase (PI3-K) (Araki et al., 1996) which
phosphorylates phosphoinositides on position D-3 of the inositol ring. Some
isoforms of PI3-K are recruited to ITAM motifs on Fcγ-receptors via their SH2
domain and can be activated by Syk and Src PTKs (Chacko et al., 1996). Their
enzyme activity is especially required for membrane extension and closure of
the phagosome (Cox et al., 1999). This might be connected to the importance
of 3’-polyphosphoinositides in recycling of endosomes (Gruenberg, 2001;
Simonsen et al., 2001), which could contribute to membrane availability at the
phagocytic cup. Furthermore, PI3-Ks can interact with the Rho GTPases
Cdc42 and Rac1 (Keely et al., 1997) that play an important role in actin
polymerization and membrane ruffling (see below). Finally, Dynamin2 and an
isoform of amphiphysin, well known to regulate a variety of membrane
scission events, have been identified as downstream effectors of PI3-K during
phagocytosis (Gold et al., 1999).
PI(4)P5-kinase (PI(4)P-5K) phosphorylates the D-5 position of PI(4)P, resulting
in the formation of PI(4,5)P2. Early in phagocytosis the levels of PI(4,5)P2 in the
membrane of the phagocytic cup are elevated (Botelho et al., 2000). These
PI(4,5)P2 domains recruit proteins that contain Pleckstrin Homology (PH)
domains, Epsin N-terminal homology (ENTH) domains and Lys/Arg-rich
effector domains (Martin, 2001). Proteins that contain such domains include
8
Caveolae-mediated endocytosis of SV40
the myristoylated alanine-rich C kinase substrate protein (MARCKS) (Laux et
al., 2000), Phospholipase C, ADP Ribosylation Factor GTPases (ARF GTPases)
(Donaldson and Jackson, 2000), proteins of the Wiskott Aldrich Syndrome
Protein family (WASP/N-WASP) (Higgs and Pollard, 2001)
and
Dynamin2/amphiphysinIIm (Gold et al., 2000).
Although PI4-kinases (PI4-K) have not been directly studied in the process of
phagocytosis, they must play a crucial role because they synthesize PI(4)P, the
precursor of PI(4,5)P2, by phosphorylation of PI on position D-4 . PI4-Ks have,
however, been studied in the context of secretion (Martin, 2001; Simonsen et
al., 2001), where they are recruited to the trans-Golgi network (TGN) by ARF
GTPases. The resulting 4’-polyphosphoinositide domains on the TGN
membrane directly interact with several components of the vesicular
transport machinery required for secretory vesicle budding, including coat
proteins and actin-interacting proteins. ARF6, a plasma membrane-specific
type III ARF GTPase involved in membrane ruffling and phagocytosis (see
below), is the prime candidate to recruit PI4-Ks to the phagocytic cup.
During the progression of the phagocytic event, the amount of PI(4,5)P2 below
the particle decreases. This results mainly from the action of PLC which
hydrolyzes PI(4,5)P2 into inositol triphosphate (IP3) and diacylglycerol (DAG)
(fig. 3). IP3 binds to the IP 3 -receptor/Ca2+ -channels located in the ER
membrane, which leads to rapid Ca2+ release from the ER into the cytosol, and
DAG activates Protein Kinase C (PKC), a serine/threonine kinase. Both a high
level of intracellular Ca2 + and the activation of PKC, regulate actin
polymerization as follows: The major substrate of PKC is MARCKS which is
recruited to the membrane during phagocytosis through strong interactions
with PI(4,5)P2 domains, cross-links F-actin, and thus links actin filaments to
the membrane (Hartwig et al., 1992; Laux et al., 2000). Both phosphorylation
of MARCKS by PKC and Ca2+-mediated binding of calmodulin to MARCKS,
results in dissociation of MARCKS from PI(4,5)P2 domains, and inhibits its
actin-crosslinking ability. High intracellular Ca2+ levels further reduce actin
polymerization by activating gelsolin, a protein that severs actin filaments
and caps the fast-growing plus end (so-called barbed end) of the actin
Chapter 1. Introduction: The many faces of endocytosis
9
filament, thereby breaking up the cross-linked actin network (Harris and
Weeds, 1984; Yin et al., 1981).
1.2.4. Small GTPases of the Rho family in phagocytosis
A second group of molecules that plays an important role in phagocytosis
consists of monomeric GTPases, especially the members of the Rho family of
GTPases, of which Cdc42, RhoA and Rac1 are the best-characterized
members. These are key regulators of the actin cytoskeleton during processes
such as cell adhesion, cell spreading, membrane ruffling and stress fiber
formation (Bishop and Hall, 2000). However, they now emerge as more
general regulators of cell proliferation, gene transcription and secretion. Early
studies using dominant-negative mutants showed that Rac1 and Cdc42 play
an important role in Fcγ-receptor mediated phagocytosis and RhoA in
complement-receptor mediated phagocytosis (Caron and Hall, 1998). Similar
to other small GTPases, they cycle between an inactive GDP-bound and an
active GTP-bound form, which is mediated by specific GDP/GTP exchange
factors (RhoGEFs) and GTPase activating proteins (RhoGAPs). Several
RhoGEFs and RhoGAPs have been identified which are regulated by many
signaling pathways, including growth factor receptor signaling, adhesiondependent signaling and cell cycle progression (Bishop and Hall, 2000). Until
now no direct link between the phagocytic signal and the activation of
RhoGEFs or RhoGAPs has been found, but RhoGEFs are likely recruited to
the PI(4,5)P2-enriched membrane of the phagocytic cup via their PH domains
present on all currently identified RhoGEFs (Bishop and Hall, 2000).
Active Rho GTPases have a wide variety of downstream effector proteins,
including many Ser/Thr kinases, the previously mentioned phosphoinositide
kinases (i.e. PI3-kinases, PI(4)5-kinases), phospholipases (phospholipase C
and D) and many adaptor proteins involved in actin organization (Bishop and
Hall, 2000). However, not all of these are directly involved in phagocytosis.
During Fcγ-receptor mediated phagocytosis, active Rac1 and Cdc42 bind to
proteins with exposed Cdc42/Rac Interactive Binding (CRIB) domains which
are present on several Focal Adhesion Kinases (FAKs) (Bagrodia and Cerione,
1999) and on WASP family members (Machesky and Insall, 1998). Especially
the role of WASP/N-WASP proteins in actin polymerization has been
10
Caveolae-mediated endocytosis of SV40
characterized in detail (see below) (fig. 2). Furthermore, the Rac1 effector
protein Por-1 activates ARF6 (D'Souza-Schorey et al., 1997), providing a link
between two different families of GTPases. Activated ARF6 functions through
the activation of a plasma membrane-associated phospholipase D (PLD)
which catalyzes the production of phosphatidic acid (PA) from membrane
phospholipids (fig. 2). PA can then recruit several actin-interacting proteins
and activates a type I PI(4)P5-K and perhaps a PI4-K, establishing an entry
point into the phosphoinositide system (Honda et al., 1999).
RhoA is well known for its role in the formation of focal adhesion contacts in
response to growth factors (Bishop and Hall, 2000). Focal adhesions are
complex structures that attach the actin cytoskeleton to the substratum or to
other cells via integrins. Because complement receptors are integrins and
RhoA plays an important role in complement-receptor mediated
phagocytosis, the detailed understanding of focal adhesions might provide
clues for some forms of phagocytosis. RhoA, but also Cdc42 and Rac1, are
believed to regulate focal adhesion assembly through several FAKs, which are
effector proteins of the Rho GTPases (Bishop and Hall, 2000). Indeed, several
FAK substrates have been shown to accumulate on the phagocytic cup during
phagocytosis. These include talin, α-actinin, vinculin and paxillin (fig. 2)
(Allen and Aderem, 1996).
1.2.5. Actin dynamics in phagocytosis
After the recruitment of several actin-interacting molecules to domains
beneath the particle, actin polymerization is the next step in the phagocytic
process (fig. 2). The main actin nucleator in cells is the Actin related protein
2/3 complex (Arp2/3 complex) which consists of 7 subunits (Higgs and
Pollard, 2001) and is recruited to the phagocytic cup by activated WASP
family members (Lorenzi et al., 2000). WASP/N-WASP is activated by Cdc42
and PI(4,5)P2, which bind synergistically to WASP/N-WASP, resulting in a
partial unfolding of WASP/N-WASP, by which a binding site for the Arp2/3
complex is revealed in the C-terminus (Rohatgi et al., 1999). The Arp2/3
complex subsequently binds monomeric actin and stimulates the formation of
new actin polymers (Higgs and Pollard, 2001). Another mechanism of Arp2/3
recruitment involves activated Rac1, which binds, via IRSp53, the
Chapter 1. Introduction: The many faces of endocytosis
11
WAVE/Scar protein that is related to WASP/N-WASP (Machesky and Insall,
1998; Machesky et al., 1999). Furthermore, some myosins can recruit both
Arp2/3 and cortactin, an actin filament binding protein (May and Machesky,
2001), but this has not been studied in detail. Coronin, a protein that binds Factin and is recruited to the phagocytic cup, might be capable of inducing
actin polymerization independent of Arp2/3 (Goode et al., 1999). Also cofilin,
an actin severing and nucleating protein is involved.
Fig. 2. Overview of transduction of some phagocytic signals to the actin cytoskeleton.
Activated Rac1 can activate PLD via ARF6, which results in the production of phosphatidic acid (PA)
from phospholipids. Src-, Syk- and PA-activated PI-kinases produce PI(4,5)P2 which, together with
activated Cdc42 and Fcγ-R-recruited Grb2 and Nck, binds to N-WASP (Cdc42 to the Cdc42/Rac1
interacting domain or CRIB). This leads to membrane recruitment of N-WASP and the exposure of the
Arp2/3 binding domain. The Arp2/3 complex nucleates the formation of new actin filaments (F-actin)
and cross-links existing actin filaments, as does cortactin. Actin filaments are attached to the
membrane via MARCKS which binds actin and PI(4,5)P2. Opsonized-particle-attached integrins
which recruit several focal adhesion proteins such as the actin-binding protein talin to the site of
phagocytosis, couple the force of actin polymerization to particle enwrapment. This is further
stabilized by the PI(4,5)P2-regulated and talin- and actin-binding protein vinculin.
Besides initial actin nucleation, actin bundling and cross-linking also takes
place. Similar to the actin network observed in lamellipodia, actin filaments
12
Caveolae-mediated endocytosis of SV40
underneath the phagocytic cup appear as a ‘dendritic’ network (Mullins et al.,
1998). This is established by the cross-linking abilities of the Arp2/3 complex
itself, α-actinin and cortactin (Allen and Aderem, 1996). The actin-interacting
proteins in focal adhesions are believed to link the force generated by the
actin cytoskeleton to the bound particle, leading to the actual engulfment of
the particle (May and Machesky, 2001). Excellent candidates for this are talin
which binds integrins and F-actin and links complement-receptors to the actin
cytoskeleton and vinculin which binds to talin and F-actin and is regulated by
PI(4,5)P2 (fig. 2) (Allen and Aderem, 1996). Paxillin which has multiple
binding partners, most prominently vinculin, might translate the actingenerated force to particle engulfment during Fcγ-receptor mediated
phagocytosis, where it is heavily phosphorylated (Greenberg et al., 1994;
Underhill and Ozinsky, 2002).
Besides actin polymerization and cross-linking, actin severing and
depolymerization play important roles at the end of the phagocytic process.
This is mediated by a reduction in PI(4,5)P2 levels, the activation of PLC and
PKC, and probably the action of PI3-K, which all leads to a reduced molecular
scaffold density on the membrane leading to actin dissociation (fig. 3)
(Aderem and Underhill, 1999; May and Machesky, 2001; Underhill and
Ozinsky, 2002). Proteins that are directly involved in actin severing and
depolymerization are the previously mentioned gelsolin and cofilin. These
proteins sever actin filaments and cap their free ends, thereby inhibiting
further polymerization.
1.2.6. Sealing the phagosome
Relatively little is known about the completion of phagocytosis leading to an
internalized phagosome. However, the activity of PI3-Ks, actin
depolymerization (described above), and the action of dynamin2 and
amphiphysinIIm are crucial. As the pseudopods extend through progressive
interactions of Fcγ-receptors with the particle, or as the particle is dragged
into the cell in the case of complement-receptor mediated phagocytosis, the
membranes meet at the top of the particle where they fuse. PI3-Ks are
believed to play a role here because their inhibition does not interfere with
pseudopod extension but prevents completion of phagocytosis (Araki et al.,
Chapter 1. Introduction: The many faces of endocytosis
13
1996), perhaps through upregulation of membrane recycling or secretion.
Dynamin2 and amphiphysinIIm are believed to provide the mechanical force
needed to seal the phagosome (Gold et al., 2000; Gold et al., 1999), similar to
their role in pinching off clathrin-coated pits (see below), but this has not yet
been proven.
Fig. 3. Some steps involved in actin removal from internalized phagosomes
After internalization, activated PLC hydrolyzes PI(4,5)P2 into IP3 and DAG. IP3 activates Ca2+-pumps
that rapidly increase the cytosolic Ca2+ concentration. DAG activates PKC, which phosphorylates
MARCKS. Together with Ca2+-bound calmodulin which binds to MARCKS this leads to a rapid
detachment of MARCKS from the membrane and F-actin. Exposed fast-growing ends of F-actin are
capped by Ca2+-activated gelsolin. As a result, F-actin depolymerizes and is removed from the
phagosome.
In general, it is believed that when the phagosome is released into the cytosol,
actin filaments rapidly depolymerize (May and Machesky, 2001). This view
has been in part challenged by recent studies which showed that actin
polymerization continues (Defacque et al., 2000), leading to the formation of
an actin ‘tail’ that might propel the phagosome further into the cell, as
observed on endocytic vesicles and macropinosomes (Merrifield et al., 1999;
Rozelle et al., 2000). The continued polymerization of actin is not completely
14
Caveolae-mediated endocytosis of SV40
understood but might involve other proteins than those that mediate the
initial actin polymerization. Actin polymerization at this stage requires
membrane proteins of the ezrin/radixin/moesin (ERM) family that link actin
filaments to membranes (Defacque et al., 2000). Annexins, which accumulate
on phagosomes as they mature by interacting with the endocytic pathway,
can also link actin to membranes (Desjardins et al., 1994; Kaufman et al.,
1996).
1.2.7. Scaffolds in phagocytosis
As is clear from above, phagocytosis involves several multi-protein scaffold
complexes. Phagocytosis might thus be regarded as a good example of a
scaffold-organized event, where Phosphotyrosine/srcPTK-based scaffolds,
phosphoinositide-based scaffolds and actin-based scaffolds sequentially play
a role. Recently, another multi-protein scaffold system, consisting of so-called
adaptor proteins, has been implicated in phagocytosis (Izadi et al., 1998; May
and Machesky, 2001). Adaptor proteins (Cbl, Nck, Grb2 among others)
contain multiple SH2 and SH3 domains which respectively bind
phosphotyrosine-rich sequences (pYXXL) and proline-rich sequences (PXXP)
(Thomas and Brugge, 1997). Because such sequences are particularly present
in the molecules of the different scaffolds, these adaptor proteins could
function as signal integrators, linking the different scaffolds functionally to
each other. How this might work has been characterized in the case of Pak1
(Bagrodia and Cerione, 1999). Upon recruitment by Rac1 or Cdc42, Pak1 binds
to Nck and subsequently to cofilin and myosin which can remodel the actin
cytoskeleton. Furthermore, Nck and Grb2 can interact with WASP/N-WASP
and trigger activation of the Arp2/3 complex (fig. 2). Nck also interacts with
RhoA via Prk2 (Quilliam et al., 1996). Thus, the adaptor proteins form a multiprotein signaling scaffold that collects the signals from multiple upstream
sources and translates it to the actin cytoskeleton.
At first sight, the involvement of all these scaffolds raises the complexity of
potential downstream interactions to a near-indecipherable point. This may
explain why a comprehensive model of phagocytosis has still not been put
forward, although the number of molecules and molecular interactions
implicated is huge. Maybe development of a comprehensive model requires
Chapter 1. Introduction: The many faces of endocytosis
15
the identification and characterization of all molecules and interactions
involved. This, however, is not likely to happen in the near future, and it may,
in cases, merely fill in the remaining gaps without providing new conceptual
insights. Computer-aided, systematic integration of the signaling cascades,
scaffolds and actin polymerization machineries identified today, may in the
end be necessary to establish a comprehensive model for phagocytosis.
Likely, this would allow a more targeted search for missing factors.
16
Caveolae-mediated endocytosis of SV40
1.3. Clathrin-mediated endocytosis
In most animal cell types and under normal conditions, the uptake of
receptor-bound ligands and extracellular fluid results mainly from the
formation of clathrin-coated vesicles (CCVs) (Mellman, 1996). The defining
feature of this endocytic process is that during internalization, a specialized
protein coat is assembled at the cytoplasmic leaflet of the plasma membrane.
It induces the formation of an invagination or clathrin coated pit (CCP), and
subsequently a CCV. The coat consists of polymers of the cytoplasmic protein
clathrin and several associated proteins. The formation of the CCV is a multistep process, orchestrated by many cellular proteins in a highly regulated
manner. Currently, the interactions between many of these proteins are being
analyzed by X-ray crystallography and NMR spectroscopy (Kirchhausen,
2000; Marsh and McMahon, 1999).
1.3.1. Formation and internalization of clathrin-coated vesicles
A present understanding of the basic sequence of events leading to the
internalization of a CCV is summarized below (see figures 4 and 5 for an
overview). In general, the recruitment of so-called adaptors to the cytosolic
site of the membrane is regarded as the first step in the formation of a CCV
(Brodsky et al., 2001). In contrast to earlier models, it is now clear that the
initial adaptor recruitment occurs irrespective of the presence of
transmembrane receptors. Adaptors are heterotetrameric complexes, existing
in several combinations called AP1, AP2, AP3 and AP4. While AP1, AP3 and
AP4 are involved in transport from the trans-Golgi network (TGN), AP2 is
involved in clathrin-mediated endocytosis at the plasma membrane. AP2
consists of two large (~100 kDa) subunits, α and β2, and two smaller chains,
µ2 (50 kDa) and σ2 (20 kDa). AP2 associates with the plasma membrane via
the α-subunit that has a high affinity for phosphoinositides, and binds
clathrin via the β2-subunit, thereby recruiting clathrin to the membrane
(Kirchhausen, 1999).
When encountering a trans-membrane receptor that contains a low-affinity
endocytic sorting signal (tyrosine- or lysine-based sorting signals, such as
NPXY, YXXΦ , LL; Φ stands for a hydrophobic residue), AP2 binds to this
Chapter 1. Introduction: The many faces of endocytosis
17
receptor (The YXXΦ motif is bound by the µ2-subunit, binding to the other
motifs is less clear) (Brodsky et al., 2001). A trans-membrane protein called
synaptotagmin stabilizes this low affinity complex and serves as a ‘docking
protein’ (Haucke and De Camilli, 1999). This membrane-associated complex,
stabilized by multivalent protein-protein and protein-lipid interactions, is
now ready to initiate clathrin assembly (fig. 4).
Besides synaptotagmin, other adaptor-like molecules have been identified.
These usually interact with AP2 or clathrin, and frequently with both. They
include adaptor protein 180 (AP180) (McMahon, 1999), EGF receptor pathway
substrate 15 (Eps15) (de Beer et al., 1998; Wendland and Emr, 1998; Wendland
et al., 1998), and arrestins (Ferguson, 2001). AP180 likely plays a general role
in CCV formation by cooperating with AP2 to achieve maximal efficiency of
coat assembly. Eps15 contains an N-terminal region with EH (Eps15
homology) domains, a central coiled-coil domain involved in Eps15
oligomerization, and a C-terminal AP2 binding domain. Via the EH domains,
Eps15 interacts with Epsin. Both Eps15 and Epsin are crucial for clathrindependent endocytosis of several receptors and ligands (Chen et al., 1998), but
their precise function is not yet clear. They might play a role in AP2
recruitment to the plasma membrane or be involved in the dynamic
rearrangement of the coat (see below) during pit invagination and fission
(Brodsky et al., 2001). Arrestins specifically link certain G protein coupled
receptors (GPCR) to CCVs.
After AP2 complexes are in place and initial clathrin monomers are recruited,
the ability of clathrin to assemble into closed basket-like structures is believed
to be the driving force behind membrane deformation into an invaginated pit
(Kirchhausen, 2000). This assembly reaction is regulated through
dephosphorylation/phosphorylation of the β2-subunit of AP2 (Brodsky et al.,
2001). When AP2 is recruited to the membrane, a serine residue in the
clathrin-binding domain is dephosphorylated by an unknown serine
phosphatase. Only then, clathrin is nucleated and the assembly reaction
occurs.
18
Caveolae-mediated endocytosis of SV40
Fig. 4. Basic steps in the formation of a clathrin-coated vesicle.
AP2-complexes are recruited to the membrane via PI(4,5)P2 and synaptotagmin. Once positioned on
the membrane, AP2-complexes can recruit clathrin triskelions, regulated by poorly understood
phosphorylation and dephosphorylation events. Cargo receptors might be incorporated into this
complex via association with the µ2 subunit of AP2. The recruitment of clathrin triskelions leads to the
rapid formation of a clathrin coat that drives the invagination process. Once the coated pit is deeply
invaginated, a complex series of events takes place that finally leads to the formation of an internalized
Chapter 1. Introduction: The many faces of endocytosis
19
Fig. 4. continued
vesicle. This event is orchestrated by dynamin which binds with its PH-domains to PI(4,5)P2 in the neck
region. The PRDs of dynamin now recruit several proteins involved in cell signaling, including PKC,
PLC and SrcPTKs. Through amphiphysin, dynamin is linked to AP2. Intersectin, a modular protein with
five SH3-domains and two EH-domains, can establish a scaffold onto which many interacting proteins
can dock, including Epsin and Eps15 that might regulate clathrin coat deformation. Strong membrane
curvature can be introduced by dynamin-recruited endopholin which converts cone-shaped into inverse
cone-shaped lipids. Through profilin and cortactin, dynamin recruits F-actin which can assemble into an
actin tail (see fig. 5).
The functional units of a clathrin coat are trimers made of dimers of a heavy
(180 kDa) and a light (30-35 kDa) chain. These have the typical triskelion
appearance, and can self-assemble into empty cages in vitro and in the absence
of ATP. In living cells, however, clathrin coats are often observed on the
plasma membrane as large, flat, hexagonal lattices. Whether invaginated pits
are formed from these lattices or de novo is not clear. In order to form a pit, at
least 12 pentameric orientations are needed. That such drastic rearrangements
take place within a lattice seems energetically unfavorable. In agreement with
this, the assembly of a CCP in living cells is energy-dependent (Schmid and
Carter, 1990) and involves ATP-consuming chaperones (see further). AP180
and Eps15 have also been implicated in this process although their function is
unclear (Brodsky et al., 2001). Furthermore, it has been proposed that the
hexagonal clathrin lattices function as stores for clathrin, and that pits are
formed de novo at the edges of these lattices (Kirchhausen, 2000). Clearly, the
exact sequence of events leading to the assembly of a clathrin coat on a CCP in
living cells is not known and likely awaits the development of higher
resolution live imaging techniques.
Once CCPs are deeply invaginated and the clathrin coat is almost complete,
the large GTPase dynamin is recruited to the CCPs via PI(4,5)P2-domains and
the AP2- and clathrin-binding protein amphiphysin (fig. 4) (De Camilli et al.,
1995; Hinshaw, 2000; Sever et al., 2000). Dynamin releases the invaginated
coated pits from the membrane. It has been suggested that it acts as a
‘pinchase’ by self-assembling into a collar around the neck of an invaginated
coated pit which constricts upon GTP hydrolysis. This has been re-evaluated
and it seems now more likely that dynamin acts as a switch to regulate
downstream effectors that mediate the scission step (Fish et al., 2000).
20
Caveolae-mediated endocytosis of SV40
The Proline Rich domain (PRD) of dynamin is responsible for effectorrecruitment. Among these are G-proteins, amphiphysin, endophilin,
intersectin, syndapin, Grb2, cortactin, Src PTKs, PI3-Ks, PLCγ and profilin
(Hinshaw, 2000; Huttner and Schmidt, 2000). G-proteins recruit dynamin
when their internalization is induced, Grb2, Src PTKs, PI3-Ks and PLCγ link
dynamin to signaling systems as described in the section on phagocytosis
and, together with syndapin, cortactin and profilin, to the actin cytoskeleton.
Intersectin serves as a cross-linker of Epsin, Eps15 and dynamin.
Amphiphysin activates PLD that hydrolyses phospholipids generating
phosphatidic acid which in turn activates PI(4)P-5Ks that generate PI(4,5)P2.
Endophilin exhibits lysophosphatidic acid acyl transferase activity and can
convert an inverted cone-shaped lipid into a cone-shaped lipid in the
cytoplasmic leaflet of the bilayer. This may facilitate the formation of a strong
membrane curvature observed at the neck of a CCP (Huttner and Schmidt,
2000). Clearly, dynamin is a central player in the late events of CCV
formation.
As the CCV is internalized, the vesicle is rapidly uncoated by the action of the
ATPase Hsc70 and auxilin (fig. 5). Hsc70 is a chaperone of the Hsp70 family
and hydrolyses ATP during removal of clathrin. Auxilin contains a
chaperone-like DnaJ domain and has phosphatase activity. Hsc70 binds to
clathrin and the DnaJ domain of auxilin (Newmyer and Schmid, 2001), and
auxilin binds to AP1, AP2 and clathrin (Lemmon, 2001).
Another protein that functions in the release of clathrin from CCVs is
synaptojanin. Synaptojanin interacts with several dynamin effectors,
including syndapin, Grb2, amphiphysin and intersectin, as well as with Eps15
and clathrin (Haffner et al., 2000; Harris et al., 2000). Although the exact
mechanism of uncoating is unclear, it is likely that the chaperone activities of
Hsc70 and auxilin destabilize the clathrin coat and that the phosphatase
activities of auxilin and synaptojanin result in the dissociation of the adaptor
proteins AP2 and AP180 from lipids (Cremona et al., 1999; Newmyer and
Schmid, 2001). Furthermore, phosphorylation of AP2 on the clathrin-binding
domain of the β-subunit plays an important role as well (Wilde and Brodsky,
1996). Upon dissociation of the clathrin coat, the small GTPase rab5 is rapidly
21
Chapter 1. Introduction: The many faces of endocytosis
recruited to an unknown receptor that has been already incorporated in the
CCV (Zerial and McBride, 2001). This is necessary to mediate the targeted
fusion of the uncoated vesicle with early endosomes (see later).
Fig. 5. Overview of the
uncoating of a clathrincoated vesicle.
After pinching off, a dynamin-,
cortactin- and Hip1R recruited
actin tail on the CCV pushes
the CCV through the dense
cortical actin cytoskeleton into
the cytosol. There, the
chaperones auxilin and Hsc70
rapidly bind to the CCV and,
by consuming ATP, destabilize
the clathrin coat. Synaptojanin
binds to AP2 and other
components of the CCV and
dephosphorylates PI(4,5)P2,
which results in the
dissociation of PI(4,5)P2binding proteins, most
importantly AP2, from the
membrane of the CCV. After
uncoating, the membrane is
exposed to the introduction of
Rab5 by Rab-GDI. Rab5
mediates fusion of the
uncoated vesicle with the early
endosome.
1.3.2. Regulation of clathrin-coated vesicle formation
The series of events described above are subjected to extensive regulation.
Thus, although clathrin-mediated endocytosis itself occurs constitutively
(there are no external stimuli required to start CCV formation), the processes
themselves, i.e. recruitment of cargo and the assembly of the clathrin-coat, are
regulated.
22
Caveolae-mediated endocytosis of SV40
A major site of regulation is the phosphorylation/de-phosphorylation cycle of
AP2. AP2 can be phosphorylated on its α-, β2- and µ2-subunits (Brodsky et
al., 2001). The phosphatases are not known, but a recent study has identified
one of the kinases. It is a novel serine/threonine kinase of the Prk/Ark family,
which has been termed Adaptor Associated Kinase 1 (AAK1) and specifically
phosphorylates the µ2-subunit of AP2 during endocytosis (Conner and
Schmid, 2002). Because phosphorylation of µ2 by AAK1 mediates tight
binding of AP2 to endocytic sorting motifs (Ricotta et al., 2002), but
dephosphorylation is needed for internalization, it seems likely that
phosphorylation of µ2 by AAK1 allows the initial concentration of receptors
that are then rapidly internalized upon dephosphorylation by an unknown
phosphatase.
The affinity of AP2 for polyphosphoinositides allows further regulation of
clathrin assembly by phosphoinositide kinases. Indeed, it has been shown that
a type II PI3-K binds specifically to clathrin (Kirchhausen, 2000). Because
polyphosphoinositides and their kinases are located in specific membrane
microdomains, it has been suggested that AP2 is recruited to these specific
sites (Brodsky et al., 2001). Furthermore, it has been speculated that a
phosphatase resides in these membrane microdomains, which
dephosphorylates AP2 and initiates internalization. Indeed, such ‘hot spots’
for clathrin assembly have recently been found (Gaidarov et al., 1999).
In special cases, the cargo molecules themselves might regulate CCV
formation. This has been implicated for two classes of receptors, namely
tyrosine kinase receptors and GPCRs. The epidermal growth factor (EGF)
receptor is the classical example for a tyrosine kinase receptor: Upon binding
of EGF, autophosphorylation within the receptor leads to the exposure of two
consensus internalization sequences at the cytosolic site that are normally
cryptic. Eps15 and AP2 can now bind, clathrin is nucleated, and a coated pit is
assembled. GPCRs bind arrestins, which can also recruit clathrin and induce
coat assembly (Brodsky et al., 2001).
A second layer of regulation is found in the protein interactions of dynamin,
synaptojanin and amphiphysin, which are all inhibited by phosphorylation
Chapter 1. Introduction: The many faces of endocytosis
23
(Hinshaw, 2000). Furthermore, the possible molecular interactions and
signaling pathways initiated by dynamin and its effector proteins could allow
several signaling systems to regulate the completion of CCV formation. It is
likely that these show significant overlap with the regulation mechanisms
described previously in the section on phagocytosis, since several of the
involved signaling factors are the same.
1.3.3. Clathrin-coated vesicle formation and the actin cytoskeleton
In order to allow the formation of CCVs and passage into the cytosol, the
dense cortical actin network below the plasma membrane needs to be locally
rearranged. In support of this, mild treatment with actin depolymerizing
drugs enhance clathrin-mediated uptake. Harsher treatment with these drugs
reduce clathrin-mediated uptake, suggesting that an actin polymerization step
is also required (Fujimoto et al., 2000).
An active role for actin in clathrin-mediated uptake is suggested by the fact
that CCV components interact with actin-binding proteins and many
regulatory components of CCV formation appear to be involved in actin
rearrangements (figures 4 and 5). Mammalian clathrin binds ankyrin and
Hip1R. Ankyrin has a well-known function in linking actin to the plasma
membrane, but is also known to interact with microtubules, Na+- and Ca2+channels, and PKC (Rubtsov and Lopina, 2000). Hip1R, the yeast homologue
of which functions in actin-dependent endocytosis, binds F-actin and
phosphoinositides (Engqvist-Goldstein et al., 1999). Actin binds cortactin
which in turn binds dynamin (Lee and De Camilli, 2002). The dynamininteracting proteins syndapin and pacsin play a role in CCV formation and
interact with WASP/N-WASP (Qualmann and Kelly, 2000) which can
nucleate actin tails through Arp2/3. Indeed, small actin tails have been
observed on CCVs, which might allow passage through the cortical actin
network (Frischknecht and Way, 2001). Furthermore, as mentioned
previously, WASP/N-WASP is preferentially recruited to phosphoinositideenriched membrane domains, just like AP2, synaptotagmin, amphiphysin and
dynamin. Finally, actin might play a role in vesicle scission through the
interaction with amphiphysin and dynamin (Wigge and McMahon, 1998).
24
Caveolae-mediated endocytosis of SV40
1.4. Destinations after internalization
Once the uptake carriers, phagosomes as well as uncoated CCVs, have
transported their cargo into the cell, they rapidly fuse with endocytic
organelles. The endocytic pathway comprises many different compartments
and it is difficult to define true boundaries between them. A simplified, but
still astonishingly well functioning model proposes 3 main endocytic
organelles (fig. 6): 1) early endosomes (EEs), 2) recycling vesicles or recycling
endosomes (REs) and 3) late endosomes (LEs) (Mellman, 1996).
EEs are the first station in the endocytic pathway and their primary function
is sorting. All incoming ligands internalized via CCVs or phagosomes first
accumulate here. EEs are preexisting structures that contain a mildly acidic
pH (~6.5). This pH mediates rapid receptor-ligand uncoupling (Kornfeld and
Mellman, 1989) and the fusion of the envelope of some viruses with the
endosomal membrane (Helenius et al., 1980; Marsh et al., 1983; Schmid et al.,
1989). Subsequently, nutrient receptors (Transferrin receptor, LDL receptor)
that are recycled back to the surface are sorted to REs, whereas material
destined for degradation (LDL or down-regulated receptors such as EGF
receptor and certain GPCRs) is sorted to LEs that contain a more acidic pH
(<5.5). From REs, molecules are targeted to the plasma membrane and from
LEs, molecules are targeted to the lysosome (Mellman, 1996).
1.4.1. Sorting in the endocytic system
The physical separation of the different components destined for REs and LEs
is partially accomplished by a geometric mechanism. Tubular extensions that
are rich in membrane but depleted of fluid accumulate transmembrane
receptors and membrane-associated molecules, while vacuolar regions
accumulate primarily fluid phase (Geuze et al., 1987; Geuze et al., 1983).
Tubular elements are then sorted to REs and vacuolar elements are sorted to
LEs. This can, however, not be the only mechanism because several
transmembrane or membrane-associated molecules travel to LEs and
lysosomes. Indeed, several different lysosomal targeting signals have been
identified in cytosolic domains of transmembrane proteins, but it is unclear
how these mediate sorting (Gruenberg, 2001).
Chapter 1. Introduction: The many faces of endocytosis
25
Recent reports are starting to provide some insight into one such targeting
mechanism, in which the attachment of monoubiquitin provides the targeting
signal (Hicke, 2001b). Upon ubiquitination, the cytosolic domain of a protein
destined for degradation can interact with the ubiquitin-interacting motif
(UIM) of hepatocyte growth factor-regulated tyrosine kinase substrate (Hrs).
Hrs also contains a clathrin-binding motif, which associates with flat clathrin
coats present on EEs (Raiborg et al., 2002). This coat presumably prevents the
ubiquitinated proteins from recycling, which is the default pathway for
membrane proteins.
Ubiquitination also plays an important role later in the endocytic pathway,
when vesicles bud into the lumen of the LEs, giving rise to MVBs (Gruenberg,
2001; Hicke, 2001a). These vesicles within the MVBs are subsequently
degraded upon fusion of MVBs with the lysosome. Proteins that remain in the
limiting membrane of the MVBs are not degraded. Thus, sorting into these
inward budding pits is necessary for efficient degradation of transmembrane
proteins. Interestingly, the machinery involved in this process appears to
share homology with the machinery that mediates the budding of several
retroviruses at the plasma membrane, a budding process that is topologically
similar to the inward budding at the MVB (Hicke, 2001a). More intriguingly
even, it now appears that some viruses that were believed to bud only at the
plasma membrane in fact mainly bud at MVBs utilizing the MVB machinery,
and are released into the extracellular fluid by exocytosis of MVBs (M. Marsh,
unpublished results). Although the exact mechanisms have not yet been
elucidated, the ongoing characterization of proteins that contain UIMs will
likely provide more insight soon. The fact that all UIM-containing proteins
identified to date also contain other protein-binding domains such as clathrin
binding, Epsin Homology, and FYVE fingers (see later), suggests that
ubiquitination is functionally linked to specific membrane transport
machineries (Hicke, 2001a).
Another set of proteins involved in sorting in the endocytic system is the
Coatomer I complex (COPI). Similar to biosynthetic COPI, endocytic COPI is
recruited to endosomes via ARF1 but lacks the β- and γ-subunits (Gruenberg,
2001). Endocytic COPI is believed to play a particularly important role in
26
Caveolae-mediated endocytosis of SV40
transport between EEs and LEs/MVBs, but whether it functions as a true coat
similar to biosynthetic COPI, or whether it contributes to the assembly of a
scaffold that prevents certain proteins from recycling, similar to the flat
clathrin lattices on endosomes mentioned above, is not clear.
1.4.2. RabGTPases in the endocytic pathway
Extensive research on the different sorting mechanisms in the endocytic
pathway during the past years has put forward several small GTPases of the
Rab family as important regulators of these events (Gruenberg, 2001; Zerial
and McBride, 2001). Rabs are attached to the cytosolic surface of membranes
via 2 prenyl groups. Because rabs exert their functions on a specific membrane
transport pathway between two membranous compartments, their
distribution is restricted to these compartments and they can thus be
considered as ‘inter-organelle’ markers. They function as molecular switches
by recruiting effector proteins when they are in their active, GTP-bound form.
Rabs that function in the endocytic pathway are rab5 (on plasma membrane
and EEs), rab4 (on EEs and REs), rab7 (on LEs and lysosomes), rab9 (on LEs
and the Golgi complex) and rab11 (on REs and the Golgi complex) (fig. 6)
(Zerial and McBride, 2001). Rabs are now believed to recruit a large number
of different effector proteins, thereby building platforms at specific sites of the
membrane. These platforms can then specifically concentrate cargo through
interactions with scaffolding proteins, recruit necessary fusion and fission
machineries, anchor the membranes at specific sites in the cell through
interactions with the actin and microtubule cytoskeleton, and initiate
directional movements by recruiting specific motor proteins.
A good example of how this is mediated is the action of the RabGTPase rab5.
Inactive (GDP-bound) rab5 is recruited to CCVs and EEs by a specific but
unidentified rab receptor, thereby releasing rab GDP dissociation inhibitor
(GDI) from inactive rab5. On the EE membrane, rab5 is unstable and cycles
continuously between an inactive and an active (GTP-bound) form (Rybin et
al., 1996). However, upon binding of the rabaptin-5-rabex-5 complex to active
rab5, a positive feedback loop is introduced, since rabex-5 is a nucleotide
exchange factor of rab5. Rabex-5 can activate an additional rab5 molecule
Chapter 1. Introduction: The many faces of endocytosis
27
which in turn recruits an additional rabaptin-5-rabex-5 complex (Horiuchi et
al., 1997).
A second feedback loop is introduced by the cooperativity between rab5effectors. Active rab5 interacts with hVPS34-p150 PI3-K, thus coupling PI(3)P
production to rab5 localization (Christoforidis et al., 1999b). The presence of
both active rab5 and PI(3)P allows the recruitment of the effectors Early
Endosome Antigen 1 (EEA1) and rabenosyn-5, which contain FYVE fingers,
zinc-fingers that specifically bind to PI(3)P (Christoforidis et al., 1999a; De
Renzis et al., 2002). EEA1, rabaptin-5 and rabex-5 can form high-molecular
weight-oligomers on the EE membrane. As these these oligomers contain the
nucleotide exchange factor for rab5, the effectors in this way feed back on
themselves resulting in the clustering of rab5 within a limited area on the EE
membrane (Zerial and McBride, 2001). These domains may in fact be the sites
where membrane fusion/fission reactions and linkage to components of the
cytoskeleton are regulated. The EEA1-rabaptin-5-rabex-5 oligomers
incorporate N-ethyl-maleimide-sensitive fusion protein (NSF) (McBride et al.,
1999), and EEA1 can interact with two soluble NSF attachment protein
receptors (SNAREs): syntaxin6, necessary in TGN-EE transport (Simonsen et
al., 1999), and syntaxin13, required for EE fusion and recycling (McBride et al.,
1999). Furthermore, active rab5 regulates motility of early endosomes on
microtubules (Nielsen et al., 1999) and interacts with rabankyrin-5, a rab5
effector that binds to F-actin (M. Zerial, unpublished results).
Although the other rabs of the endocytic system are not as well characterized
as rab5, it is likely that these also form specific domains in the endocytic
system, contributing to the extensive sub-compartmentalization. However, a
successful coordination of recycling and sorting events requires
communication between these subdomains. Interestingly, both rabaptin-5 and
rabenosyn-5 are divalent rab-effectors that interact with rab5 and rab4. An
initial study shows that divalent rab-effectors are important in coordinating
the recycling of proteins via EE and REs back to the plasma membrane and
might thus define an additional layer of regulation within the endocytic
system (De Renzis et al., 2002).
28
Caveolae-mediated endocytosis of SV40
1.4.3. Linking endocytic and biosynthetic pathways
Hydrolytic enzymes that function in lysosomal degradation are transported
via clathrin-coated vesicles from the TGN to LEs. This is mediated by
mannose-6-phosphate receptors (MPRs) that recognize mannose-6-phosphate
moieties on a variety of enzymes involved in lysosomal degradation (Griffiths
et al., 1988). The adaptor protein complex AP1 mediates the formation of
these TGN-derived CCVs, similar to how AP2 mediates the formation of
plasma membrane-derived CCVs (Kirchhausen, 1999).
The transport route is further regulated by the action of the rabGTPase rab9
(Schimmoller et al., 1998). Since MPRs cycle between LEs and the TGN
(Mellman, 1996), it was suggested that other molecules might travel from LEs
to the TGN utilizing this bi-directional transport pathway. Evidence for this
comes from the observation that TGN38, a TGN-resident protein, cycles
rapidly to the plasma membrane and via an endocytic compartment back to
the TGN (Reaves et al., 1993). When PI3-Ks were inhibited, TGN38
accumulated in EEs/LEs, indicating that 3’-polyphosphoinositides also play
an important role in transport from the endocytic compartments to the TGN
(Chapman and Munro, 1994).
Endosome acidification and Arf1 are also believed to play a role in the EETGN transport step (Chapman and Munro, 1994; Reaves et al., 1993). Also
certain toxins, especially Shiga toxin, are taken up via clathrin-coated pits into
EEs and are retrogradely transported via the Golgi complex to the ER.
Perhaps, they use a similar route as TGN38 to reach the TGN (Falnes and
Sandvig, 2000). From there a KDEL-like signal on the toxin mediates retrieval
from the TGN back to the ER (Sandvig et al., 1992). Unfortunately, the
endosome-TGN transport route is still poorly characterized.
Chapter 1. Introduction: The many faces of endocytosis
29
1.5. Clathrin-independent endocytosis
As should be clear from above, phagocytosis and clathrin-mediated
endocytosis have been characterized in striking detail over the past 30 years.
Not only are the mechanisms of internalization understood to a large extent,
also the many ways of regulating these events have been partially revealed.
Furthermore, the itinerary after internalization is known, and the mechanisms
of molecular sorting in the endocytic pathway are being unraveled. The next
step, namely to understand these biological events on the structural level, has
already been taken and significant progress is also being made there.
In striking contrast to this stands our present understanding of alternative
uptake pathways. Although there is no doubt about their existence, we know
almost nothing concrete about them. Clearly, clathrin-mediated endocytosis
constitutes the main uptake pathway in most cell types, but alternative means
of internalization that bypass the classical endocytic routes may have
important functions in the cell. The next section focusses on these potential
alternative pathways, and tries to integrate the rather heterogeneous and
sometimes confusing present knowledge to a consistent picture.
Structures that might be involved in clathrin-independent endocytosis include
small uncoated plasma membrane invaginations present in most cell types
that resemble caveolae as originally described by Palade and Yamada in the
1950s (Palade, 1953; Yamada, 1955) and macropinosomes which are large
vacuoles derived from membrane extensions (or membrane ruffles) and
which passively engulf large amounts of fluid phase. Preliminary evidence
suggests that additional uptake pathways exist, namely those that are lipid
raft-dependent (see below) but caveolae-independent (Lamaze et al., 2001).
Furthermore, lipid-raft mediated uptake can be dynamin-dependent or
dynamin-independent (Nichols and Lippincott-Schwartz, 2001). Whether
indeed five (or more) different uptake pathways (fig. 6) exist in parallel in the
same cell is not clear. Furthermore, some of the above-mentioned uptake
mechanisms might in fact be slight variations of one theme.
30
Caveolae-mediated endocytosis of SV40
Currently, a particular uptake pathway is studied after inactivation of other
pathways, preferentially by knocking out an essential gene or overexpressing
a mutant gene. However, as recognized early on, inactivation of one pathway
can lead to rapid, compensatory stimulation of other pathways (Damke et al.,
1995). Together with the fact that specific ligands are generally lacking,
clathrin-independent pathways are difficult to study; Hence, detailed
information is scarce (Dautry-Varsat, 2000; Falnes and Sandvig, 2000; Nichols
and Lippincott-Schwartz, 2001).
Fig. 6. Summary of known and potential endocytosis routes in mammalian cells
Upon internalization via phagocytosis, macropinocytosis and clathrin-mediated endocytosis, the
internalized phagosomes, macropinosomes and CCVs fuse in a rab5-dependent manner with early
endosomes. A clathrin- and caveolae-independent internalization mechanism which can be dynamindependent or dynamin-independent supposedly also meets with early endosomes, although the
involvement of alternative structures is not ruled out. Internalized caveolae take an unknown pathway,
presumably to a subcompartment of the endoplasmic reticulum (ER). Indicated are also the major
rabGTPases involved. Rab5 regulates fusion with early endosomes (EEs), Rab4 regulates transport
Chapter 1. Introduction: The many faces of endocytosis
31
Fig. 6. continued
from EEs to recycling endosomes (REs) and rab7 regulates transport from EEs to late endosomes
(LEs). Rab11 regulates transport from REs back to the plasma membrane. LEs are thought to mature
into multivesicular bodies (MVBs) that fuse with lysosomes. A connection between the biosynthetic
route and endosomes is established by rab9 which regulates the bi-directional transport between the
Golgi complex and LEs.
1.5.1. Macropinocytosis
Macropinocytosis is restricted to a few cell types such as dendritic cells
(Steinman and Swanson, 1995), but can be induced in other cell types by
pathogenic microbes (Cardelli, 2001). Because macropinocytosis is an
aselective process, no specific cargo molecules exist. In dendritic cells, it is
involved in antigen uptake and targeting to Major Histocompatibility
Complex (MHC) class I and class II pathways. It is also utilized by several
bacterial pathogens such as Legionella and Salmonella and by Human
Immunodeficiency Virus (HIV) to enter into macrophages (Marechal, 2001).
Stimulation of cells with epidermal growth hormone also induces
macropinocytosis (Haigler, 1979). Interestingly, a recent report describes that
Adenovirus type 2 induces macropinocytosis during its infectious uptake
(Meier, 2002). Since macropinocytosis is dependent on the induction of
membrane ruffles, it is regulated by Rac1 and by ARF6, in a similar manner as
described above for certain forms of phagocytosis. In agreement with this,
macropinocytosis is also dependent on the action of PI3-Ks and PLC. Specific
inhibitors of macropinocytosis appear to be certain derivatives of amiloride
that block Na+/H+ exchange, but these reagents may also inhibit other uptake
mechanisms. Other than in cell types such as dendritic cells and NIH 3T3
fibroblasts, macropinocytosis does not appear to occur constitutively.
1.5.2. Evidence for caveolae-mediated endocytosis
Caveolae are small, flask-shaped plasma membrane invaginations of 50-80 nm
diameter observed in many cell types (Palade, 1953; Yamada, 1955). Although
caveolae might have specific functions in different cell types, in epithelial
tissue culture cells they are open to the extracellular medium and appear as
single indentations or grape-like structures (Montesano et al., 1982; Rothberg
et al., 1992). Significant progress in their characterization has been made by
the discovery of caveolins, integral membrane proteins that are enriched in
caveolae (Dupree et al., 1993; Monier et al., 1995; Rothberg et al., 1992).
Interestingly, caveolins are visible on the cytosolic surface of caveolae as part
32
Caveolae-mediated endocytosis of SV40
of parallel, shallow ridges (Peters et al., 1985; Rothberg et al., 1992), which has
led to the suggestion that caveolins might represent another type of coat
proteins, similar to clathrin and COPs. Although this was further
strengthened by the fact that no typical caveolae are observed in the absence
of caveolins (Fra et al., 1995), their role as coat proteins has never been
studied. Another defining feature of caveolae is that their lipid composition
resembles that of lipid rafts, being enriched in cholesterol and sphingolipids
(Brown and London, 1998; Simons and Toomre, 2000). Consistent with this,
caveolae resist solubilization with non-ionic detergents at 4ºC. One may thus
define caveolae as caveolin-containing plasma membrane invaginations rich
in raft lipids.
Although caveolae have long been proposed to mediate the specific
internalization and transcytosis of certain molecules (Anderson, 1998; Fielding
and Fielding, 1997; Ikonen and Parton, 2000; Kurzchalia and Parton, 1999;
Parton, 1996; Smart et al., 1999) direct evidence is lacking in most cases.
Ligands or membrane constituents that are reported to be internalized via
caveolae include cholera toxin (Montesano et al., 1982; Parton et al., 1994),
folic acid (Anderson et al., 1992; Rothberg et al., 1990), serum albumin
(Schnitzer et al., 1994), autocrine motility factor (AMF) (Benlimame et al.,
1998), and alkaline phosphatase (Parton et al., 1994).
It has been suggested that during the internalization process caveolin-1 moves
along with the vesicles into the cytosol (Parton et al., 1994; Stang et al., 1997),
but this has not been proven. Indirect evidence for this comes from the
observation that upon extraction or oxidation of plasma membrane
cholesterol, caveolins re-localize to intracellular structures that can be
endosomes, the Golgi complex, or the ER (Carozzi et al., 2000; Smart et al.,
1994). That caveolae have endocytic capacity is also suggested by the fact that
molecules involved in membrane traffic, have been found in fractions
enriched for caveolae. These include dynamin and several components of the
molecular machinery for vesicle docking and fusion (Henley et al., 1998; Oh et
al., 1998; Schnitzer et al., 1995).
Chapter 1. Introduction: The many faces of endocytosis
33
1.5.3. Caveolae as sites of Simian Virus 40 entry?
While some enveloped viruses have developed a mechanism of cell entry by
direct fusion of their envelope with the cell membrane, most viruses, both
enveloped and non-enveloped, have evolved intricate ways to hijack the host
cell endocytosis machineries in order to gain entry into the cell. This makes
them excellent tools to study these cellular processes. A major advantage of
utilizing the cell’s uptake machinery is the fact that the virus particles
themselves do not need to pass the cortical actin cytoskeleton underneath the
membrane, which is a complicated process (see the previous section on
phagocytosis). Once the internalized vesicles are released into the relatively
unimposing cytosol, viruses penetrate the membrane of the vesicular
compartment into which they were internalized. Not surprisingly, most
viruses utilize clathrin-mediated endocytosis, simply because it constitutes
the main pathway for internalization in most cell types (Whittaker et al.,
2000).
Given the existence of several alternative uptake pathways, some viruses
should have adapted to utilize these alternative routes for infectious entry
into the host cell. Indeed, this was first recognized in electron microscopic
images which showed Simian Virus 40 and the related Polyoma Virus
entering cells via uncoated plasma membrane invaginations, distinct from
clathrin-coated pits (Hummeler et al., 1970; Mackay and Consiligi, 1976).
More recent studies showed that the small invaginations contained caveolin-1
(Stang et al., 1997). Also drugs that extract cholesterol from the plasma
membrane blocked SV40 infection (Anderson et al., 1996; Stang et al., 1997).
These observations suggested that virus particles enter the cells via caveolae.
A more direct link between caveolae and SV40 infection was shown by the
fact that overexpression of an N-terminally truncated mutant of caveolin-3
inhibits formation of caveolae and also SV40 infection (Roy et al., 1999).
However, whether caveolae serve as the endocytic structures that internalize
SV40 particles as a first step in the infectious entry had not been studied.
SV40 and polyoma virus are non-enveloped double-stranded DNA viruses of
the papova virus family with an icosahedral capsid of 50 nm in diameter.
SV40 contains three virus-encoded proteins, VP1, VP2 and VP3. VP1 (41 kDa,
34
Caveolae-mediated endocytosis of SV40
364 amino acids) forms the 72 pentameric capsomers that build the shell of the
virion (Liddington et al., 1991). VP2 (352 amino acids) and VP3 (234 amino
acids) are minor proteins located inside the capsid associated with the viral
DNA (Barouch and Harrison, 1996). VP2 is myristylated and thought, on the
basis of mutant studies, to play a role in the virus entry process (Cole et al.,
1977; Sahil et al., 1993). While no disulfide bonds exist in VP1 or between
subunits of a pentamer, disulfides do occur between pentamers (Stehle et al.,
1996). They may, together with Ca2+, play a role in stabilizing the particle,
since it is known that reducing and Ca2+-chelating agents promote capsid
disassembly (Christiansen et al., 1977).
SV40 is a structurally well-characterized virus. Together with polyoma virus
it is, in fact, one of few larger animal viruses for which the structure is known
at atomic resolution (Stehle et al., 1996). The initial receptors for SV40 are
MHC class I molecules (Breau et al., 1992). Since it was shown that antibody
cross-linking of MHC class I molecules leads to their clustering in caveolaelike domains, it was proposed that SV40 virus particles get sequestered into
caveolae by clustering of MHC class I molecules (Stang et al., 1997). However,
another study showed that MHC class I molecules do not enter together with
SV40 particles (Anderson et al., 1998), suggesting that another, secondary
receptor might be involved.
Approximately six hours after entry, SV40 was found to accumulate in a
smooth ER compartment, which is unusual for viruses (Kartenbeck et al.,
1989). Most virus particles resided here and the addition of a large number of
virus particles to the extracellular medium lead to an expansion of the smooth
ER. Finally, the genome of SV40 has to reach the nucleus in order to be
transcribed and replicated. It appears that an essential step in these later
stages of SV40 entry is the passage through the cytosol, which allows the viral
genome to reach nuclear pore complexes, through which the genome is
imported into the nucleus (Greber and Kasamatsu, 1996). In order to achieve
this, SV40 particles must penetrate the membrane of presumably the
endoplasmic reticulum, but this has not been studied. Although the initial
uptake of SV40 particles seems fairly efficient, the time it takes for early gene
expression to start, namely at least 12 hours, and the fact that the infectious
Chapter 1. Introduction: The many faces of endocytosis
35
unit to particle ratio is estimated to be very low, namely 1 infectious unit per
1,000 particles, suggests that especially the later stages are inefficient
processes (Kasamatsu and Nakanishi, 1998).
1.6. Outline of this thesis
This thesis addresses three related questions: (1) Do caveolae internalize? (2)
What is the itinerary of internalized caveolae? (3) How is internalization of
caveolae established? The ligand used to study caveolae-mediated
endocytosis was SV40. The virus was chosen for three reasons. First, SV40
particles appear to avoid clathrin-mediated uptake (Kartenbeck et al., 1989),
suggesting that they specifically utilize an alternative uptake pathway,
presumably via caveolae (Anderson et al., 1996; Stang et al., 1997). Second,
viruses have proven to be excellent tools to study specific pathways of
endocytosis (Helenius et al., 1980; Marsh et al., 1983; Simons, 1982). Finally,
the studies were likely to shed light on the infectious entry of non-enveloped
viruses, of which relatively little is known, particularly in the case of papova
virus family members.
First, a method was developed that allowed visualization of the uptake of
single virus particles in living cells with video microscopy. Chapter 2 reports
the caveolae-mediated uptake of SV40 and the identification of a new twostep vesicular transport pathway from plasma membrane caveolae to the ER
via a novel organelle that was termed the ‘caveosome’. Subsequently, a
combination of biochemistry, genetics and live fluorescence microscopy was
used to characterize the mechanism of caveolae internalization in more detail.
Chapter 3 shows that internalization of caveolae is ligand-triggered, requires
tyrosine phosphorylation and involves transient recruitment of dynamin2 and
extensive rearrangement of the actin cytoskeleton. Chapter 4 reports that
caveosomes contain several lipid-raft ligands and puts the findings into
perspective with other recent reports on clathrin-independent endocytosis. In
chapter 5, concluding remarks are given.
Chapter 2
Caveolar endocytosis of Simian virus 40 reveals a
new two-step vesicular transport pathway to the ER
1
1
Lucas Pelkmans, 2Jürgen Kartenbeck and 1Ari Helenius
Institute of Biochemistry, Swiss Federal Institute of Technology (ETH) Zürich
2
German Cancer Research Center (DKFZ) Heidelberg
Nature Cell Biology, volume 3, May 2001, pages 473-483
38
Caveolae-mediated endocytosis of SV40
Abstract
Simian Virus 40 (SV40) is unusual among animal viruses in that it enters cells
via caveolae, and the internalised viruses accumulate in a smooth
endoplasmic reticulum (ER) compartment. Using video-enhanced, dualcolour, live fluorescence microscopy, we could visualise the uptake of
individual virus particles in CV-1 cells. After associating with caveolae, SV40
was seen to leave the plasma membrane in small caveolin-1 containing
vesicles. Within minutes the viruses entered larger, peripheral organelles that
had a non-acidic pH. While rich in caveolin-1, these did not contain markers
for endosomes, lysosomes, the ER, or the Golgi complex, nor did they acquire
ligands of clathrin-coated vesicle endocytosis. After several hours in these
organelles, the viruses were sorted into tubular, caveolin-free membrane
vesicles that moved rapidly along microtubules. They were deposited within
40 min in perinuclear, syntaxin 17-positive, smooth ER organelles. The
microtubule-disrupting agent nocodazole inhibited the formation and
transport of the tubular carriers, and blocked viral infection. Taken together,
the results demonstrated the existence of a two-step transport pathway from
plasma membrane caveolae, via an intermediate organelle ,termed
caveosomes, to the ER. The pathway bypasses endosomes and the Golgi
complex, and is part of the productive infectious route used by SV40.
Chapter 2. A new intracellular pathway revealed
39
2.1. Introduction
Many animal viruses take advantage of receptor-mediated endocytosis to
enter their host cells. Typically, they are internalised by clathrin-coated
vesicles and penetrate the membrane in endosomes via acid-activated
processes (Helenius et al., 1980; Marsh and Helenius, 1989). That SV40, a nonenveloped DNA virus of the papova virus family, deviates from this pattern
was first observed by electron microscopy (Griffith and Consigli, 1984;
Hummeler et al., 1970; Kartenbeck et al., 1989). On the cell surface, the viruses
were found to be trapped in small tight-fitting invaginations later found to
represent caveolae (Anderson et al., 1996; Palade, 1953; Stang et al., 1997).
Next, the viruses were seen in small non-clathrin-coated vesicles in the
cytosol, and after 15-30 min increasingly within tubular membrane-bound
organelles that contained multiple virus particles (Kartenbeck et al., 1989).
Starting 4-6 hours after uptake, the viruses accumulated in an anastomising,
tubular membrane network associated with the ER where it remained for a
long time (Kartenbeck et al., 1989; Maul et al., 1978). When a large number of
viruses was added to cells, these smooth ER networks expanded in size
reaching several micrometers in diameter (Kartenbeck et al., 1989).
It is now well established that SV40 uses MHC class I antigens as its cell
surface receptor, and that the productive infectious pathway involves
caveolae. Accordingly, infection is blocked by addition of antibodies to MHC
class I antigens (Breau et al., 1992; Stang et al., 1997), by administration of
cholesterol-depleting drugs that inhibit formation of caveolae (Anderson et
al., 1996), and by expression of dominant-negative mutants of caveolin-3, a
major protein component of caveolae (Roy et al., 1999).
To what extent caveolae participate in constitutive endocytic processes in the
cell is still unclear. However, caveolar internalisation of gold-conjugated
albumin, GPI-anchored folate receptor, (GM1) ganglioside-bound cholera
toxin, and GPI-anchored alkaline phosphatase has been reported (Anderson
et al., 1992; Montesano et al., 1982; Parton et al., 1994; Schnitzer et al., 1994).
From caveolae, the gold-conjugated albumin is thought to travel to
endosomes (Schnitzer and Bravo, 1993), cholera toxin via endosomes and the
40
Caveolae-mediated endocytosis of SV40
Trans Golgi Network to the ER (Lencer et al., 1999), and alkaline phosphatase
via unidentified vesicular structures back to the plasma membrane (Parton et
al., 1994). Also, movement of caveolin-1 from the plasma membrane to
intracellular compartments has been observed after cholesterol oxidation or
depletion (Carozzi et al., 2000; Smart et al., 1994). That N-terminal truncation
mutants of caveolin-3 localise to intracellular vesicles distinct from early
endosomes has suggested the presence of a unique, intermediate organelle in
membrane trafficking processes involving caveolae (Roy et al., 1999).
To analyse caveolar endocytosis in living cells, we investigated SV40
internalisation using dual-colour video enhanced, live microscopy with
Texas-Red labelled virus and GFP-tagged caveolin-1 and tubulin. The results
showed that SV40 transport from caveolae to the ER involves two distinct
phases, and a unique intermediate sorting-compartment distinct from
endosomes, lysosomes and the Golgi complex.
Chapter 2. A new intracellular pathway revealed
41
2.2. Results
2.2.1. Texas red-labelled SV40 and caveolin-1-GFP behave normally.
To analyse the uptake of SV40 in live cells, we labelled purified virus with
Texas Red-X (TRX-SV40). After re-purification, SDS-PAGE and fluorography
showed that the label was exclusively coupled to the VP1 protein, and
absorbance spectra indicated that there were about 103 molecules of dye per
virion (fig. 1a). The virus preparation was mono-disperse. Plaque assays
showed no loss in infectivity in CV-1 cells compared to unlabeled virus (3x108
plaque forming units (pfu) per µg of viral protein). Under a confocal (not
shown) or wide-field microscope (fig. 1b), the virus suspension was visible as
spots of uniform size that likely corresponded to individual virions. The
course of early infection was normal as judged by the expression of T-antigen,
20 hours after virus addition (see below). Since Texas Red staining in the cells
overlapped exactly with the pattern seen by immunofluorescence with an
anti-SV40 polyclonal antibody, we concluded that the dye did not dissociate
from the virus during the course of infection (not shown). This was consistent
with our previous observation that degradation of incoming viral proteins is
extremely slow (Kartenbeck et al., 1989).
To follow the distribution of caveolin-1 in live cells, we constructed plasmids
encoding enhanced GFP either at the N- or the C-terminus of canine caveolin1. When cells were transfected, a similar overall distribution of fluorescent
caveolin-1 was observed for both fusion proteins (see below). However,
whereas the cells expressing C-terminally GFP-tagged caveolin-1 allowed
normal SV40 infection, the N-terminally tagged caveolin served as a dominant
negative inhibitor. It prevented SV40 uptake into cells, and inhibited Tantigen expression (not shown). This confirmed that the N-terminus of
caveolin is crucial for caveolae-mediated uptake processes (Roy et al., 1999).
The C-terminally tagged caveolin-1 (caveolin-1-GFP) was used for all
subsequent experiments.
When viewed in live cells with wide-field microscopy, caveolin-1-GFP
staining appeared in two distinct patterns. One pattern consisted of small,
42
Caveolae-mediated endocytosis of SV40
scattered spots on the plasma membrane (fig. 2a, left panel and enlargement)
that, most likely, represented individual caveolae, and as larger and more
brightly stained, intracellular organelles scattered through the cytoplasm (fig.
2a, right panel). The amount of caveolin-1-GFP expression in the transfected
cells was analysed by immunoblotting with a polyclonal antibody against the
N-terminus of caveolin-1 (N20) that recognised both endogenous and GFPtagged caveolin-1 (fig. 2b, panel I). Quantification of the bands showed that
the expression of caveolin-1-GFP was app. 0.25 of that of the endogenous
protein. Since fluorescence microscopy showed that about 50% of the cells
were transfected, we estimated that the amount of expressed caveolin-1-GFP
in the cells was half of the endogenous caveolin-1. High-speed centrifugation
of solubilised lysates in sucrose gradients showed that the majority of
endogenous and GFP tagged caveolin-1 were present as oligomers (fig. 2b,
panel III). Furthermore, immunoprecipitation with an anti-GFP antibody
brought down both GFP-tagged and endogenous caveolin-1, indicating that
they were present as hetero-oligomers (fig. 2b, panel II).
Indirect immunofluorescence with the N20 antibody showed that the overall
distribution of caveolin-1 was not changed by expression of caveolin-1-GFP
(fig. 2c, open arrowheads). When regions of the plasma membrane were
viewed in enlargements, it was apparent that the N20 antibody and the
caveolin-1-GFP signal overlapped both at 37ºC (fig. 2c, left panels) and at 0ºC
(not shown). Complete overlap was also observed in intracellular organelles
of most cells (fig. 2c, right panel enlargements). However, in a small fraction
of cells that showed particularly high levels of caveolin-1-GFP expression, we
observed in addition a perinuclear accumulation of caveolin-1-GFP that was
not stained by the N20 antibody (fig. 2c, asterisk in right panel).
Immunofluorescence against mannosidase II showed that it corresponded to
the Golgi complex (fig. 2d). This indicated that caveolin-1-GFP, like
endogenous caveolin-1 (Dupree et al., 1993), has a different conformation
when present in caveolae and intracellular spots than in the Golgi complex.
We concluded that caveolin-1-GFP was not massively over-expressed in the
cells, that it was a reliable reporter of caveolin-1 distribution, and that it was
functional.
Chapter 2. A new intracellular pathway revealed
43
Fig. 1. Texas Red labels outer capsid proteins of SV40
a, Analysis of SDS-PAGE separated Texas Red-X labelled SV40 by fluorometry (left lane) or
coomassie (right lane). VP1 (41 kDa) is the only protein fluorescently labelled. b, Wide field
fluorescence analysis of Texas Red-X labelled SV40 suspension showing individual spots of uniform
size. Some larger spots are likely 2 or 3 virus particles in too close proximity to be individually seen.
Scale bar in b represents 2.7 µm.
44
Caveolae-mediated endocytosis of SV40
Fig. 2. Caveolin-1-GFP behaves as endogenous caveolin-1.
a, Wide field fluorescence analysis of CV-1 cells transiently expressing caveolin-1-GFP for 16 hours
showing a spotty pattern on the membrane (z-axis position is 0.0 µm corresponding to the apical plasma
membrane, see methods), which are somewhat smaller than spots on fixed cells (enlargement and c) and
intracellular vesicles (right panel) (z-axis position is -0.5 µm). b, (panel I) Western blots of SDS-PAGE
separated lysates of CV-1 cells transiently expressing caveolin-1-GFP for 16 hours with an anticaveolin-1 antibody (N20), showing relative expression levels of caveolin-1-GFP and caveolin-1.
(panel II) Immunoprecipitation with an anti-GFP antibody of lysates of CV-1 cells expressing
caveolin-1-GFP and subsequent western blotting against caveolin-1 (N20) (first lane), shows that
caveolin-1-GFP is complexed with endogenous caveolin-1. In immunoprecipitations of control lysates
(GFP) (second lane) no endogenous caveolin-1 was precipitated. Western blotting of precipitates with
an anti-GFP antibody (last 2 lanes) detected only caveolin-1-GFP or GFP (control lysates, GFP). (panel
III) High speed centrifugation through a 5-50% sucrose gradient of solubilised lysates and western
blotting against caveolin-1 (N20) shows that a large part of both caveolin-1-GFP and caveolin-1 are
present in higher weight complexes (positions of molecular weight standards in parallel gradients are
indicated above the blots). c, Laser scanning confocal immunofluorescence analysis of CAV1-GFP
expressing CV-1 cells using an antibody which recognises the caveolar form of caveolin-1. CAV1-GFP
(green) co-localises completely with endogenous caveolin-1 (red) in its caveolar form, both on the
plasma membrane (left panel and enlargements; z-axis position is 0.0 µm) and on intracellular vesicles
(right panel, z-axis position is –0.5 µm, corresponding to 0.5 µm below the apical plasma membrane,
and pinhole size 0.2 a.u, see methods) Note that non-transfected cells show similar patterns (open
arrowheads) and that the occasionally seen Golgi-pool (see d) of CAV1-GFP (asterisk) is not stained by
the N20 antibody. d, Laser scanning confocal immunofluorescence analysis of cav1-GFP expressing
CV-1 cells (z-axis position is –0.5 µm) against Mannosidase II shows that the occasionally visible
perinuclear accumulation of caveolin-1-GP (asterisk) localises to the Golgi complex. Note that not in all
cells CAV1-GFP is present in the Golgi complex (filled arrowheads). Scale bars in a represent 10 µm
(left and right) and 1 µm (enlargement), in c 7 µm (whole cells) and 1 µm (enlargements) and in d 10
µm.
Chapter 2. A new intracellular pathway revealed
45
2.2.2. After binding to the cell surface, SV40 moves to stationary caveolae in
the membrane, and is internalised.
To analyse the initial stages of virus-cell interaction, we allowed TRX-SV40 to
bind to CV-1 cells at 4ºC at which temperature no virus internalisation occurs
as judged by electron microscopy (Kartenbeck et al., 1989). When the cells
were fixed, a characteristic spotty fluorescence pattern was seen on the
plasma membrane (fig. 3a, left panel). Only few of the TRX-SV40 spots colocalised with caveolin-1-GFP indicating that the majority of viruses did not
bind directly to caveolae (fig. 3a, middle panel). This was confirmed by
electron microscopy (fig. 3c).
In live cells at 37ºC, the TRX-SV40 spots were stationary, except for 10-20%
that were laterally mobile (movie 1). The stationary viruses were now
associated with caveolin containing structures. This redistribution occurred so
rapidly that by the time a coverslip was moved from ice to a stage of a
microscope thermostated at 37ºC for live fluorescence analysis, most of the
TRX-SV40 already co-localised with caveolin-1-GFP (fig. 3a, right panel).
Electron microscopy confirmed that after 15 min at 37ºC, the majority of virus
particles localised to caveolae (fig. 3c).
While the double-positive spots were largely stationary, they could suddenly
become mobile, and move within the time frame of 3-6 sec out of the focus
plane and into the cell. Using dual-colour live microscopy with a confocal
microscope, this movement was particularly well seen as spots suddenly
disappearing from the focal plane (fig. 3b and movie 2).
The images and video recordings were consistent with a process whereby,
after binding to MHC Class I antigens, virus particles are mobile in the plane
of the membrane until trapped in stationary caveolae. The virus-containing
caveolae are subsequently internalised leaving no detectable caveolin-1
behind in the plasma membrane. At least within the time frame of 3 min, we
could not observe internalisation of caveolae devoid of virus, suggesting that
the virus somehow induced or accelerated the process of caveolar
internalisation.
46
Caveolae-mediated endocytosis of SV40
2.2.3. Entry in caveosomes - intermediate organelles in caveolar endocytosis.
After 0.2-4 hours of association with cells at 37ºC, the majority of SV40 is
known to be intracellular, and localised in irregularly-shaped, membranebounded organelles (Kartenbeck et al., 1989). When viewed by wide field
microscopy in live cells incubated with TRX-SV40, these organelles could be
seen as brightly fluorescing spots dispersed throughout the cytosol
undergoing slow, non-directional movements (fig. 4a right panel and movie
3). Most of them were positive for caveolin-1-GFP (fig. 4a left panel). They
resembled in size and distribution the caveolin-containing, intracellular
organelles also seen in the absence of SV40 (fig. 2a or 2c). Following the entry
of TRX-SV40 into live cells expressing caveolin-1-GFP, we observed that the
majority of the caveolin-1-GFP positive organelles that the TRX-SV40 entered
were, indeed, already present in the cytoplasm prior to virus addition (fig.
4b). In other words, most of the viruses were transported to pre-existing
caveolin-1 containing organelles.
To investigate whether the organelles were part of the endosomal/lysosomal
system, we incubated cells with TRX-SV40 and at the same time with two
ligands known to be internalised by clathrin-mediated endocytosis [BODIPYFL labelled transferrin and fluorescein labelled Semliki Forest Virus (FLXSFV)] or a fluorescent fluid phase marker (FITC-dextran). After
internalisation, these had a distribution similar to that of TRX-SV40, but no
overlap with the TRX-SV40 containing organelles was seen (fig. 4c). That the
SV40-containing organelles were distinct from classical endosomes was also
indicated by their lack of staining in fixed cells with antibodies to EEA1, a
marker protein of early endosomes (fig. 4c). The distribution of caveolin-1GFP was, in fact, distinct from that of EEA1, whether in the presence or
absence of SV40 (not shown). Furthermore, the virus-containing organelles
did not accumulate Lysotracker green, a lysosomal marker (not shown), and
they were not stained by antibodies to TGN46 (fig. 4c) or mannosidase II (not
shown). These results indicated that they were not related to lysosomes or the
Golgi complex, confirming our previously published electron microscopic
observations (Kartenbeck et al., 1989). Finally, at this intermediate stage of
uptake, the virus did not overlap with markers of the ER or the intermediate
compartment (not shown, see below).
Chapter 2. A new intracellular pathway revealed
47
Fig. 3. After binding to the cell surface, SV40 moves into stationary caveolae that are
subsequently internalised.
a, (left), Wide field fluorescence analysis of plasma membrane bound TRX-SV40 on fixed CV-1
cells, showing the spotty pattern of surface bound virus particles (z-axis position is 0.0 µm).
(middle), Fluorescence analysis of fixed CV-1 cells expressing CAV1-GFP (green) immediately
after binding of TRX-SV40 at 4ºC(red). The distribution shows that TRX-SV40 does not bind
directly to CAV1-GFP microdomains. Note that one virus is already localised to a caveola (open
arrowhead; z-axis position is 0.0 µm). The picture is an enlargement of the square indicated in the
inset. (right), Laser scanning confocal live fluorescence analysis (z-axis position is 0.0 µm, pinhole
1.0 a.u.) of cells like those above rapidly shifted to 37ºC. In a short period of time (15 min), the
majority of TRX-SV40 spots (red) have re-localised to CAV1-GFP microdomains (green) (indicated
by open arrowheads). The picture is an enlargement of the square indicated in the inset. b, Selected
frames of a laser scanning confocal live fluorescence recording (z-axis position is 0.0 µm, pinhole
1.0 a.u.) at 37ºC of a part of the membrane of CV-1 cells expressing CAV1-GFP and having bound
TRX-SV40. One spot at the arrowhead and two spots in the circle (left upper 2 spots) suddenly
disappear, while others remain in place (time is in min:sec). c, electron microscopy of CV-1 cells
immediately fixed after having bound SV40 for 2 hours at 4ºC or fixed after a shift to 37ºC for 15
min. At 4ºC, virus particles are bound to the membrane (open arrowheads left pictures), but not yet
sequestered into caveolae (left lower picture). After 15 min at 37ºC, the majority is now sequestered
into caveolae (open arrowheads right pictures), and not in clathrin coated pits (CCP). Scalebar in a
left panel represents 10 µm, in the other panels and in b 2 µm and in c 500 nm (upper pictures) and
100 nm (lower pictures).
48
Caveolae-mediated endocytosis of SV40
Fig. 4. SV40 enters caveosomes: Intermediate organelles in caveolar uptake.
a, (left) Live fluorescence analysis of internalised TRX-SV40 in non-transfected CV-1 cells1 hour
after virus binding (z-axis position is 0.0 µm, pinhole 1.0 a.u.). It shows small vesicles after
internalisation and accumulation of virus in larger organelles. (right) Laser scanning confocal live
microscopy of CAV1-GFP expressing CV-1 cells 2 hours at 37ºC after virus binding shows that the
majority of internalised virus is in larger organelles that are strongly positive for CAV1-GFP. Note
that some virus is still present in smaller CAV1-GFP positive vesicles. b, Selected images of laser
scanning confocal live fluorescence recordings of CAV1-GFP expressing CV-1 cells during addition
and subsequent uptake of virus, while focussing on intracellular CAV1-GFP containing vesicles (zaxis position is 0.0 µm, pinhole 0.2 a.u.). Images shown were taken before virus addition, and 2
hours after virus addition, which show that the virus accumulates in pre-existing CAV1-GFP positive
organelles. Note a similar accumulation of virus in surrounding non-transfected cells. c, Fluorescence
analysis (z-axis position is –0.5 µm) of CV-1 cells incubated with fluorescent transferrin (BodipyFL-Tfn), Semliki Forest Virus (FLX-SFV) and FITC-dextran (MW 42,000) during virus binding and
internalisation for 1 hour at 37ºC showing that the organelles containing SV40 do not accumulate
other endocytic markers. Immunofluorescence analysis of CV-1 cells, 2 hours at 37ºC after virus
binding, against a marker of early endosomes (EEA1) or the Trans Golgi Network (TGN46), shows
that the virus does not accumulate in these organelles. d, thin section electron microscopic image of a
typical caveosome, 3 hours after virus binding and shift to 37ºC, containing multiple virus particles
and showing an irregular shape. Note that the membrane lies tightly around the virus particles but is
continuous (arrowheads). The inset shows a small vesicle directly after internalisation (30 min at
37ºC), which contains only one virus particle. Scalebars in a-c represent 5 µm and in d 500 nm and
50 nm (inset).
Chapter 2. A new intracellular pathway revealed
49
Thin section electron microscopy showed that two hours after uptake, the
majority of the intracellular organelles that contained the virus, were
membrane-bounded, had irregular shapes and sizes, and contained numerous
virus particles (fig. 4c, last panel; see also (Kartenbeck et al., 1989)). Although
some of the organelles were located near the plasma membrane, most were
deeper in the cytoplasm. They clearly differed in size and complexity from the
primary endocytic vesicles that contained a single virus particle observed at
earlier timepoints (see inset fig. 4c, last panel).
To determine whether the virus-containing organelles constituted deep
invaginations still connected to the plasma membrane, we tested whether
extracellular ions could reach the virus after internalisation. For this purpose,
we labelled SV40 with the fluorophore fluorescein-X (FLX), because its
excitation spectrum is pH-sensitive (Ohkuma and Poole, 1978). After binding
of FLX-SV40 in the cold, or after internalisation at 37ºC for 3 hours, the CV-1
cells were resuspended in buffers of different pH (pH 4.0, 7.0 and 8.0) and
immediately analysed with a fluorometer (Maxfield, 1985). As expected, the
spectrum of membrane-bound FLX-SV40 was sensitive to the pH (fig. 5a). In
contrast, the fluorescence emitted by the internalised FLX-SV40 did not
change (fig. 5b). After addition of ionophores monensin and nigericin
(Maxfield, 1982), the spectrum of the internalised FLX-SV40 changed, as the
pH of the intracellular organelles was now clamped with the extracellular pH
(fig. 5b). This result clearly established that the virus-containing
compartments were not connected to the plasma membrane.
When we compared the unclamped and the pH 7.0-clamped spectrum of
internalised FLX-SV40 with each other, and with that of the fluid phase
marker FITC-dextran (the fluorescence of which is similarly pH dependent), it
was, moreover, clear that the viruses were present in a compartment of
neutral pH (Mellman, 1985). We concluded that the organelles in which the
virus was trapped had a pH close to neutrality, and that they were not
connected with the plasma membrane.
Taken together, the results showed that the majority of incoming SV40
entered a population of pre-existing, caveolin-1 rich, intracellular organelles
50
Caveolae-mediated endocytosis of SV40
with a pH around 7.0. These were distinct from organelles of the classical
endocytic and secretory pathways. Since they appeared to be unique to the
caveolar uptake pathway, and since they contained caveolin-1, we named
them ‘caveosomes’.
2.2.4. SV40 is sorted from caveosomes.
SV40’s passage within the caveolar endocytic pathway was considerably
slower than the transport of ligands within the coated vesicle-mediated
endocytic pathway. Typically, it took about 20 min before the virus was
internalised from the plasma membrane and 20 to 40 min before it reached the
caveosomes. Further 2-4 hours were needed before the virus started to leave
the caveosomes. At this time, video microscopy indicated that the caveosomes
became more dynamic; their morphology underwent rapid changes and some
of them fused with each other. Fission reactions also occurred, and viruses
could be seen to exit the caveosomes in vesicular and tubular structures (see
below).
The departure of viruses from caveosomes could be readily visualised with
dual colour time lapse microscopy in live caveolin-1-GFP expressing cells
incubated with TRX-SV40 at 37ºC for 4-6 hours. From caveosomes, which
were yellow in colour, mobile, tubular, red-coloured extensions could be seen
to emerge (fig. 6a, open arrowheads and movies 4, 4a, 4b and 4c). Frequently,
these broke off and moved swiftly away from the caveosome. The average
length of the tubular structures detaching from the caveosomes was about 1
µm, but some were as long as 7 µm (fig. 6b and movies 5, 5a and 5b). Since the
TRX-SV40 containing tubules were red, they were devoid of detectable
caveolin-1-GFP (fig. 6d, upper series). Their formation thus involved
molecular sorting within the caveosomes. We could, in fact, sometimes
observe the separation of red and green signals within a single caveosome
suggesting that caveolin-1-GFP and TRX-SV40 were being separated into
distinct domains (fig. 6d; lower series), which could both detach from the
caveosome as distinct vesicles. Thin section electron micrographs showed the
presence of virus-containing tubular structures, which could be seen attached
to vesicular organelles devoid of virus (fig. 6c, upper panel), or as apparently
detached structures (fig. 6c, lower panel).
Chapter 2. A new intracellular pathway revealed
51
Fig. 5. caveosomes are not
accessible to extracellular proton
ions and contain a neutral pH.
a, Fluorescence excitation profile
analysis of FLX-SV40 bound to the
membrane of CV-1 cells for 2
hours at 4ºC showing that the
spectra were sensitive to the
extracellular pH (pH4, 7 or 8). b,
The same analysis of FLX-SV40
after internalisation for 3 hours at
37ºC showed that the spectra were
not sensitive to the extracellular pH
(pH 4, 7, 8). Addition of 10 µM of
monensin and nigericin (+
ionophores) to the same samples,
resulted in rapid equilibration of
the spectra to the corresponding
extracellular pH. Note that nonclamped spectra had the same
profile as the spectrum clamped at
pH 7. c, Fluorescence excitation
profile analysis of FITC-dextran
internalised for 3 hours at 37ºC in
CV-1 cells showing the same
spectrum at different extracellular
pH (pH 4 and 8). After addidition
of 10 µM monensin and nigericin
(+ ionophores), the spectra were
clamped to the corresponding
extracellular pH. Note that the nonclamped spectra have a much more
acidic profile than FLX-SV40
(compare with b).
52
Caveolae-mediated endocytosis of SV40
Fig. 6. SV40 is sorted in caveosomes.
a, Laser scanning confocal live fluorescence analysis
(z-axis position is -0.5 µm, pinhole 0.2 a.u.) of CAV1GFP transfected CV1-cells 4 hours at 37ºC after virus
binding, showing polarisation of CAV1-GFP and
virus in organelles (open arrowheads) that have
tubular extensions containing only virus (enlargement
of first picture indicated by white box). b, Live
fluorescence recording (z-axis position is –0.5 µm) of a CV-1 cell (left), 4 hours at 37ºC after virus
binding showing long tubules in the cell (open arrowheads) and the dynamics of tubule formation from
a bigger virus containing vesicle (selected frames). c, Thin section electron microscopy of CV-1 cells,
5 hours at 37ºC after virus binding, showing tubular extensions containing several virus particles
(upper panel) emerging from vesicular structures (asterisks) devoid of virus or tubular carriers (lower
panel). Frequently, microtubules (open arrowheads) were observed in close proximity. Note that the
virus suspension used for these experiments contain both full and empty (lacking DNA) virus
particles. d, Selected frames of part of a CV-1 cell expressing CAV1-GFP 4 hours at 37ºC after virus
binding showing formation and dissociation of tubular carriers containing only virus from a central
caveosome (asterisk, filled and open arrowheads, upper series) and polarisation of a multi-domain
organelle (open arrowheads, top series). Scalebar in a represents 10 µm (left) and 3 µm (right), in b 10
µm (left) and 3 µm (selected frames) and in c 150 nm.
Chapter 2. A new intracellular pathway revealed
53
Often they were seen in close proximity of microtubules (fig. 6c, open
arrowheads).
2.2.5. SV40 entry is a two-step process
The tubular SV40 carriers that emerged from the caveosomes were flexible
(suppl info movie 5, 5a and 5b), and moved through the cytosol in the
direction of their long axis. Compared to slow-moving early vesicles and
caveosomes, the movement was rapid, averaging 0.5 µm per sec (fig. 7a and
b). Although individual carriers could be seen to move in both retrograde and
anterograde directions, net movement was in the direction of the perinuclear
area, which contained the larger organelles in which the fluorescent virus
accumulated (see below). A merged picture of all frames taken over a period
of 3-4 min (see methods) showed that the majority of the tracks were positive
for TRX-SV40, negative for caveolin-1-GFP (fig. 6a, right panel), and that
movement was in the direction of the perinuclear area.
We next measured the amount of the virus in the different cellular locations at
different times of entry (fig. 7c). As described in the methods, we quantified
the fluorescent virus signal that was present on the plasma membrane, in
caveolae, in caveosomes, and in the perinuclear organelles. This confirmed
that during the first 20 min, the plasma membrane-bound virus first
associated with caveolae, and that they were internalised at 37ºC into the
caveosome with a half time of 60 min. After 2.5 hours, co-localisation with
caveolin-1-GFP rapidly decreased, and soon thereafter the virus accumulated
in perinuclear organelles.
2.2.6. After sorting, SV40-containing carriers travel along microtubules
To investigate whether microtubules played a role in the movement from
caveosomes to the perinuclear area, we followed the fate of TRX-SV40
containing carriers in PtK2 cells expressing YFP tagged α-tubulin. The timecourse of T-antigen expression, and its sensitivity to cholesterol depletion,
indicated that SV40 infected these cells, like CV-1 cells, via the caveolar
pathway (not shown). Video sequences showed that fast moving virus
carriers were tracking along pre-assembled microtubules (fig. 8a movie 6 and
6a). Although movement appeared to occur in both antero- and retrograde
54
Caveolae-mediated endocytosis of SV40
directions, the virus eventually accumulated in larger perinuclear organelles
(see below). Closer analysis showed that the virus-containing tubular
structures budding from caveosomes were already attached to microtubules
(movie 6b and 6c).
Nocodazole, a drug that dissociates the microtubule cytoskeleton, did not
inhibit virus uptake into CV-1 or PtK2 cells, nor its transport to caveosomes
(fig. 8a). However, the virus-containing caveosomes did not enter the
dynamic phase, and tubular extensions or tubular carriers were not formed.
Transport of the virus to the perinuclear accumulation sites was inhibited (fig.
8a and movie 7). Some virus-containing vesicles budded from the caveosomes
but these did not move away rapidly (movie 7a). While the microtubule
network was not needed for plasma membrane to caveosome transport of
SV40, it was evidently essential for efficient caveosome-to-ER transport.
To investigate whether the microtubule-dependent transport step was needed
for productive SV40 infection, T-antigen expression after 20 hours was
analysed in the presence or absence of nocodazole. We found that in the
presence of nocodazole, T-antigen expression was blocked (fig. 8b, middle
row). No expression of this viral protein was seen even after 48 hours (not
shown). When the nocodazole was removed after 20 hours, the virus
accumulated in the smooth ER (see below), and, within 12 hours, T-antigen
expression could be observed (fig. 8b, lower row). We concluded that the
microtubule dependent transport of virus to the ER was part of the
productive infectious pathway, and that nocodazole blocks infection at a late
state of the entry process.
2.2.7. After sorting, SV40 rapidly accumulates in the smooth ER.
After leaving the caveosomes, TRX-SV40 accumulated in large, heterogeneous
stationary organelles in the perinuclear region of the cell. That net transport
occurred to these could be shown by fluorescence recovery after
photobleaching (FRAP) (fig. 9a). Four hours after TRX-SV40 addition, the
perinuclear regions containing accumulated virus were bleached with a high
power laser beam. Upon subsequent incubation at 37ºC, accumulation of new
Chapter 2. A new intracellular pathway revealed
55
Fig. 7. SV40 entry is a two-step process
a, Live fluorescence analysis (wide field microscopy, left picture and dual colour laser scanning
confocal microscopy in CAV1-GFP transfected cells, right picture) of internalised virus 4 hours after
virus binding, depicted as a merged recording (50 frames, 500 ms exposure, 2.5 sec intervals; see
methods), showing long tracks of fast moving perinuclear oriented carriers (left, z-axis position is -0.5
µm) that only contain virus and originate from caveosomes (right, z-axis position is -0.5 µm, pinhole
0.2 a.u.). b, (left)Quantification of the average speed (see methods) of moving virus at the plasma
membrane within the first 0-10 min after virus binding and shift to 37ºC (PM) (9 tracks, 4 cells), 30
min to 1 hour at 37ºC after virus binding (early vesicles) (36 tracks, 4 cells), 2-3 hours after virus
binding (caveosomes) (42 tracks, 3 cells) and of tubular carriers at 4-5 hours after virus binding
(tubules) (38 tracks in 6 cells). Error bars are standard errors of the mean. Internalised virus is more
dynamic than the small dynamic pool on the plasma membrane and speed increases about 5 fold when
tubular carriers haven been formed. (right) Quantification of the amount of TRX-SV40 signal
overlapping with caveolin-1-GFP signal (see methods) and the amount of TRX-SV40 signal present in
specific regions of interest (see methods), plasma membrane (PM), cell periphery (peripheral) and
close to the nucleus (perinuclear) as a function of time after a shift to 37ºC. Values are expressed as
mean percentage of total virus signal (at least 4 different cells per time-point). showing that when
speed of intracellular carriers is relative low (see b) the majority of TRX-SV40 is in the cell periphery
and overlaps with caveolin-1-GFP (up to 3 hours after shift to 37ºC). When speed of carriers increased
(see b), overlap with caveolin-1-GFP rapidly dereased and TRX-SV40 accumulated in a perinuclear
area (between 3-6 hours after shift to 37ºC. Scalebar in a represents 10 µm.
56
Caveolae-mediated endocytosis of SV40
Fig. 8. After sorting, SV40 travels along preassembled microtubules to its accumulation
site, which is necessary for infection.
a, Live fluorescence analysis (z-axis position is
–0.5 µm) of stable YFP-α -tubulin (green)
transfected Ptk2 cells, 4 hours at 37ºC after virus
binding under normal conditions or in the presence
of 10 µM nocodazole, depicted as merged
recordings (52 frames, 4 sec intervals, see
methods). Initial virus entry is not dependent on an
intact microtubule skeleton (right, nocodazoletreated), but movement towards a perinuclear site
is (left). Note the long tracks of moving virus in
control cells that run along pre-assembled
microtubules (open arrowheads in left
enlargements), that are devoid in nocodazoletreated cells (right enlargements). b, Wide-field
immunofluorescence analysis (z-axis position is
–0.5 µm) of T-antigen expression in CV-1 cells,
20 hours at 37ºC after virus binding, in normal
growth medium (upper row) or in growth medium containing 1 µM nocodazole (middle row),
showing that the majority of TRX-SV40 is in the perinuclear accumulation site when T-antigen
expression occurs. When transport from caveosomes to the smooth ER is blocked by nocodazole, no
accumulation of TRX-SV40 and T-antigen expression is seen (middle row, asterisks indicate the
nuclei). After removal of nocodazole and incubation for 12 hours in normal medium at 37ºC, all TRXSV40 has now accumulated in the smooth ER and T-antigen expression is observed. Scale bars
represent 10 µm.
Chapter 2. A new intracellular pathway revealed
57
fluorescent virus in the same organelles could be observed using quantitative
live fluorescence microscopy (fig.9a and movie 8).
To better characterise the compartment in which the incoming SV40
accumulated, we allowed TRX-SV40 to enter CV-1 cells for 16h. As mentioned
above, at this time, the virus was present mainly in the heterogeneous
organelles located in the perinuclear region (fig.9b). The largest of these
reached a size of more than 8 µm. They were clearly devoid (9%) of caveolin1-GFP (fig. 9b, top row), and they were not stained with antibodies to
caveolin-1 (not shown). When TRX-SV40 loaded cells were analysed by
immunofluorescence using antibodies to marker proteins, it was found that
many of the virus containing perinuclear organelles (80%) were positive for
syntaxin17 (Syn17) (fig. 9b, second row), a recently identified smooth ER
marker (Steegmaier et al., 2000; Steegmaier et al., 1998). Some also stained
positively for ER markers such as BiP, calnexin (not shown, see also
(Kartenbeck et al., 1989)) and PDI (70%) (fig 9b, third row), and for the
intermediate compartment marker ERGIC-53 (54%) (not shown). No
significant co-localisation was seen with the Golgi marker mannosidase II
(ManII) (5%) (fig. 9b, bottom row). The organelles did not accumulate
lysotracker green (not shown) indicating that they were not lysosomes. We
concluded that the tubular network seen in the ER as the site of virus
accumulation, indeed, corresponded to a smooth ER compartment, a result
consistent with previous electron microscopic observations (Kartenbeck et al.,
1989).
58
Caveolae-mediated endocytosis of SV40
Fig. 9. SV40 finally accumulates in a smooth ER compartment
a, Fluorescence Recovery After Photobleaching (FRAP) experiment (z-axis position is –0.5 µm) in CV1 cells 4 hours at 37ºC after binding of virus showing transport to a distinct tubulovesicular organelle
(yellow dashed line). Quantification of relative fluorescence intensity (see methods) of the indicated
area shows that recovery is 90% after app 50 min. b, Laser scanning confocal fluorescence and
immunofluorescence analysis (z-axis position is –0.5 µm) of the tubulovesicular structures in which
SV40 finally accumulates, 12 hours at 37ºC after virus binding, showing that they are not positive for
CAV1-GFP and that the majority of them is positive for syntaxin17 (Syn17, enlargements, see
arrowheads). The compartment is more perinuclear than the reticular rER compartment (PDI) and is
partially overlapping, although extensions can be seen that do not overlap with PDI (PDI, enlargements,
red lines). SV40 does not accumulate in the Golgi complex (ManII; see arrowheads). Scale bars
represent 5 µm.
Chapter 2. A new intracellular pathway revealed
59
2.3. Discussion
Although several ligands and pathogens are known to be internalised via cell
surface caveolae (for review see (Anderson, 1998)), the pathways that they
follow intracellularly have remained elusive. Here, we have analysed the
entry of SV40 particles in live cells, and visualised their transfer from caveolae
in the plasma membrane to the smooth ER. Taken together with previous
observations (Anderson et al., 1996; Kartenbeck et al., 1989; Stang et al., 1997),
our results showed that after binding to the cell surface, the viruses move
laterally into caveolae. They apparently trigger a change in the caveolae that
results in the induction or acceleration of membrane fission reactions that
detach the caveolae from the plasma membrane and allow them to move into
the cytoplasm within small caveolin-1 containing vesicles. The viruses are
delivered by these small vesicles to pre-existing, stationary, membranous
organelles that we have called caveosomes. Delivery most likely occurs by
direct membrane fusion.
Caveosomes are distributed throughout the cytoplasm, they are numerous,
and they are heterogeneous in size and shape. Being considerably larger than
the primary, caveolae-derived vesicles, they can contain numerous SV40
particles. Their membrane contains caveolin-1 in a conformational state
similar to that in the plasma membrane, but different from that in the Golgi
complex. The pH in caveosomes is neutral, and our preliminary experiments
with filipin-staining suggest that, like caveolae, they are rich in cholesterol
(unpublished results). The location of caveosomes in the cytosol, the lack of
visible connections to the plasma membrane, and the inaccessibility to
extracellular protons confirmed that they are not connected to the
extracellular space.
Two to four hours after virus arrival, caveosomes become more dynamic.
Although still stationary in the cytoplasm, they begin to undergo rapid shapechanges including the formation of long, tubular, virus-containing, membrane
extensions. After detaching form the caveosomes, these extensions serve as
carrier vesicles that transport the viruses from the caveosomes to the smooth
ER compartment where the virus accumulates. The formation and
60
Caveolae-mediated endocytosis of SV40
intracellular movement of these carrier vesicles requires active participation
of microtubules.
Unlike the primary endocytic vesicles that carry SV40 from the plasma
membrane to the caveosomes, they do not contain caveolin-1. The caveolin-1
is at this stage excluded from virus containing parts of the caveosomes, and
seems to cycle back to the plasma membrane in vesicles devoid of virus.
Although we do not yet know the reason for the gradual change in caveosome
dynamics and the activation of its sorting function, there is reason to believe
that the caveolar endocytosis pathway is under the control of complex
regulation (Parton et al., 1994).
That the caveolar pathway constitutes the infectious route of SV40 is
suggested by the effects of cholesterol-depleting drugs that inhibit both the
formation of caveolae and virus infection (Anderson et al., 1996; Stang et al.,
1997). Expression of N-terminally truncated caveolin-3 (Roy et al., 1999) and
of the N-terminally GFP-tagged caveolin-1 construct described here, also
inhibit infection. Although other endocytic mechanisms remain active in the
presence of these mutant caveolins, SV40 fails to be internalised from the
plasma membrane. We found that infection was also blocked by nocodazole,
which disrupted the microtubule cytoskeleton and prevented transport of
virus from caveosomes to the ER. This suggested that to be infectious, the
virus must not only be internalised by caveolae but be transported to the ER.
Productive infection of SV40 is generally thought to involve penetration of the
virus particles into the cytosol, and transport through the nuclear pore
complexes (see (Greber and Kasamatsu, 1996; Yamada and Kasamatsu, 1993)).
Using the techniques described here, we have not observed free virus
particles in the cytosol or in the nucleus, nor did we observe import of
fluorescent SV40 VP-1 into the nucleus. Further studies are needed to
determine how the genome of the virus moves from the smooth ER to the
nucleus.
Although it is clear that the caveolae and caveosomes exist in the absence of
virus infection, our results do not provide clues as to the physiological role of
Chapter 2. A new intracellular pathway revealed
61
a caveolae-caveosome-ER pathway. It is possible that it plays a role in
cholesterol homeostasis: caveolae are rich in cholesterol and raft-lipids, and
have been implicated in cholesterol regulation (Brown and Goldstein, 1999;
Fielding et al., 1997; Fielding and Fielding, 1997; Lange et al., 1999; Parton,
1996; Simons and Ikonen, 1997). Since caveolae are also rich in signalling
receptors (Kurzchalia and Parton, 1999), an internalisation pathway that bypasses the degradative organelles may have other regulatory functions.
Recently, it was found that FimH expressing E. Coli use caveolae to enter
phagocytic mast cells (Shin et al., 2000). The intracellular compartment in
which these bacteria reside and replicate may be related to caveosomes as
they do not fuse with endosomes or lysosomes. It is, moreover, likely that
viruses other than SV40 utilise the caveolar route. Potential candidates
include picornaviruses and other non-enveloped viruses known to have pHindependent entry mechanisms. Polyoma virus, a virus related to SV40, seems
to enter a pathway similar to that of SV40 (Griffith and Consigli, 1984; Griffith
et al., 1988; Mackay and Consiligi, 1976), although recent observations have
suggested that caveolin and dynamins may not be involved (Gilbert and
Benjamin, 2000).
62
Caveolae-mediated endocytosis of SV40
2.4. Methods
Antibodies and other reagents
Polyclonal antisera against Mannnosidase II, Syntaxin17 and TGN46 were
described previously (Ponnambalam et al., 1996; Steegmaier et al., 1998;
Velasco et al., 1993). Polyclonal antibodies against PDI were purchased from
StressGen Biotechnologies Corp (Victoria, BC, Canada). Monoclonal
antibodies against human EEA1 were purchased from Transduction
Laboratories (Lexington, KY, USA). Polyclonal antibodies against the Nterminus of caveolin-1 (N20) were from Santa Cruz Biotechnology Inc. (Santa
Cruz, CA, USA). Monoclonal antibodies against GFP were purchased from
Boehringer Mannheim (Indianapolis, IN, USA). Polyclonal antiserum against
SV40 was raised in rabbits immunised with 200 µg of heat-inactivated SV40 in
Freud’s adjuvans according to standard procedures. This serum was able to
detect all three structural proteins. FLX-labelled Semliki Forest Virus (FLXSFV) was prepared as previously described (Helenius et al., 1980), and
according to supplier’s instructions (Molecular Probes BV, Leiden, the
Netherlands). Bodipy-FL-transferrin, Texas Red-X and Fluorescein-X were
purchased from Molecular Probes BV (Leiden, The Netherlands). FITCdextran (MW 42,000) was obtained from Sigma-Aldrich Chemie (Steinheim,
Germany). Media and reagents for tissue culture were purchased from
GibcoBRL Life Technologies (Basel, Switzerland). All other chemicals were
purchased from Sigma-Aldrich Chemie (Steinheim, Germany).
Preparation of fluorophore labelled SV40
SV40 was purified by a modification of existing protocols (Khoury and Lai,
1979; Sahli and Beard, 1994). Briefly, 24 T175 flasks with confluent CV-1 cells
were infected by a stock of SV40 at a MOI of 0.01. After 10 days, cells were
freeze-thawed 3 times, and debris spun down at 10,000 x g for 10 min at 4ºC.
Supernatant was saved and pellet resuspended in 1/50 volume of
supernatant. This suspension was freeze/thawed 3 times again and debris
repelleted. Both supernatants were combined and 20 ml loaded on a 10 ml
cushion of Hepes buffered CsCl (ρ = 1.4 g/ml). Virus was banded in the CsCl
cushion by centrifugation at 24,000 rpm for 3 hours at 4ºC in a SW28 rotor
(Beckman). The banded virus was isolated, checked for its density (ρ = 1.34
Chapter 2. A new intracellular pathway revealed
63
g/ml) and diluted 5 times into new Hepes buffered CsCl (ρ = 1.34 g/ml). This
suspension was centrifuged to equilibrium at 40,000 rpm for 16 hours at 4ºC
in a 70.1 Ti rotor (Beckman). The lower virus band was isolated and dialysed
extensively against 0.1 M carbonate buffer, pH 8.3. The virus was pelleted by
centrifugation at 50,000 rpm for 45 min at RT in a SW-55 Ti rotor (Beckman)
resuspended in 0.1 M carbonate buffer, pH 8.3 and stored in aliquots at –80ºC.
1 mg of virus (1 mg/ml) was labelled with 33 µl Texas Red-X-succinimidyl
ester or Fluorescein-X-succinimidyl ester (10 mg/ml in DMSO) according to
supplier’s instructions (Molecular Probes BV, Leiden, The Netherlands).
These fluorophores react exclusively with free amines, resulting in a stable
carboxamide bond and contain a seven-atom aminohexanoyl spacer (X),
which allows higher degrees of labelling without functional perturbance of
the virus. Labelled virus was repurified with CsCl as above, dialysed against
virion buffer (10 mM Hepes, pH 7.9, 150 mM NaCl, 1 mM CaCl2) and stored
in 2 µg aliquots at –20ºC.
Construction and expression of caveolin-1-GFP
Caveolin-1-GFP was constructed by PCR amplification of the canine caveolin1 cDNA from pBluescript-VIP21 (Kurzchalia et al., 1992) using a standard T3
primer and CGGTACCGTTGTTTCTTTCTGCATGTTGATGCG, followed by
cloning of the EcoR1/Kpn1 fragment from the PCR product into a pEGFP-N1
expression vector (Clontech Laboratories AG, Basel, Switzerland). The
resulting construct contains the canine caveolin-1 gene in frame and upstream
of the EGFP sequence with a spacer sequence of 30 nucleotides,
CCGCGGGCCCGGGATCCACCGGTCGCCACC, corresponding to the amino
acid sequence Pro-Arg-Ala-Arg-Asp-Pro-Pro-Val-Ala-Leu. CV-1 cells were
grown to 70-90% confluency on 12 mm Alcian Blue coated coverslips and
transiently transfected with 1.5-3 µg of plasmid DNA using superfect reagent
according to the manufacturer’s instructions (Qiagen AG, Basel, Switzerland),
resulting in app. 50% of transfected cells. Cells, which showed relative low
levels of expression, were analysed after 16-20 hours.
Western blots, sucrose gradients and immunoprecipitations
CV-1 cells in 6-cm dishes were transiently transfected with caveolin-1-GFP
and after 16 hours lysed in 10 mM Tris-Cl, pH 7.5 containing 1% Triton-X-100
64
Caveolae-mediated endocytosis of SV40
and a protease inhibitor cocktail. After lysis, nuclei were spun down at
5,000xg for 5 min at 4ºC and the postnuclear supernatant either analysed
directly by electrophoresis and immunoblotting or heated for 10 min at 37ºC
to dissolve raft domains completely. These lysates were loaded on top of a 550% linear sucrose gradient, containing 10 mM Tris-Cl pH 7.5 and centrifuged
at 4ºC for 18h at 38,000 rpm in a SW60 rotor (Beckman). 500 µl aliquots were
collected, TCA-precipitated and analysed for caveolin-1 and caveolin-1-GFP
by electrophoresis and immunoblotting. Immunoprecipitation of cell lysates
with an anti-GFP antibody was done as described previously (Pelkmans and
Helenius, 1999).
Immunofluorescence microscopy
CV-1 cells were grown to confluency (app 105 cells) on 12 mm Alcian Blue
coated coverslips and incubated with 1 µg (103 pfu/cell) of TRX-SV40 in Rmedium (RPMI 1640, 10mM Hepes pH 6.8, 0.2% BSA) for 2 hours at 4ºC. cells
were washed extensively with ice-cold R-medium and incubated in total
growth medium (DMEM, 10% FCS, 1x Glutamax, 1x PenStrep) at 5% CO2 and
37ºC. After indicated time-points, cells were fixed in 4% formaldehyde,
quenched with 50 mM NH4Cl and permeabilised with 0.05% (w/v) saponin
and subsequently incubated with appropriate primary and secondary
antibodies. Coverslips were mounted in Moviol containing DAPCO and
examined on a Leica confocal microscope (Leica DM IRBE) or a Zeiss Axiovert
wide-field microscope (see analysis and quantification of images).
Thin section electron microscopy
Fixation, embedding and sectioning of CV-1 cells was performed as described
previously (Kartenbeck et al., 1989).
pH dependent excitation scans of FLX-SV40
CV-1 cells on 6 cm dishes were incubated with 10 µg of FLX-SV40 or 5 mg of
FITC-dextran (MW: 42,000) for 2 hours at 4ºC in R-medium. Cells were
extensively washed with ice-cold R-medium and either directly resuspended
in PBS of different pH, or further incubated for 3 hours in normal growth
medium. Cells were transferred to a quartz cuvette with a small stirrer at RT
and excited with a fluorometer between 420 and 500 nm with 1-nm
Chapter 2. A new intracellular pathway revealed
65
increments. The emitted light at 520 nm was measured and plotted as light
intensity units. After this, 10 µM monensin and nigericin were added to the
same cuvette, incubated for 10 min at RT and the same scan was repeated.
Values obtained were averages of 4 independent experiments, which showed
less than 2% variation.
Time-lapse live fluorescence microscopy
TRX-SV40 was bound to CV-1 cells, or CV-1 cells expressing caveolin-1-GFP
and internalised as above. Labelling of endocytic/lysosomal structures was
done by binding TRX-SV40 in the presence of the appropriate markers
(Bodipy-FL-transferrin, FITC-dextran, FLX-SFV, Lysotracker green) and
subsequent internalisation as above. At indicated times, coverslips were
transferred to custom-built aluminium microscope slide chambers (workshop
Biochemistry/ETH) for live analysis in CO2-independent medium, placed on
a heated stage and analysed at 37ºC using wide-field or confocal microscopy.
For wide-field microscopy, cells were analysed with a Zeiss Axiovert
microscope using a 100x/1.40 plan-Apochromat lens. Images were collected
with a cooled charge-coupled-device (CCD) camera (Hamamatsu Inc.) at 3-sec
intervals using a computer-controlled shutter with standard FITC/Texas red
filter set and exposure times of 0.5-1 sec per image. For confocal microscopy,
cells were analysed with an inverted Leica microscope (DM IRBE) using a
100x/1.40 plan-Apochromat lens, and computer-controlled excitation with an
Argon laser at 568 nm (Texas red) and 488 nm (GFP). Signals were collected at
3-sec intervals with a photomultiplier tube (PMT), digitised (1024x1024 pixels)
and converted into images by the Leica TCS software. The amount of
photobleaching in each channel was analysed by calculating the total
fluorescence intensity in each frame of a movie sequence and was comparable
for both signals (20% for GFP and 25% for Texas Red).
Analysis and quantification of images and video sequences
For both wide-field and confocal microscopy, CV-1 and PtK2 cells (their
periphery is extensively spreaded and has a thickness of 1-1.5 µm) were first
focussed at the apical plasma membrane and this was set as the reference zaxis position of 0.0 µ m, and used for plasma membrane studies. For
intracellular analysis of cells, the z-axis position was computer-controlled set
66
Caveolae-mediated endocytosis of SV40
at –0.5 µm below the apical membrane (-0.5 µm). In confocal images, the
detector pinhole was set at 1 airy-disk unit (a.u.), resulting in a focal plane
thickness of app. 120 nm. In some cases, as indicated, the pinhole was set at
0.2 a.u. to further eliminate out-of-focus signal.
Processing and analysis of movies was performed using the Openlab software
package (v. 2.0.7, Improvision, Coventry, UK). For single image presentation
of single colour (TRX-SV40) movie sequences, all images were merged,
resulting in tracks that indicate movement. For single image presentation of
dual colour movie sequences (TRX-SV40 and EGFP-caveolin-1 or TRX-SV40
and YFP-α-tubulin), the TRX-SV40 signal in each image was coloured red, the
EGFP-caveolin-1 or YFP-tubulin signal in each image coloured green and all
images merged, resulting in tracks coloured red, green or yellow, indicating
movement of single-positive or double positive structures. Movements were
quantified by marking the position of a specific carrier in at least 20
consecutive images. Average and maximum speeds were calculated by
measuring the distance travelled between marked positions in 2 subsequent
images along the whole track. Values of speed were expressed as the mean of
at least 4 carriers/movie in at least 3 different cells.
Quantification of amounts of overlap in confocal images was done using
Adobe Photoshop (v. 5.0, Adobe Systems Inc.), by subtraction of the green
channel from the red channel, resulting in a picture of non-overlapping red
elements. Total intensity of the non-overlapping red signal was measured,
subtracted from the total red signal, and the resulting overlapping signal
presented as mean percentage of total red signal (of at least 4 different cells
per time-point). Quantification of bulk flow of virus signal was performed by
marking regions of interest in at least 4 different cells at each time-point (all
cells imaged at a z-axis position of –0.5 µm) corresponding to plasma
membrane, the cell periphery and a perinuclear area as described before
(Nakano and Greber, 2000). However, when a prominent accumulation near
the nucleus was observed, this whole structure was included in the
perinuclear region of interest. Total intensity was measured in regions of
interest and plotted as percentage of total cellular signal.
Quantification of fluorescence recovery after photobleaching (FRAP) was
performed by measuring the fluorescence intensities of the whole cell and of
the bleached area before, directly after and during recovery of bleaching using
Chapter 2. A new intracellular pathway revealed
67
the Openlab (v. 2.0.7) software. Relative fluorescence intensity of the bleached
area over time (R t) was calculated according to the following equation:
Rt =
Itotal ( o )
Ibleached ( 0 )
×
Ibleached ( t )
, whereby Itotal(0) is the total intensity of the cell before
Itotal ( t )
bleaching, Ibleached(0) is the total intensity of the bleached area before bleaching,
Ibleached(t) the intensity of the bleached area over time (so directly after bleaching
and recovery) and Itotal(t) the intensity of the whole cell over time. In this way
one corrects for overall bleaching during the experiment.
68
Caveolae-mediated endocytosis of SV40
2.5. Movies
Movie 1:
Dynamics of TRX-SV40 bound to the plasma membrane of CV-1 cells directly
after shifting to 37ºC recorded with wide field microscopy. The movie shows
a part of the membrane in which both mobile as stationary spots can be
discerned (recorded at 0.33 Hz, shown at 20 Hz, 50 frames). Note the mobile
spots (e.g. arrow) move randomly in the membrane. There is considerable
bleaching due to high exposure. Scalebar: 2 µm.
Movie 2:
Dynamics of TRX-SV40 and CAV1-GFP on the plasma membrane of CV-1
cells directly after shifting to 37ºC recorded with confocal laser scanning
microscopy. The movie shows part of the membrane in which TRX-SV40 (red)
virions are localised to CAV1-GFP positive microdomains (green), resulting in
yellow spots on the membrane (recorded at 0.33 Hz, shown at 20 Hz, 80
frames). Note that these spots are stationary in the membrane, but that three
of them (indicated by arrowheads) suddenly disappear, while others remain
in place. Note also that CAV1-GFP is present as a diffusive dynamic staining
between the spots. Scalebar: 2 µm.
Movie 3:
Dynamics of TRX-SV40 being internalised in CV-1 cells, incubated for 1 hour
at 37ºC after virus binding before recording with wide field microscopy
(recorded at 0.33 Hz, shown at 10 Hz, 50 frames). Note that small vesicles
gain in speed (arrow), in contrast with spots (of approximately same size) on
the membrane (asterisk), which have not been internalised yet. Some larger
vesicles are already visible (arrowhead). Note also that the movements of
vesicles do not share a common orientation. Scalebar: 7 µm.
Movie 4:
Dual colour live fluorescence laser scanning confocal microscopy experiment
recorded in CAV1-GFP transfected CV-1 cells 4 hours after virus binding and
shifting to 37ºC (recorded at 0.33 Hz, shown at 10 Hz). The movie shows part
of a cell (part of nucleus is visible in upper left corner) in which virus has
Chapter 2. A new intracellular pathway revealed
69
accumulated in CAV1-GFP containing organelles, but is also present in
unlabeled vesicles. Arrowheads indicate the formation of multi-domain
organelles and separation of virus from a CAV1-GFP containing vesicle. Note
how dynamic the vesicles are and how frequently they interact in a ‘kiss-andrun’ fashion. Note also that high-speed travelling structures are only positive
for virus (red). Scalebar: 2 µm.
Movie 4a:
Same as movie 4, especially focussed on tubule formation from a viruspositive vesicle (lower arrowhead), on which some CAV1-GFP is still located
(upper arrowhead). Note also that double positive vesicles (yellow, left upper
region) do not move. Scalebar: 2 µm.
Movie 4b:
Similar to movie 4, however shown at 20 Hz. This enlargement shows the
dynamics of sorting indicated by arrowheads. Visible are the separation of
green and red domains on one organelle (upper arrowhead) and the fast
sorting and detachment of virus-positive tubules from double positive
vesicles. Scalebar: 5 µm
Move 4c: Similar to movie 4b. Upper arrow, indicates sorting of virus, middle
arrow indicates partial fusion of one double positive vesicle with another
(upper arrow) and subsequent fusion. Right arrow indicates dynamics of
several small organelles, showing fusion and fission reactions. Scalebar: 5 µm.
Movie 5:
Dynamics of TRX-SV40 targeted to a perinuclear accumulation 6 hours after
virus binding (recorded at 0.5 Hz, shown at 10 Hz, 50 frames). Note the large
tubular carriers that can span a large part of the cell (arrows) and move both
towards and away from the perinuclear accumulation site. In the periphery,
the dynamics of tubule formation can be observed (white box, shown in
movie 7). Scalebar: 5 µm.
70
Caveolae-mediated endocytosis of SV40
Movie 5a:
White box in movie 5, showing the dynamic formation of tubules from a
vesicular structure (arrowhead). Note how tubules extend from the vesicular
structure and ‘search for their way out’. Scalebar: 2 µm.
Movie 5b:
The same as movie 5a, from another part of the same cell.
Movie 6:
Dual colour live fluorescence microscopy experiment recorded in PtK2 cells
expressing YFP-α-tubulin, 6 hours after virus binding and shifting to 37ºC
(recorded at 0.25 Hz, shown at 10 Hz). Note the intricate network of
microtubules. Virus has started to accumulate near the nucleus. Tubules
containing virus move along pre-assembled microtubule tracks (arrows).
Vesicular structures in the periphery are not moving (asterisk, bottom left).
There is considerable bleaching of YFP-α-tubulin. Scalebar: 10 µm.
Movie 6a:
Same as movie 6, only a selection of the upper left region movie 9. Scalebar: 10
µm.
Movie 6b and 6c:
Same as movie 6, focussed on vesicular structures from which tubules emerge
(arrowheads). The tubules grow along the aligning microtubules. Scalebar: 5
µm.
Movie 7:
Same as movie 6, but the cells have been treated with nocodazole throughout
the whole experiment. Note that virus has entered the cells and has
accumulated in vesicular structures, but does not move to a site near the
nucleus. Vesicular structures are dynamic, but show no net movement. Note
also the relative large amount of small vesicles. Scalebar: 10 µm.
Chapter 2. A new intracellular pathway revealed
71
Movie 7a:
Selection of movie 7 (see white box) indicating that fission of bigger vesicular
structures occurs resulting in the formation of multiple small vesicles (arrow),
instead of larger tubules. Scalebar: 2 µm.
Movie 8:
Fluorescence Recovery After Photobleaching (FRAP) experiment in CV-1 cells
recorded with laser scanning confocal microscopy (recorded at 0.017 Hz,
shown at 10 Hz) 6 hours after virus binding and shifting to 37ºC. The movie
shows a cell which has accumulated a large amount of virus at the perinuclear
site, but still contains a lager amount of carriers. Note that in the second
frame, the prominent accumulation (indicated by yellow line) completely
disappeared (due to bleaching with a high power laser beam) and
subsequently recovers with newly targeted virus. Note also the similar
morphology of the structure before bleaching and after recovery. Scalebar: 5
µm.
Acknowledgements
We thank all lab members for helpful discussions and suggestions throughout
the work, M. Kowarik, K. Breiner and M. Molinari for critical reading of the
manuscript and A. Mezzacasa for help with the microscopes. We also thank
the following people for providing reagents: Ulla Lathinen for cDNA
encoding caveolin-1, Derek Toomre for Ptk2 cells expressing YFP-α-tubulin
and M. Steegmaier for antibodies against syntaxin17. This work was
supported by the Swiss National Science Foundation and by ETHZ.
Chapter 3
Local actin polymerization and dynamin recruitment
in SV40-induced internalization of caveolae
Lucas Pelkmans, Daniel Püntener and Ari Helenius
Institute of Biochemistry, Swiss Federal Institute of Technology (ETH) Zürich
Science, volume 296, 19 April 2002, pages 535-539
74
Caveolae-mediated endocytosis of SV40
Abstract
Simian Virus 40 is a non-enveloped dsDNA virus of the polyomaviridae that
utilizes endocytosis through cell surface caveolae for infectious entry. To
understand this uptake mechanism better, we investigated the early events of
entry and observed that after binding to caveolae, the virus particles induced
a transient breakdown of actin stress fibers. Actin was recruited to the virusloaded caveolae as small actin patches that served as sites for the formation of
actin ‘tails’. Dynamin2 was also transiently recruited to virus-loaded caveolae.
These virus-triggered events depended on the presence of cholesterol and on
the activation of tyrosine kinases that phosphorylated proteins in caveolae.
They were necessary for formation of caveolae-derived endocytic vesicles and
their transport to caveosomes, as well as for virus infection. The results
showed that caveolar endocytosis is a ligand-triggered event that involves
extensive rearrangement of the actin cytoskeleton.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
75
3.1. Introduction
Simian Virus 40 (SV40), a non-enveloped dsDNA virus of the papova family,
is well known for its ability to cause tumors in laboratory animals and to
induce neoplasmic transformation of cells in culture (Norkin, 1999). Although
the molecular basis for these processes are well characterized, less is known
about the cell biology of replication. Recent studies have shown that SV40
deviates from most other animal viruses in that it uses cell surface caveolae
for entry (Anderson et al., 1996; Stang et al., 1997). Caveolae are flask-shaped
plasma membrane invaginations enriched in cholesterol, sphingolipids, GPIanchored proteins, and integral membrane proteins called caveolins
(Anderson, 1998; Palade, 1953). Their functions are thought to include
transcytosis, signal transduction, and the uptake of plasma membrane
components and external cellular ligands (Anderson, 1998).
The process by which SV40 enters the host cell occurs in several steps
(Anderson et al., 1996; Breau et al., 1992; Kartenbeck et al., 1989; Pelkmans et
al., 2001; Stang et al., 1997; Yamada and Kasamatsu, 1993). First, the virus first
binds to MHC Class I antigens and moves laterally in the membrane until
trapped in caveolae. The virus-loaded caveolae pinch off from the plasma
membrane as small, tight-fitting vesicles containing a single virus and with
caveolin-1 still attached. These vesicles then fuse with larger, preexisting
organelles, caveosomes, which can accumulate numerous viruses. From the
caveosomes, the viruses are sorted into larger, often tubular membrane
extrusions that detach as vesicles and move with the help of microtubules to
the smooth endoplasmic reticulum (ER) where they fuse. The viruses exit the
ER lumen and enter the nucleus most likely via the cytosol, through nuclear
pore complexes. Thus, SV40, unlike most other viruses, bypasses the classical
endosomal pathway during infectious entry.
In this study we have addressed the question, whether the virus actively
induces caveolar endocytosis, and if so, what the involved signals and cellular
components are. While there is good evidence that caveolae have the capacity
to pinch off as endocytic vesicles and internalize a variety of ligands
(Anderson, 1998; Nichols and Lippincott-Schwartz, 2001), the process does
76
Caveolae-mediated endocytosis of SV40
not seem to occur efficiently under normal culture conditions. Uptake has
usually been seen only after some form of perturbation such as cross-linking
of GPI anchored proteins combined with phosphatase inhibitor treatment,
incubation in hypertonic medium, or after addition of SV40 or Polyoma Virus
(Aoki et al., 1999; Kang et al., 2000; Parton et al., 1994; Richterova et al., 2001;
Stang et al., 1997). Using live fluorescence microscopy we recently observed
efficient uptake of individual SV40-loaded caveolae whereas empty caveolae
stayed at the plasma membrane (Pelkmans et al., 2001). This suggested that
the virus particle might induce its own uptake.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
77
3.2. Results
3.2.1. Internalization of SV40 depends on active kinases, an intact actin
cytoskeleton and dynamin2.
To follow the internalization process quantitatively, we prepared radioiodinated SV40 to which we coupled biotin via a spacer containing a disulfide
bond (Sulfo-NHS-SS-Biotin) (see methods). Control experiments showed that
the modified virus followed the normal pathway to the ER, and that it
retained one third of the original infectivity. After saturating the surface with
roughly 1,300 virus particles per cell at 4°C, CV-1 cells were incubated at 37°C
for different times to allow internalization, without washing away unbound
virus. To remove the biotin specifically from the surface-exposed fraction of
virus particles, the cells were then treated with the membrane-impermeable
reducing agent Tris(2-carboxyethyl)phosphine (TCEP). After quenching
excess reductant by alkylation, the cells were solubilized with detergent, and
the lysates were subjected to immunoprecipitation with anti-biotin antibodies
to isolate the biotin-containing, internalized fraction of viruses.
The assay showed, that internalization of surface-bound SV40 was efficient,
but relatively slow (Fig. 1A, 1st graph). Uptake started after a lag time of 20
min and continued for about 3 hours with half maximal internalization at 90
min. Maximal internalization leveled off at about 1,500 particles per cell
although plenty of unbound viruses were still present in the medium, which
was somewhat unexpected. Apparently, there was only a single cycle of
uptake.
Next, the effects of potential inhibitors on internalization were tested. They
were all used at concentrations that were non-toxic to the cells as determined
by colony formation of the drug-treated cells (see methods). The combination
of the cholesterol-sequestering drug Nystatin (Simons and Toomre, 2000) and
the cholesterol synthesis inhibitor Progesterone (Smart et al., 1996)
(Nys/Prog), which reduced cellular free cholesterol levels to 30% (assayed
with Amplex Red, not shown), blocked virus uptake almost completely (9095%) (Fig. 1A, 1st graph). This was expected since cholesterol depletion is
known to block SV40 infection (Anderson et al., 1996) (see also Fig. 1C).
78
Caveolae-mediated endocytosis of SV40
Staurosporin (STP), a general kinase inhibitor, and Genistein, a tyrosine
kinase inhibitor, also known to reduce infection (Chen and Norkin, 1999) (Fig.
1C), were found to reduce SV40 uptake by 70% (Fig. 1A, 2nd graph). Okadaic
acid (OA), a general phosphatase inhibitor (Cohen et al., 1990), and vanadate,
a tyrosine phosphatase inhibitor (Rozelle et al., 2000), enhanced uptake and
eliminated the lag in the uptake process (Fig. 1A, 3rd graph). LatrunculinA
(LatA), an actin monomer sequestering drug (Coue et al., 1987), and
Jasplakinolide (Jas), an actin-polymer stabilizing drug (Bubb et al., 1994),
reduced virus internalization by 60-65% (Fig. 1A, 4th graph). The effects on
SV40 internalization observed for LatA roughly mirrored its effect on
infection, measured by the fraction of cells expressing T antigen after a 20h
infection period (Fig. 1C). Jas however, blocked virus infection much more
efficiently then LatA, suggesting that it additionally interfered with a later
step in the entry process (Fig. 1C). None of the inhibitors were found to
reduce initial binding of virus to the cells (Fig. 1B).
The cholesterol depletion effect was most likely caused by the loss of rafts and
caveolae (Simons and Toomre, 2000). The effects of Genistein, STP, vanadate
and OA implied involvement of a tyrosine kinase in the internalization
process. Since the phosphatase inhibitors increased the amount of internalized
virus well beyond the single cycle (160% for OA and 220% for vanadate) it is
moreover possible that phosphorylation not only promoted internalization of
the virus, but also supported recycling of essential cellular components
needed in the uptake process. This was consistent with the observation that
they slightly increased initial binding (Fig. 1B). Finally, from the effects of
LatA and Jas, we concluded that actin played a key role.
In order to test the effect of dominant-negative constructs in the
internalization process, we also developed an assay for internalization that
could be used for quantitative analysis of single, transfected cells. When
fluorescein-labeled virus is used, the internalized fraction can be selectively
visualized by adding low pH buffer to the extracellular space, thus
eliminating the fluorescence from viruses on the cell surface and in caveolae
(Pelkmans et al., 2001). Quantification was achieved by measuring pixel
intensities of the fluorescein signal in single cells with fluorescence
Chapter 3. Uptake is ligand-triggered via actin and dynamin
79
Fig. 1. Internalization of SV40 is dependent on active kinases, an intact actin cytoskeleton and
Dynamin 2
(A) Drug-treated or control CV-1 cells in suspension were incubated with Biotin-SS-SV40-125I for 2
hours at 4°C. Cells were subsequently shifted to 37°C in continuous presence of Biotin-SS-SV40-125I.
At different time points, cells were washed and spun down, treated twice with TCEP at 4°C, quenched
with iodoacetamide, lysed, and immunoprecipitated with Protein A coated beads and anti-biotin
antibodies. The amount of precipitated Biotin-SS-SV40-125I was background-corrected and expressed as
percentage of initially bound Biotin-SS-SV40-125I (results obtained from 3 independent experiments).
STP, Genistein, LatA and Jas all inhibit SV0 internalization, while OA and Vanadate accelerate
internalization. (B) Drug-treated CV-1 cells in suspension where incubated with 125I-SV40 for 2 hours at
4°C under saturating conditions, washed and the cell-associated radioactivity determined. Values (from
3 independent experiments) are expressed as percentage of non-treated cells. 20x excess of unlabeled
SV40 was used as a negative control (SV40). Treatment with OA and vanadate slightly increase
binding, while other drug treatments have no influence. (C) CV-1 cells on coverslips were pretreated
with the indicated drugs, infected with unlabeled SV40 at a multiplicity of infection of 10 in continuous
presence of the drugs and stained with antibodies against T antigen 20 hours after infection. In control
experiments, the drugs were washed out after 20 hours and cells stained against T antigen 20 hours after
the washout. Amount of T antigen-expressing cells from 4 randomly chosen fields are presented as
percentage of untreated cells (which is 50% of total number of cells). (D) CV-1 cells co-tranfected with
DsRed-N1 and Dyn2wt, Dyn2K44A, Eps15DIII∆2 (Eps15ctr) or Eps15E∆95/295 (Eps15mut) were
incubated with FLX-SV40 for 3.5 hours at 37°C and analyzed with fluorescence microscopy in a
medium of pH 4.5. The fluorescein signal of 18-25 cells expressing the cDNA was quantified and
expressed as percentage of the signal in untreated cells. Representative images are shown on the left. To
analyze infection, cells were fixed 20 hours after virus binding and immunostained with antibodies
against T antigen. Amount of T antigen-expressing cells from 4 randomly chosen fields are presented as
percentage of non-transfected cells in the same field. Only Dyn2K44A inhibits SV40 internalization and
infection. Scalebar, 50 µm.
80
Caveolae-mediated endocytosis of SV40
Fig. 1. continued
(D) Drug-treated CV-1 cells on coverslips were incubated
with FLX-SV40 for 3.5 hours at 37°C in the continuous
presence of the drugs. Cells were incubated in medium of pH
4.5, which quenches all extra-cellular virus signal and
analyzed with fluorescence microscopy. Representative
images are shown on the left. The fluorescence signal of 38-65 cells was quantified from images with
low exposure times to ensure a linear range of intensities (12bit 1-4095) using Openlab 3.0.4
(Improvision) and expressed as percentage±standard deviation of the signal in non-treated cells.
Scalebars, 20 µm. (E) CV-1 cells co-tranfected with DsRed-N1 and Dyn2wt, Dyn2K44A,
Eps15DIII∆2 (Eps15ctr) or Eps15E∆95/295 (Eps15mut) (41) were incubated with FLX-SV40 for 3.5
hours at 37°C and analyzed with fluorescence microscopy in a medium of pH 4.5. The fluorescein
signal of 18-25 cells expressing the cDNA was quantified and expressed as percentage of the signal
in untreated cells. Representative images are shown on the left. To analyze infection, cells were fixed
20 hours after virus binding and immunostained with antibodies against T antigen. Amount of T
antigen-expressing cells from 4 randomly chosen fields are presented as percentage of nontransfected cells in the same field. Only Dyn2K44A inhibits SV40 internalization and infection.
Scalebar, 50 µm.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
81
microscopy (see methods). The results using this assay showed that
internalization of SV0 was similarly disturbed by the drugs as observed in the
biochemical assay, described above (Fig. 1D). To identify transfected cells, we
co-expressed a red fluorescent protein (DsRed) together with the protein of
interest. Since Dynamin2, a member of the dynamin-GTPase family, is known
to be involved in the internalization of clathrin coated pits and caveolae
(Henley et al., 1998; Oh et al., 1998), we tested a dominant negative dynamin2
construct with a mutation in the active GTPase site (dyn2K44A) (Fish et al.,
2000). The expression of this nonfunctional GTPase drastically reduced
endocytosis of the virus and resulted in the loss of infectivity (Fig. 1E). As a
control, we used a dominant-negative, truncated mutant of Eps15 (E∆95/295),
known to specifically block clathrin-coated pit assembly (Benmerah et al.,
1999). It did not block uptake or infection of SV40 (Fig. 1E). These
observations indicated that dynamin2 was involved in the infectious uptake
of SV40 via caveolae.
3.2.2. Caveolae-sequestered SV40 particles are entrapped in the membrane
by the actin cytoskeleton.
Alexa Fluor 594-labeled virus particles (AF-SV40) (40) were next used to
follow the uptake process at the level of individual virus particles (Pelkmans
et al., 2001). We allowed AF-SV40 to bind to cells at 4°C, and tracked their
movements after warming to 37˚C using fluorescence microscopy and timelapse video recording (see methods). Immediately after warming, the virus
particles were found to move laterally in the plasma membrane in random
pathways with an average speed of 0.17 µm/sec and an average range of 4.04
µm (Fig. 2A, movie 1). The movement abruptly stopped within the first 10
minutes as the viruses became fixed in place (Fig. 2A, movie 2, 3and 3A). The
viruses are now arrested in small, immobile caveolin-1-GFP-labeled spots,
which correspond to caveolae (Pelkmans et al., 2001).
To investigate the involvement of actin in the sequestration process, we
recorded trajectories of mobile virus particles in LatA- or Jas-treated cells. Jas
did not have any detectable effect on virus particle movement or colocalization with caveolin-1-GFP (Cav1-GFP) (Fig. 2A,B). However, in the
presence of LatA, the virus particles moved somewhat faster than in control
82
Caveolae-mediated endocytosis of SV40
cells (average speed of 0.24 µm/sec) and they failed to stop moving (Fig. 2A,
movie 4), although they had reached the cav1-GFP positive spots (Fig. 2B).
Evidently, the viruses were trapped in caveolae, but instead of being
stationary these were now mobile (compare movie 5 and 5A). Also, in cells
that had not bound SV40, LatA caused the caveolae to become laterally
mobile (compare movie 6 and 6A). A similar effect has been described for
clathrin-coated pits in LatA-treated cells (Gaidarov et al., 1999). We concluded
that actin filaments are not necessary for the diffusion of virus particles on the
membrane and their sequestration into caveolae, but that they play a role in
restricting the lateral mobility of caveolae in the plasma membrane.
3.2.3. Caveolae-sequestered SV40 induces the formation of actin tails.
To follow actin more directly, we made use of cells expressing GFP-β-actin.
We observed that between 10 and 20 min after initiating virus entry, the
number of actin stress fibers in the cell decreased dramatically. Instead, small
actin patches and tails appeared on the upper surface of the cell where most
virus particles were located (Fig. 3A, movie 7). These changes were all
transient; after 120 min, the actin cytoskeleton was back to the normal pattern.
The actin tails were on average 1.3 µm long, and one end was generally
tapered. Repetition of the experiment with AF-SV40 showed that the majority
of tails (65±11%) had a virus particle at their thick end (Fig.3B). Live
recordings revealed that the tails used virus-associated actin patches as
polymerization sites, and that the tails were quite dynamic (Fig. 3C and movie
7A, B). In cells expressing YFP-β-actin and cav1-CFP, it became clear, that the
sites for actin tail formation corresponded to cav1-CFP domains, which had
sequestered an AF-SV40 particle, namely to virus-loaded caveolae (Fig. 3D).
Whether the tails were present also on departing vesicles could not be
resolved. However, the caveosomes, into which the incoming viruses
accumulated, were devoid of tails (Fig.3B). The dissociation of stress fibers
and the formation of actin tails were clearly SV40-induced; when fewer virus
particles were added, the changes were less prominent (Fig. 3A). In control
cells without virus, no changes in the actin distribution were observed (movie
8).
Chapter 3. Uptake is ligand-triggered via actin and dynamin
83
Fig. 2. The actin cytoskeleton traps SV40 particles in the membrane after sequestration into
caveolae.
(A) AF-SV40 particles were bound to CV-1 cells at 4°C and analyzed with live fluorescence
microscopy at 37°C. From 100 seconds long recordings, moving virus particles were tracked. 3
representative tracks are depicted on the right. The maximum range of trajectories were calculated
and presented on the left. Time after warming to 37ºC is indicated in min:sec. Directly after virus
binding, virus particles are mobile in the membrane (0:00-1:40) but are rapidly trapped in place. 10
minutes after binding, most virus particles are fixed in the membrane (10:00-11:40). When cells are
treated with LatA, particles are not trapped in place (10:00-11:40+LatA). Pretreatment of cells with
Jas has no effect on virus particle entrapment (10:00-11:40 min + Jas). Scalebar, 2 µm. (B) AF-SV40
particles were bound to cav1-GFP expressing CV-1 cells at 4°C. After 20 min of incubation at 37°C,
cells were fixed and analyzed with fluorescence microscopy. Representative enlargements of the
plasma membrane are depicted on the left. The extent of co-localization between AF-SV40 signal and
cav1-GFP signal was calculated and presented as percentage of plasma membrane-bound AF-SV40
signal overlapping with cav1-GFP signal. Treatment with LatA or Jas does not influence the colocalization of AF-SV40 with cav1-GFP. Scalebars, 2.5 µm.
84
Caveolae-mediated endocytosis of SV40
Fig. 3. SV40 induces the formation of actin tails on virus-loaded caveolae in the plasma
membrane.
(A) SV40 was bound to GFP-b-actin expressing CV-1 cells for 2 hours at 4°C, shifted to 37°C for
different periods of time, fixed and analyzed with fluorescence microscopy. Initially, most GFP-βactin is distributed to stress fibers (0 min). Next, small actin foci appear on the dorsal site of the cell
(10 min), and subsequently small actin tails are observed and the amount and intensity of stress fibers
is reduced (20 min). After further incubation, actin tails disappear and stress fibers reappear (120
min). CV-1 cells were incubated with SV40 at different multiplicities of infection (MOI) at 4°C,
shifted to 37°C for different periods of time, fixed and scored for the presence of actin tails on the
dorsal membrane by fluorescence microscopy (graph). Between 60-75 cells from 3 different
experiments were counted and data are expressed as percentage of cells that showed actin tails.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
85
Fig. 3. continued.
Between 20-30 minutes after virus binding, app. 70% of cells have actin tails. 2 hours after virus
binding, the actin cytoskeleton has returned to a normal pattern. At lower MOI, the actin cytoskeleton
returns to a normal pattern sooner and less cells show actin tails. Scalebars, 10 µm. (B) Closer
examination of plasma membrane regions of GFP-β-actin expressing CV-1 cells that have bound AFSV40 and were incubated at 37ºC for 20 minutes show that the actin tails co-localize with one end to
virus particles (arrowheads). After 120 minutes, AF-SV40 is in larger caveosomes, which do not have
actin tails. Scalebars, 2 µm. (C) Selected frames (time indicated relative to first frame in
min:sec:msec) of live fluorescence recordings show, that actin tails polymerize from sites where virus
particles are attached to the membrane (arrowheads) Scalebar, 5 µm. (D) Closer examination of
plasma membrane regions of CV-1 cells expressing YFP-β-actin and cav1-CFP that have bound AFSV40 at 4°C and were shifted to 37°C for 20 min before fixation, show that actin tails are localized to
cav1-CFP domains which have sequestered AF-SV40 particles (arrowheads). Scalebar, 3 µm.
3.2.4. SV40-induced actin tail formation depends on functional caveolae and
active tyrosine kinases.
To study whether actin tail formation was indeed dependent on virus
reaching the caveolae, we depleted cells of cholesterol with Nys/Prog, and
found that the redistribution of actin was completely inhibited (Fig. 4A, B).
Since co-localization studies showed that Nys/Prog inhibited formation of
cav1-GFP domains and impaired localization of virus with caveolae (Fig. 4C),
the result suggested that SV40 has to associate with caveolae or rafts to induce
changes in actin,. Treatment with STP and Genistein also inhibited the SV40induced changes in the actin cytoskeleton (Fig. 4A, B). Thus, kinases seemed
to be involved in virus-induced triggering of the observed changes in the
cytoskeleton. That OA and vanadate accelerated the formation of actin tails
when cells were exposed to SV40 (Fig. 4A, B) was consistent with this. Note
that OA and vanadate by itself induced the loss of actin stress fibers (Fig. 4A),
but did not induce actin tails.
Because the actin inhibitors and the kinase inhibitors did not interfere with
SV40 sequestration into caveolae (Fig. 2B and 4C), it was likely that kinase
activation only occurred when SV40 was trapped. Immunofluorescence using
antibodies against phosphorylated tyrosines (see methods) in cells expressing
YFP-β-actin and cav1-CFP that had bound SV40 showed that those caveolae,
that had actin tails were indeed immunostained (Fig. 5A, +SV40). In control
cells without virus, phosphotyrosine staining was predominantly observed at
focal adhesion contacts, and not on cav1-CFP domains (Fig. 5A, -SV40).
Binding AF-SV40 to cells expressing cav1-GFP revealed that only caveolae
86
Caveolae-mediated endocytosis of SV40
that contained viruses labeled for phosphotyrosines (Fig. 5B). This indicated,
that SV40 induces local tyrosine phosphorylation in caveolae.
3.2.5. Depolymerization/polymerization of actin and transient recruitment
of dynamin2 during SV40 internalization.
Since treatment with LatA and Jas reduced virus internalization, it was of
interest to determine whether these inhibitors interfered with the virusinduced changes in the actin cytoskeleton. When cells were pretreated with
LatA, actin patches were recruited to plasma membrane-bound virus
particles, but actin tail formation could not be detected (Fig. 6A, upper row).
In control cells without virus, no actin patches were observed (not shown).
Thus, actin patch formation is induced by SV40 and does not involve actinactin interactions while tail formation does. Live recordings showed, that the
virus-actin patches were mobile ( movie 9). Pretreatment with Jas, in contrast,
inhibited the dissociation of actin stress fibers, blocked the formation of small
actin patches, and prevented the polymerization of actin tails (Fig. 6A, lower
row). Using the ‘single cell’ assay for internalization described above, we
observed in higher magnifications of LatA-treated cells in low pH medium,
that the fraction of viruses that was internalized (app. 30% of non-treated
cells, see Fig. 1A) was transported to caveosomes (Fig. 6B, left panel), and
were evidently able (although less efficient) to infect the cell (see Fig. 1C). This
suggested, that formation of an actin tail promoted efficient closure of
caveolae, but was not necessary for transport of caveolar vesicles to
caveosomes. In contrast, virus particles that were internalized in the presence
of Jas did not move far into the cytosol. They remained just below the cell
surface (Fig. 6B, right panel), which may explain why they failed to infect the
cell (see Fig. 1C). The most likely interpretation of these results was, that the
presence of a stabilized cortical actin cytoskeleton prevented penetration of
virus-containing vesicles into the cytosol.
Finally, when cells expressing functional GFP-labeled dynamin2 (Dyn2-GFP)
(Cao et al., 1998) were studied, we observed that in control cells, Dyn2-GFP
was visible at the surface only for brief periods of time (average 8 sec) as
bright spots, giving the impression of ‘blinking’ lights in the video recording
(movie 10). Dyn2K44A-GFP in contrast, was not observed to blink,
Chapter 3. Uptake is ligand-triggered via actin and dynamin
87
Fig. 4. SV40-induced actin tail formation is dependent on functional caveolae and active
tyrosine kinases.
(A) CV-1 cells expressing GFP-actin were pretreated with Nys/Prog, STP, Genistein, OA, or
Vanadate, subsequently allowed to bind AF-SV40 at 4°C in the presence of the drugs and after that
shifted, in the presence of the drugs, for 20 min to 37°C. Images show, that cholesterol depletion with
Nys/Prog, kinase inhibition with STP and tyrosine kinase inhibition with Genistein block the
disappearance of stress fibers and the formation of actin patches and actin tails. Phosphatase
inhibition with OA or tyrosine phosphatase inhibition with vanadate results in reduced stress fiber
staining in cells without SV40 and in heavy formation of actin tails in cells having bound SV40.
Scalebars, 5 µm. (B) Quantification of the amount of drug-treated CV-1 cells with actin tails (18-30
randomly chosen cells of 2 independent experiments) at different time points after virus binding
shows, that cholesterol depletion and inhibition of tyrosine kinases completely blocks the formation
of actin tails, while tyrosine phosphatase inhibition accelerates actin tail formation. (C) Drug-treated
CV-1 cells expressing cav1-GFP having bound AF-SV40 at 4°C were shifted to 37°C for 20 min,
fixed and the amount of AF-SV40 signal overlapping with cav1-GFP signal was determined (42).
Only Nys/Prog treatment reduces the amount of overlap between AF-SV40 and cav1-GFP.
88
Caveolae-mediated endocytosis of SV40
Fig. 5. SV40 particles induce tyrosine phosphorylation on virus-loaded caveolae.
(A) Immunostaining of phosphotyrosines in YFP-b-actin and cav1-CFP expressing CV-1
cells that were either untreated (-SV40) or had bound unlabeled SV40 at 4°C and were
shifted to 37°C for 20 min before fixation (+SV40) (41). In control cells, phosphotyrosine
staining is predominantly localized to focal adhesion sites and not to cav1-CFP domains. In
cells having bound SV40, phosphotyrosine staining on focal adhesion sites has disappeared
and is now visible on cav1-CFP microdomains that have recruited actin tails (arrowheads).
Scalebars, 5 µm and 2 µm (enlargements). (B) Immunostaining of CV-1 cells, expressing
caveolin-1-GFP that have bound unlabeled SV40 at 4°C and shifted to 37°C for 20 min
before fixation against phosphotyrosines, shows that cav1-GFP domains that contain an
SV40 particle are stained against phosphotyrosines. Overlap of PTYR (blue), cav1-GFP
(green) and AF-SV40 (red) results in white domains (arrowheads). Scalebars, 5 µm and 2 µm
(enlargements).
Chapter 3. Uptake is ligand-triggered via actin and dynamin
LatA
89
Jas
Fig. 6. Effects of LatA
and Jas on actin
recruitment and release
of
SV40-containing
vesicles into the cytosol
and live analysis of
Dynamin2 recruitment
to membrane-bound
SV40 particles
(A) Closer examination of
plasma membrane regions
of CV-1 cells expressing
GFP-β-actin that have
bound AF-SV40 at 4°C
and shifted to 37°C for 20 min in the presence of LatA or Jas show that small actin patches are still
recruited to virus particles on LatA-treated cells (arrowheads) but not on Jas-treated cells. Also stress
fibers have not disappeared in Jas-treated cells. Scalebars, 5 µm. (B) Higher magnification images of
LatA-treated or Jas-treated CV-1 cells incubated for 3.5 hours with FLX-SV40 at 37°C and incubated
in a medium of pH 4.5. The small amount of unquenched signal in LatA-treated cells is in larger
more perinuclear located organelles (enlargement), while in Jas-treated cells internalized virus is
visible as single particles trapped in the periphery (enlargement). Scalebars, 20 µm and 5 µm
(enlargements). (C) Live fluorescence recordings of Dyn2-GFP expressing CV-1 cells that have
bound AF-SV40 were performed. First, an image of AF-SV40 signal was taken and subsequently
rapid recordings of the Dyn2-GFP signal. Selected frames (time indicated relative to first frame in
min:sec:msec) show that Dyn2-GFP is recruited in a blinking manner to the site where virus particles
are bound to the membrane (red circles) (control). Genistein inhibits the recruitment of Dyn2-GFP to
membrane-bound virus particles (red circles), but not to other sites on the membrane (arrowheads)
(+Genistein). Scalebars, 2.5 µm.
90
Caveolae-mediated endocytosis of SV40
but rather stayed as permanent bright spots on the membrane (not shown).
When AF-SV40 particles were present on the cell surface, about 10% of the
Dyn2-GFP spots were found to co-localize with them in a blinking manner
(Fig. 7 no treatment, movie 11). When Genistein was added to block the virusinduced signals, co-localization of AF-SV40 and Dyn2-GFP and the blinking
of Dyn2-GFP on virus particles was inhibited (Fig. 7 +Genistein, movie 12).
This indicated that the GTPase was recruited to membrane-bound viruses
through tyrosine kinase activation.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
91
3.3. Discussion
Taken together, our results showed, that SV40 induces its own endocytic
internalization. Once sequestered into caveolae, the virus particles triggered
local tyrosine phosphorylation, and a complex series of transient changes in
the caveolae and in the actin cytoskeleton are set in motion. The first change
to be observed was the dissociation of filamentous actin. This was followed by
recruitment of dynamin2 and actin to the cytosolic surface of the caveolae.
Actin was recruited as a patch, which served as the site for actin tail
formation. These events were important for the closure of the caveolae by
membrane fission and for the free passage of vesicles deeper into the
cytoplasm. We found no evidence for actin-mediated propulsion of virusloaded caveolar vesicles to caveosomes.
Actin has been shown to play an important role in phagocytosis (May and
Machesky, 2001) and the internalization of clathrin-coated pits (Brodsky et al.,
2001). It may also be involved in the intracellular movement of pinocytic
vesicles (Merrifield et al., 1999), endosomes, and vesicles derived from the
Trans Golgi Network (Rozelle et al., 2000; Taunton et al., 2000). Actin tails,
like those seen here, also play a role in the inter- or intracellular mobility of
several pathogens, such as Listeria, Shigella and Rickettsia (Dramsi and
Cossart, 1998). The induction of actin tails observed here for SV40 resembled,
in many respects, the well-characterized induction of actin tails by vaccinia
virus (Frischknecht and Way, 2001). Both processes are mediated by virus
particles bound to the outside surface of the plasma membrane, and both
depend on tyrosine phosphorylation (Frischknecht et al., 1999a; Ward and
Moss, 2001). Clearly, there are also differences: SV40 induces actin tails for cell
entry whereas vaccinia virus induces them for exit and transfer to
neighbouring cells (Ward and Moss, 2001). Furthermore, in the case of SV40,
the recruitment of actin must depend on cellular proteins located in caveolae,
while vaccinia virus makes use of a virus-encoded transmembrane protein
(Frischknecht et al., 1999b). It will now be important to identify the SV40activated kinases and their substrates, and to elucidate the signalling
pathways that SV40 uses to activate caveolae, recruit dynamin and reorganize
actin. A detailed analysis of this endocytic pathway and its regulation is
92
Caveolae-mediated endocytosis of SV40
clearly important because it is not only used by SV40 and other viruses
(Marjomaki et al., 2002; Richterova et al., 2001), but also by bacteria and
bacterial toxins (Shin and Abraham, 2001; van der Goot and Harder, 2001). It
also remains to be determined to what extent this pathway is involved in the
internalization of endogenous ligands.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
93
3.4. Methods
Iodination of SV40 and virus binding assays
For virus binding experiments purified SV40 was radiolabeled with
125
I
(Amersham) using chloramine T. Unbound radioactivity was removed by gel
filtration and sucrose gradients. A specific activity of 3x106 CPM/µg of viral
protein (corresponding to 38.5 CPM/virus particle) was obtained and virus
retained full infectivity (1x108 PFU/µg viral protein). To standardize our
experiments (36) drug-treated or untreated CV-1 cells were resuspended with
trypsin/EDTA and incubated at a density 106 cells/ml in R-medium (RPMI
1640, 10 mM Hepes pH 6.8, 0.2% BSA) with or without the drugs. Saturating
amounts of radiolabeled virus (106 CPM, of which 5% binds in control
situations) were added and cells rotated end-over-end for 2 hours at 4°C.
Control experiments performed on confluent CV-1 cells on 6-cm dishes, or on
cells brought in suspension by either scraping or treatment with 5 mM EDTA
showed, that binding efficiencies were not affected by treatment with
trypsin/EDTA, that receptors could be saturated, and that unlabeled virus
could compete with the binding of radiolabeled virus in equimolar
concentrations. After virus binding, 50 µl aliquots were taken and diluted into
1000 µl ice-cold NT buffer (50 mM Tris-Cl pH 8.6, 100 mM NaCl). Cells were
pelleted, supernatant completely removed, and cell-associated radioactivity
counted.
Generation of Biotin-SS-SV40-125I and virus internalization assays
For internalization assays purified SV40 was first labeled with Sulfo-NHS-SSBiotin (Pierce), before being iodinated (Biotin-SS-SV40-125I). We estimated that
app. 360 biotin molecules were coupled to one virus particle. Biotinylation
had no significant effect on virus infectivity (3x107 PFU/µg versus 108
PFU/µg of viral protein). 90% of the radioactivity could be specifically
immunoprecipitated with anti-biotin antibodies (Bethyl laboratories) or antiSV40 antiserum (Pelkmans et al., 2001). When the virus was treated with
Tris(2-carboxyethyl)phosphine (TCEP) (Pierce), only 5% of the radioactivity
(background) was precipitated with anti-biotin antibodies, while anti-SV40
antiserum still precipitated 90% of the radioactivity. After Biotin-SS-SV40-125I
was bound, CV-1 cells were shifted to 37°C and at the indicated time points,
94
Caveolae-mediated endocytosis of SV40
50 µl aliquots were taken and quickly diluted into 1000 µl ice-cold NT buffer.
Cells were pelleted and the amount of internalized Biotin-SS-SV40-125I was
determined as described (Schmid and Carter, 1990), except that 2x 15 mM of
the membrane-impermeable reducing agent TCEP was used and protected
Biotin-SS-SV40-125I was precipitated with Protein A beads and anti-biotin
antibodies. Background values (less then 10% of cell-associated radioactivity
at 4°C that was TCEP resistant) were subtracted from the results, which were
then expressed as the percentage of initially surface-bound Biotin-SS-SV40-125I
that was internalized.
Drug treatments
Drug treatments were performed as follows: CV-1 cells were pre-incubated
for 30 minutes at 37°C in R-medium containing 1 µM LatrunculinA
(Molecular Probes), 500 nM Jasplakinolide (Molecular Probes), 25 µg/ml
Nystatin (Sigma) plus 10 µg/ml Progesterone (Sigma), 1 µM Staurosporin
(Sigma), 100 µM Genistein (Sigma), 1 µM Okadaic acid (Sigma) or 1 mM
Sodium orthovanadate (Calbiochem). Drugs were present throughout the
experiments. Drug treatments did not result in a loss of cell viability and the
effects were reversible. Drug-treated CV-1 cells were analyzed with I.I.F. 20
hours after addition of virus for infection using monoclonal antibodies against
SV40 small and large T antigen (Pharmingen). 4 randomly chosen fields in 3
independent experiments were analyzed for the percentage of cells expressing
T antigen.
Single-cell SV40 uptake assay
CV-1 cells, co-transfected with DsRed-N1 and the appropriate cDNA (41)
were incubated with FLX-SV40 (Pelkmans et al., 2001) for 3.5 hours at 37°C,
washed and analyzed with live microscopy in a medium of pH 4.5. At this
pH, the emitted light from extracellular FLX-SV40 excited at 488 nm is
completely quenched (Pelkmans et al., 2001). Transfected cells were
visualized in the red channel (DsRed). Images in the green channel
(internalized FLX-SV40) were obtained with low exposure times to ensure a
linear range of intensities (12bit 1-4095), the signal was quantified using
Openlab 3.0.4 (Improvision) and expressed as percentage of the signal in nontransfected cells.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
95
Generation of AF-SV40
SV40 was purified and labeled as described (Pelkmans et al., 2001) with the
fluorophore Alexa Fluor 594 carboxylic acid succinimidyl ester (Molecular
Probes).
Light microscopy techniques
Cells were seeded the previous day on 18 mm Alcian blue-coated coverslips
and were either transiently transfected with the appropriate cDNA using
Superfect (Qiagen) or used directly. Transfected cells with relatively low
levels of expression were analyzed 12 hours after transfection. In cotransfection studies, cells were analyzed 24 hours after transfection, when,
according to control experiments, 95% of transfected cells expressed both
genes. In some experiments cells were fixed and processed for I.I.F. as
described previously (Pelkmans et al., 2001). A mixture of the monoclonal
antibodies 4G10 (Upstate biotechnology) and PY20 (Santa Cruz
Biotechnology) in combination with Alexa Fluor 594- or Alexa Fluor 350labeled anti-mouse secondary antibodies (Molecular Probes) was used for
immunostaining of phosphotyrosines. All microscopy experiments were
performed with a fully automated Zeiss Axiovert 100M microscope suitable
for recording of Alexa Fluor 350, CFP, GFP/Fluorescein, YFP and Alexa Fluor
594 signals. Images and movies were acquired with a cooled charge-coupleddevice (CCD) camera (Hamamatsu) and processed with Openlab 3.0.4
(Improvision) as described (Pelkmans et al., 2001). Images of each channel
were acquired separately with 500 msec exposure time. Time-lapse intervals
were 500 msec. Trajectories were obtained by marking the position of moving
virus particles in 50 consecutive images. The maximal range was acquired by
calculation of the absolute distance between each position on the trajectory
respective to the starting position. The maximum distance of 18-37 trajectories
was taken and represented as average±standard deviation.
Quantification of signal overlap
The amount of overlap between Alexa Fluor 594-labeled SV40 (AF-SV40)
signal and caveolin-1-GFP (cav1-GFP) signal was quantified using Openlab
3.0.4 (Improvision). Signals were obtained with low exposure to ensure a
linear range of intensities (12 bit, ranging from 1-4095) and of areas in the cell
96
Caveolae-mediated endocytosis of SV40
where structures were well resolved (periphery). The cav1-GFP signal was
subtracted from the AF-SV40 signal, resulting in non-overlapping AF-SV40
signal. This was subtracted from total AF-SV40 signal and the resulting
overlapping AF-SV40 signal expressed as mean percentage±standard
deviation of total AF-SV40 signal in at least 6 cells.
Chapter 3. Uptake is ligand-triggered via actin and dynamin
97
3.5. Movies
All movies were recorded with a Zeiss Axiovert 100M microscope, equipped
with a 150 W HBO lamp and computer-controlled shutters and filters for
separate or simultaneous recording of GFP and Alexa Fluor signals. A 100x
Plan Apochromat 1.4 N.A. (Zeiss) objective was used. Images were captured
with an ORCA-ER CCD camera (Hamamatsu) using 2x2 binning (camera
pixel size: 9.9 µm), and processed with Openlab 3.0.4. (Improvision).
Movies should be viewed with a Quicktime movie player (Apple Macintosh
or PC, downloadable at http://www.apple.com/quicktime/download) and
brightness of the monitor should be set to the maximal value. Movies are best
viewed with a gamma correction of 1.0.
The real time (relative to the first frame of the movie) is always indicated in
the lower left corner of the movie (hours:min:sec:msec).
Movie 1
Live fluorescence microscopy experiment recorded in CV-1 cells having
bound AF-SV40 at 4ºC and immediately transferred to the heated microscope
stage. Image size: 10x10 µm, Recording: 0.77 FPS, Playback: 20 FPS
Movie 2 and 2A
Live fluorescence microscopy experiment recorded in CV-1 cells having
bound AF-SV40 at 4ºC, and incubated for 10 min at 37ºC. Arrowheads
indicate dynamic virus particles that become trapped during the recording.
Image size: 10x10 µm, Recording: 0.43 FPS, Playback: 20 FPS
Movie 3
Live fluorescence microscopy experiment recorded in CV-1 cells having
bound AF-SV40 at 4ºC and incubated for 20 min at 37ºC. Image size: 10x10
µm, Recording: 0.43 FPS, Playback: 20 FPS
98
Caveolae-mediated endocytosis of SV40
Movie 4
Live fluorescence microscopy experiment recorded in CV-1 cells, pretreated
with 1 µM LatA, having bound AF-SV40 at 4ºC in the presence of LatA and
incubated for 20 min at 37ºC in the presence of LatA. Image size: 9x9 µm,
Recording: 0.44 FPS, Playback: 20 FPS
Movie 5
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing caveolin-1-GFP having bound AF-SV40 at 4ºC and incubated for
20 min at 37ºC. Arrowheads indicate trapped SV40 particles in caveolae.
Image size: 10x10 µm, Recording: 0.45 FPS, Playback: 20 FPS
Movie 5A
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing caveolin-1-GFP, pretreated with 1 µm LatA, having bound AFSV40 in the presence of LatA at 4ºC and incubated for 20 min at 37ºC in the
presence of LatA. Arrowheads indicate mobile, virus-loaded cav1-GFP spots.
Image size: 10x10 µm, Recording: 0.22 FPS, Playback: 20 FPS
Movie 6
Live fluorescence microscopy experiment recorded in CV-1 cells expressing
caveolin-1-GFP for 12 hours, incubated for 2 hours at 4ºC and 20 min at 37ºC.
Image size: 10x10 µm, Recording: 0.79 FPS, Playback: 20 FPS
Movie 6A
Live fluorescence microscopy experiment recorded in CV-1 cells expressing
caveolin-1-GFP, pretreated with 1 µm LatA, incubated at 4ºC and 20 min at
37ºC in the presence of LatA. Image size: 10x10 µm, Recording: 0.79 FPS,
Playback: 20 FPS
Movie 7
Live fluorescence microscopy experiment recorded in CV-1 cells expressing
GFP-β-actin for 12 hours, having bound SV40 at 4ºC and incubated for 20 min
at 37ºC., Image size: 50x50 µm, Recording: 0.07 FPS, Playback: 20 FPS
Chapter 3. Uptake is ligand-triggered via actin and dynamin
99
Movie 7A and 7B
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing GFP-β-actin for 12 hours, having bound AF-SV40 at 4ºC and
incubated for 20 min at 37ºC. Arrowheads indicate virus particles that induce
polymerization of actin tails., Image size: 10x10 µm, Recording: 0.07 FPS,
Playback: 10 FPS
Movie 8
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing GFP-β-actin for 12 hours, incubated for 2 hours at 4ºC and 20 min
at 37ºC. Image size: 50x50 µm, Recording: 0.61 FPS, Playback: 20 FPS
Movie 9
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing GFP-β-actin for 12 hours, pretreated with 1 µm LatA, having
bound AF-SV40 at 4ºC in the presence of LatA and incubated for 20 min at
37ºC in the presence of LatA. Arrowheads indicate dynamic virus particles
that have recruited actin patches. Image size: 10x10 µm, Recording: 0.09 FPS,
Playback: 20 FPS
Movie 10
Live fluorescence microscopy experiment recorded in CV-1 cells expressing
dynamin2-GFP for 12 hours. Image size: 20x20 µm, Recording: 0.54 FPS,
Playback: 100 FPS
Movie 11
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing dynamin2-GFP for 12 hours, having bound AF-SV40 for 2 hours at
4ºC and incubated for 20 minutes at 37ºC. Only one frame for the AF-SV40
signal was taken, which is merged with the subsequent dynamin2-GFP
timelapse series. Arrowheads indicate virus particles on which dyn2-GFP
spots appear and dissapear. Image size: 10x10 µm, Recording: 0.49 FPS,
Playback: 20 FPS
100
Caveolae-mediated endocytosis of SV40
Movie 12
Dual color live fluorescence microscopy experiment recorded in CV-1 cells
expressing dynamin2-GFP for 12 hours, pretreated with 100 µm Genistein,
having bound AF-SV40 for 2 hours at 4ºC in the presence of Genistein and
incubated for 20 minutes at 37ºC in the presence of Genistein. Only one frame
for the AF-SV40 signal was taken, which is merged with the subsequent
dynamin2-GFP timelapse series. Image size: 10x10 µm, Recording: 0.49 FPS,
Playback: 20 FPS
Acknowledgements
The authors thank all lab members for helpful discussions and suggestions
throughout the work. They also thank Sandra Schmid for suggesting the use
of TCEP and providing cDNA of Dyn2wt and Dyn2K44A, Marc McNiven for
providing cDNA of dyn2wt-GFP and Dyn2K44A-GFP and Alexandre
Benmerah and Alice Dautry-Varsat for providing cDNA of Eps15-DIII∆2 and
Eps15-E∆95/295. This work was supported by the Swiss National Science
Foundation and by ETHZ.
Chapter 4
Endocytosis via caveolae
‘Discussion’
Lucas Pelkmans and Ari Helenius
Institute of Biochemistry, Swiss Federal Institute of Technology (ETH) Zürich
Traffic, volume 3, May 2002, pages 311-320
102
Caveolae-mediated endocytosis of SV40
Abstract
Caveolae are flask-shaped invaginations present in the plasma membrane of
many cell types. They have long been implicated in endocytosis, transcytosis,
and cell signaling. Recent work has confirmed that caveolae are directly
involved in the internalization of membrane components (glycosphingolipids
and glycosylphosphatidylinositol (GPI)-anchored proteins), extracellular
ligands (folic acid, albumin, autocrine motility factor), bacterial toxins
(cholera toxin, tetanus toxin), and several non-enveloped viruses (Simian
Virus 40, Polyoma virus). Unlike clathrin-mediated endocytosis,
internalization through caveolae is a triggered event that involves complex
signaling. The mechanism of internalization and the subsequent intracellular
pathways that the internalized substances take are starting to emerge.
Chapter 4. Endocytosis via caveolae: Discussion
103
While clathrin-mediated endocytosis constitutes the main pathway for
internalization of extracellular ligands and plasma membrane components in
most cell types, it has been recognized for some time that alternative, parallel
uptake mechanisms also exit. These ‘clathrin-independent pathways’ have
been more difficult to study; hence, detailed information is still scarce (for
recent reviews see (Dautry-Varsat, 2000; Falnes and Sandvig, 2000; Nichols
and Lippincott-Schwartz, 2001)). In this review, we focus on one of these
alternatives: endocytosis via caveolae. Our own interest in this field was
evoked by the observation that certain non-enveloped viruses such as Simian
Virus 40 (SV40) use caveolae-mediated endocytosis to enter cells (Anderson et
al., 1996; Hummeler et al., 1970; Stang et al., 1997). However, similar uptake
processes are likely to be involved in many interesting endogenous processes
including cholesterol homeostasis, recycling of GPI-anchored proteins,
glycosphingolipid transport, and transcytosis of serum components
(Anderson, 1998; Fielding and Fielding, 1997; Ikonen and Parton, 2000;
Kurzchalia and Parton, 1999; Parton, 1996; Smart et al., 1999).
4.1. Caveolae
Caveolae were first identified in the 1950s by Palade (Palade, 1953) and
Yamada (Yamada, 1955) due to their characteristic morphology observed by
electron microscopy of thin sections. They typically appear as rounded
plasma membrane invaginations of 50-80 nm in diameter. Their composition,
appearance and function are cell-type dependent. In endothelial cells,
caveolae can be more constricted at the mouth, or they may contain a
diaphragm that might restrict diffusion (Stan et al., 1999; Stan et al., 1997). In
muscle cells, caveolae are often observed in the form of composite clusters or
linear rows of multiple flask-shaped units involved in the formation of T
tubules (Carozzi et al., 2000; Parton et al., 1997). In epithelial tissue culture
cells, caveolae are open to the extracellular medium, do not contain a
diaphragm, appear as single indentations or grape-like structures, and are on
average a bit smaller (Montesano et al., 1982; Rothberg et al., 1992). Recent
video microscopy and fluorescence recovery after photobleaching (FRAP)
analysis has shown that in these cells caveolae are stationary and held in place
in the plasma membrane by the cortical actin cytoskeleton underlying the
104
Caveolae-mediated endocytosis of SV40
plasma membrane (Pelkmans et al., 2001; Thomsen et al., 2002). As discussed
below, only upon specific signals, do they detach from the membrane as an
endocytic vesicle.
Although caveolae do not show an electron dense layer on their cytosolic
surface in thin section electron micrographs, they do have a protein ‘coat’
composed primarily of a protein called caveolin-1 (or caveolin–3 in muscle
cells) (Rothberg et al., 1992). Caveolins are integral membrane proteins of 21
kDa. They have an unusual topology in that both N- and C-terminal domains
are cytosolic connected by a hydrophobic sequence that is buried in the
membrane but does not span the bilayer. (Dupree et al., 1993; Monier et al.,
1995). Caveolin-1 is palmitoylated in the C-terminal segment (Dietzen et al.,
1995), they can be phosphorylated on tyrosine residues (Glenney, 1989), they
bind cholesterol (Murata et al., 1995), and they form dimers and higher
oligomers (Monier et al., 1995). On the cytosolic surface they can be visualized
by electron microcopy as part of parallel, shallow ridges in replicas obtained
after shadowing (Peters et al., 1985; Rothberg et al., 1992).
Caveolins are essential for the formation and stability of caveolae: in their
absence no caveolae are seen, and, when expressed in cells that lack caveolae,
they induce caveolar formation (Fra et al., 1995). In addition to the plasma
membrane, caveolins are present in the Trans Golgi Network (TGN) (Dupree
et al., 1993; Kurzchalia et al., 1992) and in a newly discovered organelle called
the ‘caveosome’ (Pelkmans et al., 2001), which will be discussed below. Upon
extraction or oxidation of plasma membrane cholesterol, caveolins re-localize
to intracellular structures that can be endosomes, the Golgi complex, or the
ER (Carozzi et al., 2000; Smart et al., 1994). When caveolins are overexpressed or retained in the ER, they can also localize to intracellular lipid
droplets (Fujimoto et al., 2001; Ostermeyer et al., 2001; Pol et al., 2001).
Immunofluorescence with conformation-specific anti-caveolin antibodies
suggests that the conformation of caveolin-1 on the plasma membrane and in
caveosomes is similar, but different from that in the Golgi complex (Dupree et
al., 1993; Pelkmans et al., 2001).
Chapter 4. Endocytosis via caveolae: Discussion
105
During endocytosis of caveolae, caveolin-1 moves along with the vesicles into
the cytosol with no visible remnant left in the plasma membrane (Aoki et al.,
1999; Kang et al., 2000; Parton et al., 1994; Pelkmans et al., 2001; Stang et al.,
1997). Whether caveolin has a direct role in the endocytic process is not clear.
On the one hand, it has been shown that N-terminally truncated or Nterminally GFP-tagged caveolin constructs serve as dominant negative
inhibitors of caveolar endocytosis during SV40 entry (Pelkmans et al., 2001;
Roy et al., 1999). On the other hand, recent data indicate that the presence of
caveolin-1 can actually slow down the endocytic process via caveolae (Le et
al., 2001; Minshall et al., 2000). This implies that the role of caveolin may be to
stabilize caveolae and thus to counteract an underlying raft-dependent
endocytic process. The finding that certain ligands internalize via a lipid raftdependent but clathrin-independent mechanism in cells that lack caveolae,
has lead to the postulation that lipid rafts can internalize independently of
caveolae (Lamaze et al., 2001). In this context, it is significant that caveolin
knockout mice survive surprisingly well although their cells do not have
detectable caveolae (Drab et al., 2001; Galbiati et al., 2001; Razani et al., 2001).
In addition to caveolins, caveolae are known to contain dynamin (Henley et
al., 1998; Oh et al., 1998), a GTPase involved in the formation of clathrincoated vesicles (Hinshaw, 2000). This molecule has been localized to the neck
of flask-shaped caveolar indentations (Henley et al., 1998; Oh et al., 1998), and
is therefore most likely involved in pinching off the caveolar vesicle in a way
similar to its role in coated vesicle fission (De Camilli et al., 1995; Sever et al.,
2000). Recent video imaging of the dynamics of GFP-tagged dynamin2 during
SV40 entry showed that it is only transiently recruited to virus-loaded
caveolae (Pelkmans et al., 2002). It appears as small spots on the site of
caveolae with a residence time of about 8 seconds. Thus, dynamin is not a
permanent component of caveolae but rather belongs to a group of factors
recruited in response to specific signals. Also, caveolae contain the molecular
machinery for vesicle docking and fusion (Schnitzer et al., 1995).
Caveolae are, moreover, rich in several receptor and non-receptor protein
tyrosine kinases and in GPI-anchored proteins (Lisanti et al., 1993; Sargiacomo
et al., 1993). Since these kinases are involved in signal transduction events,
106
Caveolae-mediated endocytosis of SV40
caveolae are thought to constitute especially active sites for signal
transmission
(Anderson, 1998; Schlegel and Lisanti, 2001; Simons and
Toomre, 2000). Furthermore, some of the kinases might be involved in
regulating the internalization of caveolae, as will be discussed below.
The lipid composition of caveolae corresponds to that of lipid rafts, i.e.
caveolae are rich in cholesterol and sphingolipids (extensively reviewed in
(Brown and London, 1998; Simons and Toomre, 2000). These are, in fact,
essential for the formation and stability of caveolae. If cholesterol is removed
from the plasma membrane, caveolae disappear (Rothberg et al., 1992).
Consistent with their high content of raft lipids, caveolae resist solubilization
by non-ionic detergents at 4ºC. One may thus define caveolae as caveolincontaining plasma membrane invaginations rich in raft lipids.
4.2. Caveolar entry of Simian Virus 40.
We will first concentrate on one of the most extensively studied ligands for
caveolar endocytosis, SV40, which uses this pathway for infectious entry into
the cell (Anderson et al., 1996; Atwood and Norkin, 1989; Breau et al., 1992;
Chen and Norkin, 1999; Clever et al., 1991; Hummeler et al., 1970; Kartenbeck
et al., 1989; Liddington et al., 1991; Maul et al., 1978; Pelkmans et al., 2001;
Pelkmans et al., 2002, ; Roy et al., 1999; Shimura et al., 1987; Stang et al., 1997;
Stehle et al., 1996; Upcroft, 1987; Yamada and Kasamatsu, 1993). The entry
process of this non-enveloped DNA virus, analyzed in several laboratories
including ours, is emerging as a useful paradigm in the field.
SV40 has several advantages as a model ligand. The particle itself is well
characterized in terms of composition and structure. The X-ray structure of
this non-enveloped DNA virus has been determined at a resolution of 3.1Å
and shows an icosahedral shell with 360 subunits of the major coat protein
VP1 (Liddington et al., 1991; Stehle et al., 1996). The diameter of the particle is
50 nm. Although replication of the virus is restricted to monkey cells, the
entry process and initial infection can be followed in most cell types.
Chapter 4. Endocytosis via caveolae: Discussion
107
Uptake of SV40 occurs specifically by the caveolar pathway. Even when cells
are incubated with amounts of virus particles exceeding by far receptorsaturation, less then 5% of internalized particles is found to pass through
clathrin-coated pits and vesicles (Hummeler et al., 1970; Kartenbeck et al.,
1989; Maul et al., 1978; Pelkmans et al., 2001). When clathrin-dependent
endocytosis was inhibited by expressing a dominant negative mutant of the
EGF receptor pathway substrate 15 (Eps15), SV40 endocytosis and infection
were not affected (Pelkmans et al., 2002). In contrast, when dominant negative
mutants of caveolin were expressed, internalization or infection was not
observed (Pelkmans et al., 2001; Roy et al., 1999).
The interaction of the incoming virus particles with cells has been studied by
morphological techniques (electron microscopy and light microscopy), by
biochemical techniques (quantitative endocytosis assays) and by virological
methods (infection assays). Most of the studies have been performed in tissue
culture cells from green African monkeys. A variety of inhibitors have been
tested, as well as expression of dominant negative mutant proteins. Extensive
use has recently been made of video microscopy in live cells with
fluorescently labeled virus particles and GFP-labeled cell proteins. This has
made it possible to follow single virus particles during their journey into the
cell. The stepwise process of SV40 entry is today known in some detail:
4.2.1. Sequestration and internalization (Fig. 1).
After binding to the plasma membrane via MHC class I antigens (Breau et al.,
1992), virus particles diffuse laterally along the membrane, until trapped in
caveolae (Pelkmans et al., 2001). Virus containing caveolae are somewhat
smaller in diameter than the virus-free ones (Stang et al., 1997). The reason
may be the tight interaction between the virus and the caveolar membrane; in
electron micrographs it almost looks like the viruses would be ‘budding’ into
the cell.
Virus particles stay in the caveolae for 20 min or more, after which the
caveolae pinch off and move as caveolin-coated endocytic vesicles into the
cytoplasm. Caveolae devoid of virus particles do not internalize (Pelkmans et
al., 2001). Interestingly, virus uptake does not involve endocytosis of MHC
108
Caveolae-mediated endocytosis of SV40
class I antigens (Anderson et al., 1998). This implies the presence of a
secondary receptor that mediates the tight binding of the virus to the caveolar
membrane, and possibly triggers the signal needed to induce the endocytic
process (see below).
Fig. 1. Initial stages of SV40 internalization via caveolae.
After binding to the membrane, virus particles are mobile until trapped in caveolae, which are linked to
the actin cytoskeleton (step 1). In the caveolae, SV40 particles trigger a signal transduction cascade that
leads to local protein tyrosine phosphorylation and depolymerization of the cortical actin cytoskeleton
(step 2). Actin monomers are recruited to the virus-loaded caveolae and an actin patch is formed (step
3). Concomitantly, Dynamin is recruited to the virus-loaded caveolae and a burst of actin
polymerization occurs on the actin patch (step 4). Virus-loaded caveolae vesicles are now released from
the membrane and can move into the cytosol (step 5). After internalization, the cortical actin
cytoskeleton returns to its normal pattern (step 6).
During the lag period preceding internalization, a complex series of virusinduced events take place. The arrival of the SV40 particle in the caveolae
triggers phosphorylation of tyrosine residues in proteins associated with the
Chapter 4. Endocytosis via caveolae: Discussion
109
caveolae (Pelkmans et al., 2002). However, the kinases and substrates
involved remain to be identified.
One effect of the activated signal transduction cascade is the disassembly of
nearby actin stress fibers. Subsequently, actin is recruited to the virus-loaded
caveolae as a small actin patch, followed by bursts of actin polymerization
resulting in the transient appearance of actin ‘tails’ (average length about 1.5
µm) emanating from the virus-loaded caveolae. Dynamin is also recruited to
the sites of virus internalization but, as already mentioned, it stays there only
for a short period of time (Pelkmans et al., 2002). Another effect is the
upregulation of the primary response genes c-myc, c-jun and c-sis (Dangoria
et al., 1996).
Studies with inhibitors and dominant negative mutants show that the
association of SV40 with caveolae, the tyrosine phosphorylation, the
recruitment of actin, the formation of actin tails, and the association of
dynamin are all necessary for efficient closure of caveolar vesicles and
ultimately for infection of the cell (Chen and Norkin, 1999; Pelkmans et al.,
2002). Most of the changes seem to be transient; once the virus particles are
internalized, the phosphotyrosines disappear, and the actin cytoskeleton
returns to a normal pattern.
4.2.2. Transport to caveosomes
The caveolar vesicles formed have a diameter of 60-70 nm and contain single
virus particles surrounded by a tight-fitting membrane (Hummeler et al.,
1970). The local depolymerization of cortical actin seems to be necessary for
their passage deeper into the cytosol (Pelkmans et al., 2002) and the actin tail
is not needed once the vesicle is released into the cytosol.
These primary endocytic vesicles transfer their viral cargo presumably by
membrane fusion to larger, more complex tubular membrane organelles that
we have termed ‘caveosomes’ (Pelkmans et al., 2001). Since the uptake of
virus particles is relatively slow (half time 90 min and maximal uptake after 3
hours), accumulation of virus particles in caveosomes becomes visible after
about 40-60 min and continues for the following 3 hours.
110
Caveolae-mediated endocytosis of SV40
Fig. 2. Subcellular distribution, morphology and constituents of caveosomes
a. (Left) Laser scanning confocal micrograph (intracellular plane) of a CV-1 cell expressing caveolin
–1-GFP (cav1-GFP, green) and Texas Red-labeled Simian Virus 40 (TRX-SV40, red). Virus particles
have been allowed to accumulate in caveosomes for 3 hours. Most caveosomes are now filled with
virus particles (yellow), but some are still empty (green). Scalebar, 10 µm. (Middle) Thin section
electron micrograph of an intracellular region of a CV-1 cell having internalized virus particles for 2
hours. Caveosomes appear as heterogeneous organelles containing tightly packed SV40 particles.
Vacuolar endosomes do not accumulate virus particles. (CCV: clathrin coated vesicle, CV: caveolar
vesicle). Scalebar, 500 nm. (Right) Higher magnification thin section electron micrograph shows the
heterogeneous morphology of caveosomes, with virus particles appearing as ‘peas in a pod’. Scalebar,
100 nm. b. CV-1 cells were transiently transfected with cav1-GFP (upper row), GPI-GL-GFP (lower
row) or not transfected. 16 hours after transfection, cells were analyzed live with bright field
microscopy. (Upper row) cav1-GFP-expressing CV-1 cells were incubated at 37ºC with 10 µM Filipin
(Sigma), which emits blue fluorescence upon binding to free cholesterol. Since it is membrane
permeable, intracellular structures are also stained. Note that the plasma membrane and intracellular
cav1-GFP labeled caveosomes are strongly stained with fluorescent filipin. (Middle row) Cells were
first incubated for 2.5 hours at 37ºC in the presence of 1 µg AF-SV40, followed by 30 min at 4ºC in the
presence of 5 µM bodipy-labeled Lactosyl Ceramide (LacCer) complexed to BSA. Cells were washed
and incubated for 30 min at 37ºC. AF-SV40 is mainly intracellular and is visible in caveosomes. After
30 min, some LacCer is internalized and colocalizes with AF-SV40 in caveosomes. (Lower row) GPIGL-GFP-expressing CV-1 cells were incubated for 3 hours at 37ºC with 1 µg AF-SV40 and
subsequently washed. Note that all intracellular AF-SV40 signal, present in caveosomes, colocalizes
with GPI-GL-GFP. Scalebar is 15 µm.
Chapter 4. Endocytosis via caveolae: Discussion
111
Caveosomes are pre-existing, caveolin-1 containing membrane organelles
distributed throughout the cytoplasm. The pH in caveosomes is neutral, and
the membrane is rich in cholesterol and glycosphingolipids (Pelkmans et al.,
2001; Puri et al., 2001), (L. Pelkmans, A. Helenius, in preparation).
Caveosomes do not accumulate ligands endocytosed via clathrin-coated pits,
or components such as transferrin that cycle between endosomes and the
plasma membrane. Nor do they accumulate detectable amounts of fluid phase
markers such as FITC-dextran or horseradish peroxidase even when these are
added to cells together with SV40 (Kartenbeck et al., 1989; Pelkmans et al.,
2001). Antibodies against markers of endosomes, lysosomes, TGN, Golgi
complex, or the ER do not stain the caveosome. Ultra-structural analysis
shows tubulovesicular structures, heterogeneous in size and shape, with the
virus particles usually present in narrow tubules like peas in a pod
(Kartenbeck et al., 1989; Pelkmans et al., 2001).
4.2.3. Molecular sorting and transport to the ER
During the accumulation of virus particles, the caveosomes become
increasingly dynamic. Longer tubular extensions filled with virus particles
but devoid of caveolin-1 emerge and detach from the caveosomes leaving
caveolin-1 behind (Pelkmans et al., 2001). These vesicles are transported along
microtubules to perinuclear membrane organelles identified as the smooth
ER. The receiving compartment can grow in size when more virus particles
are added, forming an anastomosing, tubular expansion of the ER
(Kartenbeck et al., 1989; Maul et al., 1978). Here, the majority of virus particles
remain undegraded for up to 16 hours or longer. How the viral genome is
transported from the ER to the nucleus is unclear, but it seems to passage
through the cytosol and the nuclear pore complex (Kasamatsu and Nakanishi,
1998).
In summary, the entry pathway of SV40 has revealed several key features that
seem to define the caveolar endocytic pathway and make it distinct from
clathrin-mediated endocytosis. First, SV40 uptake through caveolae is ligandtriggered. Triggering involves one or more tyrosine kinases that initiate a
phosphorylation-dependent signaling cascade. Second, the pathway taken by
the virus inside the cell bypasses the organelles involved in clathrin-coated
112
Caveolae-mediated endocytosis of SV40
vesicle endocytosis. Moreover, along the route, the pH is maintained at a
neutral level and no degradative end-station is reached. Instead, the cargo is
delivered to the ER. Thus, this pathway provides a direct route from the
plasma membrane to the ER.
Remarkable is also the role of the cytoskeleton. Actin plays a central role in
the formation of the caveolar vesicles, and in the initial transfer of the newly
formed vesicles through the cortex of the plasma membrane. Both the ability
to locally de-polymerize, but also to polymerize actin are important. This is in
contrast to the internalization of clathrin-coated pits, which appears to be
enhanced by, but is not strictly dependent on, a functional actin cytoskeleton
(Brodsky et al., 2001; Fujimoto et al., 2000).
After internalization, the microtubule cytoskeleton takes over. Initial transport
to caveosomes does not seem to be dependent on microtubules, but video
microscopy shows that early vesicles can move along them. Later,
microtubules play an essential role in the sorting of SV40 from caveosomes
and transport to the smooth ER.
4.3. Caveolar endocytosis of other ligands and membrane
constituents.
Other ligands or membrane constituents that can be internalized via caveolae
are cholera toxin (Montesano et al., 1982; Parton et al., 1994), folic acid
(Anderson et al., 1992; Rothberg et al., 1990), serum albumin (Schnitzer et al.,
1994), autocrine motility factor (AMF) (Benlimame et al., 1998), alkaline
phosphatase (Parton et al., 1994), GPI-anchored green fluorescent protein
(GPI-GFP) (Nichols et al., 2001), and the bodipy-labeled glycosphingolipid
Lactosyl Ceramide (LacCer) (Puri et al., 2001). (See Table 1 for more details).
Certain FimH expressing bacteria are also internalized in a caveolaedependent manner in mast cells (Shin et al., 2000).
SV40 is not the only virus that enters via caveolae. The closely related
polyoma virus was recently found to enter mouse cells by the caveolar route
and to have a similar dissociating effect on actin stress fibers (Richterova et
113
Chapter 4. Endocytosis via caveolae: Discussion
al., 2001). Interestingly, polyoma virus appears to specifically bind to
branched sialic acid groups present in glycoproteins with O-linked
carbohydrates and in GM1 gangliosides (Stehle et al., 1994). Echovirus 1, a
member of the picorna virus family, which binds to α2β1-integrin, also
internalizes via caveolae (Marjomaki et al., 2002). In addition, Respiratory
Syncytial virus has also been reported to associate with caveolae (Werling et
al., 1999). Finally, there is evidence that HIV-1 uses caveolae for transcytosis
across endothelia (Campbell et al., 2001). It is not unlikely that the caveolar
pathway is important for a variety of viruses for which the entry mechanism
has remained obscure.
Table 1: Materials endocytosed via caveolae or lipid rafts.
Ligand or
constituent
Receptor or
membrane
moiety
Induction Affinity
Caveolae/
Lipid rafts
References
Clathrin
Ligands
Folic acid
Albumin
Autocrine
Motility Factor
(AMF)
Interleukin-2
(IL2)
Folic acid
receptor (GPI
anchor)
gp60 (TM),
and others
AMF receptor
(TM)
IL2 receptor
(TM)
Yes
++
+
(Anderson et al., 1992;
Rijnboutt et al., 1996; Rothberg
et al., 1990)
(Minshall et al., 2000; Schnitzer
et al., 1994)
(Benlimame et al., 1998; Le et
al., 2000)
Yes
++
+
ND
++
+
ND
++
-
(Lamaze et al., 2001)
Membrane constuents
Alkaline
phosphatase
GPI-anchor
Yes
++
-
(Parton et al., 1994)
GPI-GFP
Lactosyl
Ceramide
GPI-anchor
Glycosphingo
-lipid
ND
Yes
++
++
-
(Nichols et al., 2001)
(Puri et al., 2001)
Cholera toxin
GM1
gangliosides
ND
++
+
(Montesano et al., 1982; Parton
et al., 1994; Shogomori and
Futerman, 2001; Torgersen et
al., 2001)
Tetanus toxin
GPI anchored
molecules
ND
++
-
(Herreros et al., 2001;
Montesano et al., 1982; Munro
et al., 2001)
Simian Virus 40
MHC1 (TM),
ND
Yes
++
-
Polyoma Virus
Sialic acid,
GM1
gangliosides
α2β1-Integrin
ND
ND
++
-
ND
ND
+
+
ND
ND
(Anderson et al., 1996; Breau et
al., 1992; Hummeler et al.,
1970; Stang et al., 1997)
(Gilbert and Benjamin, 2000;
Mackay and Consiligi, 1976;
Richterova et al., 2001)
(Marjomaki et al., 2002)
(Werling et al., 1999)
++
-
Toxins
Viruses
Echovirus 1
Respiratory
Syncytial virus
Bacteria (caveolae-mediated phagocytosis)
FimH expressing
E.Coli
CD48 (GPI
anchor)
Yes
(TM, transmembrane insertion; ND, not determined).
(Shin et al., 2000)
114
Caveolae-mediated endocytosis of SV40
4.4. How does it work?
The systems studied so far provide a rather heterogeneous picture of caveolar
endocytosis, and it is mot easy to define a common denominators. However,
caveolae seem to be used by cells to internalize membrane components that
are enriched in rafts, such as cholesterol, glycosphingolipids, GPI-anchored
proteins, and any ligands that bind to them. However, to be internalized,
affinity for rafts or raft components is, in general, not sufficient; some form of
induction is in addition is needed.
One way to induce caveolar uptake is by cross-linking caveolar components.
This has been shown for several GPI-anchored proteins (Mayor et al., 1994;
Parton et al., 1994). Many of the ligands in table 1 are multivalent and
therefore capable of cross-linking their receptors. Such clustering ability may
explain why multivalent ligands such as virus particles and bacteria are
sequestered into caveolae. They may also induce the formation of new
caveolae (Parton and Lindsay, 1999; Verkade et al., 2000).
Internalization of caveolae is in several cases induced upon activation of a
phosphorylation cascade (Anderson, 1998; Minshall et al., 2000; Pelkmans et
al., 2002). It is also known that phosphatase inhibitors can induce
internalization of caveolae (Parton et al., 1994; Thomsen et al., 2002). How
extracellular ligands activate phosphorylation is unclear. One way to transmit
a signal is by cross-linking transmembrane tyrosine kinase receptors, of which
several are enriched in caveolae. Another, more recently postulated
mechanism suggests that clustering of GPI-anchored proteins or
glycosphingolipids in the extracellular leaflet may stabilize lipid rafts, which
in turn may lead to the recruitment of proteins with high affinity for rafts,
such as caveolin-1 and Src family protein tyrosine kinases, on the cytosolic
leaflet (Simons and Toomre, 2000; Verkade et al., 2000).
Direct signaling via lipid raft-enriched lipids such as ceramide or
phosphatidyl inositols could also be involved. For instance, internalization of
folic acid via caveolae is regulated by the serine/threonine kinase Protein
Kinase Cα (PKCα) (Anderson, 1998), which is activated by diacylglycerol, a
Chapter 4. Endocytosis via caveolae: Discussion
115
hydrolysis product of phosphatidyl inositol 4, 5-bisphosphate (PIP2). It has
also been suggested that PKCα regulates the internalization or receptor
recycling of SV40, since the phorbol ester PMA, which mimics diacylglycerol,
reduces SV40 uptake (Anderson et al., 1996).
The actin cytoskeleton seems to play an essential role in the internalization of
caveolae. First of all, local recruitment of an actin patch to caveolae, similar to
that observed in clustered lipid rafts (Harder and Simons, 1999), is necessary
and might function as a scaffold for the buildup of the internalization
machinery (Pelkmans et al., 2002). Temporary depolymerization of the
cortical actin cytoskeleton is necessary to allow proper closure of caveolae and
transit to the cytosol (Pelkmans et al., 2002, ; Thomsen et al., 2002), (P.
Verkade and K. Simons, personal communication). The actin patches are
likely formed by actin-protein or actin-lipid interactions. At least during SV40
entry, the patches are formed in conditions where actin depolymerization is
temporarily favored. The mechanism of actin recruitment is not known, but it
may involve enrichment of PIP2 on the cytosolic leaflet of caveolae, which can
recruit the necessary machinery to locally polymerize actin (Caroni, 2001;
Martin, 2001). It likely also involves several SH2 and SH3 adapter proteins
that link the actin cytoskeleton to tyrosine posphorylation. Furthermore,
filamin, an actin-binding protein that binds caveolin-1, might play a role.
Dynamin is also involved in the internalization of caveolae. Over-expression
of a mutant dynamin, defective in GTP hydrolysis inhibits release of caveolae
from plasma membranes in an in vitro assay (Oh et al., 1998). EM images
showed that caveolae have extended necks in the presence of this mutant
(Henley et al., 1998; Oh et al., 1998). Functional data come from the
observation that this dynamin mutant inhibits uptake of cholera toxin and
muscarinic cholinergic receptors via caveolae and interleukin-2 (IL2) via lipid
rafts (Dessy et al., 2000; Henley et al., 1998; Lamaze et al., 2001; Oh et al.,
1998). It now appears that, as a result of the ligand-induced kinase activity,
dynamin is temporarily recruited to the internalization site to perform its
function in the fission of caveolar vesicles.
116
Caveolae-mediated endocytosis of SV40
Once internalized, caveolar vesicles seem to follow an intracellular route
distinct from the classical endocytic pathway (Fig. 3). They enter a caveolin-1rich, sorting compartment, the caveosome, which, as already discussed, is
distinct from endosomes. From the caveosomes, the internalized substances
are distributed to the ER, to the Golgi complex, and possibly to other
compartments. The pathway is best analyzed for SV40, GPI-GFP, and LacCer
(Nichols et al., 2001; Pelkmans et al., 2001; Puri et al., 2001). Available
information regarding internalization of cholera and tetanus toxins, albumin,
folic acid, and FimH expressing E. coli is consistent with this picture
(Anderson et al., 1992; Lalli and Schiavo, 2002; Nichols et al., 2001; Puri et al.,
2001; Schnitzer and Bravo, 1993; Shin et al., 2000). The issue is somewhat
complicated, however, by the observation that some of the ligands (cholera
toxin, albumin and the folic acid receptor) are also internalized via clathrincoated pits, and can therefore be found in endosomes as well (Rijnboutt et al.,
1996; Shogomori and Futerman, 2001; Torgersen et al., 2001).
4.5. Perspectives
Endocytosis through caveolae provides a true alternative to the clathrinmediated pathway. Being ligand-triggered, it provides an intrinsically more
selective way for uptake of specified substances, and it allows tighter control
by cell regulation. It can be used to route incoming ligands and membrane
components to organelles, such as the ER, that are not easily accessed by other
endocytic mechanisms. This advantage could, for example, be important in
the homeostasis of cholesterol for which most of the regulatory sensors reside
in the ER (Ikonen and Parton, 2000). A direct connection between the plasma
membrane and the ER could also be important in other processes such as the
immune response and in the regulation of secretion. Moreover, the pathway
bypasses low pH and digestive compartments, which may be important for
some of the internalized ligands.
The analysis of caveolar endocytosis is still in its early stages. Many
conceptual and mechanistic issues remain to be solved. It is important to
address the differences and similarities between caveolae and other non
clathrin-mediated uptake systems. Is caveolar uptake just a special form of
Chapter 4. Endocytosis via caveolae: Discussion
117
raft-mediated endocytosis? The mild phenotype of knockout mice suggests
that, while required for formation of caveolae, caveolin-1 may not be essential
for endocytosis nor for crucial signaling functions. The molecular mechanisms
of endocytosis and the connection with the cytoskeleton also need to be
addressed in detail. Here, complicated signaling and regulatory cascades are
involved, of which we now only see the tip of the iceberg.
Fig. 3. Proposed pathway for intracellular trafficking of ligands and constituents internalized via
caveolae or lipid rafts.
Depicted is a model for internalization via caveolae and lipid rafts. After internalization, caveolae- or
lipid raft-derived vesicles travel to caveosomes, which are distinct from endosomes in content and pH.
In caveosomes, internalized ligands or membrane constituents could reside, be sorted to the Golgi
complex, to the Endoplasmic reticulum (ER), or recycled back to the ER. Membrane constituents that
are sorted to the Golgi complex are GPI-GFP and Lactosyl Ceramide (LacCer). Ligands that are sorted
to the ER are Simian Virus 40 (SV40) and autocrine motility factor (AMF). Whether ligands or
constituents can cycle from caveosomes directly back to the plasma membrane has not yet been
studied. Caveolin-1 partly resides in caveosomes. Examples of ligands internalized via clathrin-coated
pits that travel to endosomes are also depicted. From endosomes, the envelope of Semliki Forest Virus
(SFV) and low-density lipoprotein (LDL) are sorted to lysosomes, Shiga toxin is sorted to the Golgi
complex and transferrin (Tfn) is recycled back to the plasma membrane. Whether ligands or membrane
constituents can travel between endosomes and caveosomes has not been studied.
118
Caveolae-mediated endocytosis of SV40
Acknowledgments
The authors would like to thank all members of the laboratory for helpful
suggestions and valuable comments. Work is supported by grants from the
Swiss National Science Foundation and the Swiss Federal Institute of
Technology.
Chapter 5
General conclusions
Lucas Pelkmans
Institute of Biochemistry, Swiss Federal Institute of Technology (ETH) Zürich
120
Caveolae-mediated endocytosis of SV40
The work presented in this thesis has confirmed that caveolae have endocytic
capacity under non-perturbing conditions (i.e. drug-treatments) and thus, a
long-standing debate has reached an end. Furthermore, the work established
that the itinerary after internalization is different from most other endocytic
processes. It involves a new organelle, called the caveosome. Caveosomes
establish a direct, long-hypothesized link between the plasma membrane and
the endoplasmic reticulum. Finally, the work has revealed that endocytosis of
caveolae is a ligand-triggered process that involves the induction of extensive
rearrangements of the actin cytoskeleton and the recruitment and action of
dynamin2. This directly explains why many attempts to study caveolaemediated endocytosis lacking a suitable ligand have given ambiguous results
in the past.
It is tempting to draw general conclusions from these observations, but it
must be stressed that the uptake of a non-endogenous, pathogenic particle
was analyzed. However, since SV40 is a non-enveloped virus built from
passive components (the capsid does not include enzymes such as proteases
or kinases), it must activate an endogenous mechanism. Unfortunately, it is
not yet known what the components of this SV40-induced signal are, or
whether physiological stimuli can accomplish the same. Preliminary evidence
not presented in this thesis (L. Pelkmans and A. Helenius, in preparation),
shows that the RhoGTPase RhoA and the phosphoinositide system are
involved in SV40 uptake. Together with the observations that, 1) the plasma
membrane tightly enwraps the virus particle, 2) the actin cytoskeleton
rearranges dramatically during SV40 internalization, and 3) PKC is possibly
involved (Anderson et al., 1996), there is clear resemblance with events
happening during phagocytosis.
Phagocytosis involves classical transmembrane receptor signaling via the
activation of Src PTKs. Since preliminary evidence shows that Src PTKs are
not involved in SV40 uptake (L. Pelkmans and A. Helenius, in preparation),
the SV40-induced signal is likely transduced by a different mechanism.
Recent observations describe that clustering from the outside of proteins or
certain glycosphingolipids enriched in lipid rafts leads to the accumulation of
actin on the cytosolic side of the plasma membrane. Interestingly, the SV40-
Chapter 4. Endocytosis via caveolae: Discussion
121
related polyoma virus is known to bind to sialic acid residues present on
certain gangliosides, which are enriched in lipid rafts. Perhaps SV40 uses a
similar mechanism by cross-linking lipid-raft components.
The clustering of MHC class I molecules could also provide a mechanism of
signal transduction. Although speculative, the SV40-induced signal might be
related to the stimulation of T cells. There, ligation of MHC class I molecules
results in tyrosine phosphorylation of PLC, a PI3-K and ZAP70, the T cell
variant of Syk, eventually leading to the upregulation of the early response
gene c-jun. These processes should now be compared to SV40 entry.
Interestingly, upregulation of c-jun is already known to take place during
SV40 infection (Dangoria et al., 1996).
How actin associates with the cytosolic leaflet of lipid rafts is not known, but
it might involve the recruitment of acylated or palmitoylated actin-binding
proteins that are specifically enriched on the cytosolic leaflet of clustered lipid
rafts. Furthermore, the overexpression of a PI(4)P-5K, and thus increased
levels of PI(4,5)P2 result in actin tail formation preferentially on lipid-raftenriched vesicles (Rozelle et al., 2000). Since preliminary evidence not
discussed in this thesis (L. Pelkmans and A. Helenius, in preparation) shows
the involvement of the phosphoinositide system in the uptake of SV40, it may
well be that SV40 induces local production of PI(4,5)P2. This can then, similar
to the mechanism of phagocytosis, result in the rapid recruitment of actininteracting and –nucleating factors. The actin layer itself may even provide a
scaffold on which the internalization machinery is assembled.
Thus, with the possible overlap with the phagocytosis machinery in mind, the
next quest should be focused on characterizing the components leading to the
activation of caveolae-mediated endocytosis, most importantly on the
transmission of the extracellular signal to the intracellular machinery.
Hopefully, this will shed light on the physiological role of caveolae-mediated
endocytosis, the most important question that arises from this work.
Another important conclusion that can be drawn is that apparently several
different parallel endocytic systems exist in animal cells. Caveosomes are not
122
Caveolae-mediated endocytosis of SV40
directly linked to the classical endocytic pathway, since they have a neutral
pH, and, surprisingly, they do not contain fluid phase markers, as would be
expected of endocytic structures. This might be explained by the possibility
that SV40 particles push away the fluid phase as observed in electron
micrographs where the membrane is seen tightly surrounding the virus
particle (Kartenbeck et al., 1989). To analyze this more carefully, smaller fluid
phase tracers should be used, preferentially in combination with a treatment
that mimics the SV40-induced signal. Interestingly, a recent study indeed
suggests that caveosomes can accumulate fluid phase in the absence of SV40
(Nichols, 2002).
Because caveosomes contain several components of plasma membrane lipid
rafts (i.e. GPI-GL-GFP, LacCer, cholesterol) it can be speculated that
caveosomes are endocytic organelles for lipid raft ligands. Indeed, this now
appears to be the case, and thus another compartment with endogenous
markers can be added to the list of intracellular organelles. Since some of
these molecules are internalized with much faster rates than SV40, it appears
as if caveosomes can receive material from different uptake routes. These
routes might derive from separate internalization mechanisms (Nichols,
2002). Although internalizing pits have not been visualized in these other
pathways, they appear to be dependent on dynamin and on lipid rafts, other
than caveolae. Preliminary evidence not described in this thesis suggests that
SV40 can be internalized in a clathrin-independent manner in cells lacking
caveolins. This might mean that, at least from an endocytosis point of view,
caveolae and other lipid rafts can perform the same tasks. In the case of the
endocytic capacity of caveolae, caveolin-1 might just be a marker protein, and
even slow down the endocytic process by stabilizing caveolae. It will now be
of interest to see whether a similar internalization machinery is in place in
caveolin-deficient cells, whether they contain caveosomes (or raftosomes?)
and whether these still support transport to the ER. Detailed analysis of the
route taken by SV40 in these cells will likely shed more light on this problem.
A related question, whether SV40 can induce ‘new’ caveolae on membranes
that normally do not show caveolae, could be studied on the apical membrane
of polarized cells, which express caveolins, but only have caveolae on their
basolateral side.
Chapter 4. Endocytosis via caveolae: Discussion
123
Caveosomes can perform molecular sorting, resulting in the separation of
SV40 from caveolin-1. Interestingly, SV40 might activate this sorting
machinery and evoke the observed gradual change in caveosome dynamics
and appearance, i.e. the formation of tubular protrusions. An alternative
explanation is that these tubules might have been there, but simply not visible
before a saturating amount of SV40 had accumulated in the caveosomes,
which takes quite some time. Furthermore, preliminary video microscopy
experiments not described in this thesis show that cholera toxin and GPI-GLGFP are also sorted in caveosomes, namely away from caveolin-1 and SV40,
on their way to the Golgi complex (L. Pelkmans and A. Helenius, in
preparation). Thus, it appears that caveosomes are capable of targeting
different organelles. It seems likely that some of the sorting machinery for
SV40 is dependent on specific motor proteins, since the SV40-filled tubules
form along microtubules and depolymerization of microtubules interferes
with sorting. Since SV40 is sorted from caveolin-1, it could moreover be that
the sorting mechanism depends on the partitioning of SV40-receptors from a
lipid-raft into a non lipid-raft domain, as is the case for sorting of secretory
proteins in the TGN to the apical membrane of polarized cells. Crucial for a
better understanding of this new vesicular pathway will now be the isolation
of its components, most importantly the caveosomes. Initial results show that
accumulation of SV40 particles makes the caveosomes denser, and provides a
way to separate caveosomes from most other cellular material. Hopefully,
these ongoing efforts will lead to a molecular characterization of the
constituents and, using in vitro assays, will allow a detailed characterization of
the sorting machinery.
Another important feature of caveosomes is that they provide a link between
the plasma membrane and the endoplasmic reticulum. A recent study
suggests that transport from caveosomes to the ER is mediated by the COPIcoatomer complex (Richards et al., 2002). This fits with the view that COPI
coats in general mediate transport back to the ER. When caveosomes are
regarded as intermediate organelles between the plasma membrane and the
ER, fusion capacity between these membranes should exist. Interestingly, the
unexpected finding of a specific pairing of a plasma membrane-localized
SNARE with an ER-localized SNARE that assemble a fusion-competent
124
Caveolae-mediated endocytosis of SV40
machinery in vitro (J. Rothman, personal communication) might have its
function in this pathway. Furthermore, very recently, it was described that
during phagocytosis, the ER can specifically fuse with the internalizing
phagocytic cup (Gagnon et al., 2002). Again this suggests that caveolaemediated endocytosis to the ER might have more in common with
phagocytosis than previously thought.
The ER subcompartment in which SV40 accumulates has a peculiar
appearance. It is connected to cisternae of the rough ER but contains no
ribosomes and consists of many anastomosing smooth tubules (Kartenbeck et
al., 1989). Such a smooth ER is normally not present in epithelial cells.
Interestingly, when a large amount of virus is added, the compartment
expands and preliminary evidence not described in this thesis shows that the
expression of the smooth ER marker syntaxin 17 is upregulated. Although it is
not known what the function of this compartment is, it could be related to the
smooth ER in hepatocytes which is involved in the detoxification of certain
drugs and in sterol synthesis. Maybe, SV40 is hijacking the cell’s mechanism
to control its cholesterol content, which supposedly needs a connection
between the plasma membrane, where the majority of cholesterol is located,
and the smooth ER, where cholesterol synthesis is regulated.
Finally, accumulating evidence shows that several viruses utilize caveolae to
enter their host cell. It will now be of interest to see to what extent the
infectious pathway of these viruses overlaps with that of SV40.
Today, many scientists regard clathrin-independent endocytosis as one of the
most exciting areas of basic cell biological research (Pfeffer, 2001). Although it
was recognized half a century ago (Palade, 1953), before clathrin-mediated
endocytosis was even discovered, significant progress has been made only in
the past decade. Striking in many cases is the connection with the biosynthetic
pathway. The fundaments for future research have now been laid, but we are
far from understanding it . Likely, SV40, as one of the founding ligands for
this uptake route, will prove a valuable tool to dissect the function, the
molecular machinery and the regulation of this clathrin-independent uptake
pathway for years to come.
125
Abbreviations
Abbreviations
FRAP
fluorescence recovery after
AAK1
adaptor associated kinase 1
GAP
GTPase activating protein
ADP
adenosine diphosphate
GDP
guanosine diphosphate
AMF
autocrine motility factor
GEF
GDP/GTP exchange factor
ATP
adenosine triphosphate
GFP
green fluorescent protein
AF
Alexa Fluor
GPCR
G-protein coupled receptor
AP
adaptor-protein complex
GPI
glycosyl phosphatidyl inositol
AP180
adaptor protein 180
GTP
guanosine triphosphate
Arf
ADP ribosylation factor
HIV
human immunodeficiency
Arp2/3
actin related protein 2/3
ATP
adenosine triphosphate
IgG
immunoglobulin G
BSA
bovine serum albumin
IP3
inositol triphosphate
Cav1
caveolin-1
ITAM
immunoreceptor tyrosine-
CCD
charge coupled device
CCP
clathrin coated pit
Jas
jasplakinolide
CCV
clathrin coated vesicle
LacCer
lactosyl ceramide
CFP
cyan fluorescent protein
LatA
latrunculin A
COP
coatomer protein
LDL
low density lipoprotein
CRIB
Cdc42/Rac interactive binding
LE
late endosome
DAG
diacylglycerol
ManII
Mannosidase II
DMEM
Dulbecco’s modified Eagle’s
MARCKS myristoylated alanine-rich C
photobleaching
virus
based activation motif
medium
kinase substrate protein
DNA
deoxyribonucleic acid
ds
double stranded
Dyn2
dynamin2
MOI
multiplicity of infection
EDTA
ethylenediaminetetraacetic acid
NMR
nuclear magnetic resonance
EE
early endosome
MPR
mannose-6-phosphate receptor
EEA1
early endosome antigen1
MVB
multivesicular body
EGF
epidermal growth factor
NSF
N-ethyl maleimide sensitive
EH
Eps15 homology
ENTH
Epsin N-terminal homology
Eps15
EGF receptor pathway
MHC
major histocompatibility
complex
factor
N-WASP neuronal Wiskott Aldrich
syndrome protein
substrate15
OA
okadaic acid
ER
endoplasmic reticulum
PA
phosphatidic acid
ERGIC
ER Golgi intermediate
PBS
phosphate buffered saline
compartment
PCR
polymerase chain reaction
ERM
ezrin/radixin/moesin
PDI
protein disulfide isomerase
FAK
focal adhesion kinase
PFU
plaque forming unit
FCS
fetal calf serum
PH
Pleckstrin homology
FITC
fluorescein isothiocyanat
PI
phosphatidylinositol
FLX
fluorescein X
PI 3-K
phosphatidylinositol-3 kinase
126
Caveolae-mediated endocytosis of SV40
PI-4K
phosphatidylinositol-4 kinase
PI(4)P-5K
phosphatidylinositol-4phosphate-5-kinase
PI(3)P
phosphatidylinositol-3phosphate
PI(4,5)P2
phosphatidylinositol-4,5bisphosphate
PKC
protein kinase C
PLC
phospholipase C
PLD
phospholipase D
PRD
proline rich domain
PTK
protein tyrosine kinase
RE
recycling endosome
SDS-PAGE
sodium dodecyl sulfatepolyacrylamide gel
electrophoresis
SFV
Semliki Forest Virus
SH2
Src homology 2
SH3
Src homology 3
SNARE
soluble NSF attachment
protein receptor
STP
staurosporin
SV40
Simian Virus 40
Syn17
syntaxin 17
TCA
trichloroacetic acid
TCEP
Tris(2carboxyethyl)phosphine
Tfn
transferrin
TGN
trans Golgi network
TRX
Texas red X
UIM
ubiquitin interacting motif
VIP21
vesicular integral membrane
protein of 21 kDa
VP
viral protein
WASP
Wiskott Aldrich syndrome
protein
YFP
yellow fluorescent protein
Bibliography
127
Bibliography
•
Adamczewski, M., Numerof, R. P., Koretzky, G. A., and Kinet, J. P. (1995).
Regulation by CD45 of the tyrosine phosphorylation of high affinity IgE
receptor beta- and gamma-chains, J Immunol 154, 3047-55.
•
Aderem, A., and Underhill, D. M. (1999). Mechanisms of phagocytosis in
macrophages, Annu Rev Immunol 17, 593-623.
•
Allen, L. A., and Aderem, A. (1996). Molecular definition of distinct
cytoskeletal structures involved in complement- and Fc receptor-mediated
phagocytosis in macrophages, J Exp Med 184, 627-37.
•
Amer, A. O., and Swanson, M. S. (2002). A phagosome of one's own: a
microbial guide to life in the macrophage, Curr Opin Microbiol 5, 56-61.
•
Anderson, H. A., Chen, Y., and Norkin, L. C. (1996). Bound simian virus 40
translocates to caveolin-enriched membrane domains, and its entry is
inhibited by drugs that selectively disrupt caveolae, Mol Biol Cell 7, 182534.
•
Anderson, H. A., Chen, Y., and Norkin, L. C. (1998). MHC class I
molecules are enriched in caveolae but do not enter with simian virus 40, J
Gen Virol 79, 1469-77.
•
Anderson, R. G., Kamen, B. A., Rothberg, K. G., and Lacey, S. W. (1992).
Potocytosis: sequestration and transport of small molecules by caveolae,
Science 255, 410-1.
•
Anderson, R. G. W. (1998). The caveolae membrane system, Annu Rev
Biochem 67, 199-225.
•
Aoki, T., Nomura, R., and Fujimoto, T. (1999). Tyrosine phosphorylation of
caveolin-1 in the endothelium, Exp Cell Res 253, 629-36.
•
Araki, N., Johnson, M. T., and Swanson, J. A. (1996). A role for
phosphoinositide 3-kinase in the completion of macropinocytosis and
phagocytosis by macrophages, J Cell Biol 135, 1249-60.
•
Atwood, W. J., and Norkin, L. C. (1989). Class I major histocompatibility
proteins as cell surface receptors for simian virus 40, J Virol 63, 4474-7.
•
Bagrodia, S., and Cerione, R. A. (1999). Pak to the future, Trends Cell Biol
9, 350-5.
•
Barouch, D. H., and Harrison, S. C. (1996). The interaction between the
major and minor coat proteins of polyomavirus, J Virol 68, 3982-3989.
128
•
Caveolae-mediated endocytosis of SV40
Benlimame, N., Le, P. U., and Nabi, I. R. (1998). Localization of autocrine
motility factor receptor to caveolae and clathrin-independent
internalization of its ligand to smooth endoplasmic reticulum, Mol Biol
Cell 9, 1773-86.
•
Benmerah, A., Bayrou, M., Cerf-Bensussan, N., and Dautry-Varsat, A.
(1999). Inhibition of clathrin-coated pit assembly by an Eps15 mutant, J
Cell Sci 112, 1303-11.
•
Bishop, A. L., and Hall, A. (2000). Rho GTPases and their effector proteins,
Biochem J 348 Pt 2, 241-55.
•
Botelho, R. J., Teruel, M., Dierckman, R., Anderson, R., Wells, A., York, J.
D., Meyer, T., and Grinstein, S. (2000). Localized biphasic changes in
phosphatidylinositol-4,5-bisphosphate at sites of phagocytosis, J Cell Biol
151, 1353-68.
•
Breau, W. C., Atwood, W. J., and Norkin, L. C. (1992). Class I major
histocompatibility proteins are an essential component of the simian virus
40 receptor, J Virol 66, 2037-45.
•
Brodsky, F. M., Chen, C. Y., Knuehl, C., Towler, M. C., and Wakeham, D.
E. (2001). BIOLOGICAL BASKET WEAVING: Formation and Function of
Clathrin-Coated Vesicles, Annu Rev Cell Dev Biol 17, 517-68.
•
Brown, D. A., and London, E. (1998). Functions of lipid rafts in biological
membranes, Annu Rev Cell Dev Biol 14, 111-36.
•
Brown, M. S., and Goldstein, J. L. (1999). A proteolytic pathway that
controls the cholesterol content of membranes, cells, and blood, Proc Natl
Acad Sci U S A 96, 11041-8.
•
Bubb, M. R., Senderowicz, A. M., Sausville, E. A., Duncan, K. L., and Korn,
E. D. (1994). Jasplakinolide, a cytotoxic natural product, induces actin
polymerization and competitively inhibits the binding of phalloidin to Factin, J Biol Chem 269, 14869-71.
•
Campbell, S. M., Crowe, S. M., and Mak, J. (2001). Lipid rafts and HIV-1:
from viral entry to assembly of progeny virions, J Clin Virol 22, 217-27.
•
Cao, H., Garcia, F., and McNiven, M. A. (1998). Differential distribution of
dynamin isoforms in mammalian cells, Mol Biol Cell 9, 2595-609.
•
Cardelli, J. (2001). Phagocytosis and macropinocytosis in Dictyostelium:
phosphoinositide- based processes, biochemically distinct, Traffic 2, 31120.
Bibliography
•
129
Caron, E., and Hall, A. (1998). Identification of two distinct mechanisms of
phagocytosis controlled by different Rho GTPases, Science 282, 1717-21.
•
Caroni, P. (2001). New EMBO members' review: actin cytoskeleton
regulation through modulation of PI(4,5)P(2) rafts, Embo J 20, 4332-6.
•
Carozzi, A. J., Ikonen, E., Lindsay, M. R., and Parton, R. G. (2000). Role of
Cholesterol in Developing T-Tubules: Analogous Mechanisms for TTubule and Caveolae Biogenesis, Traffic 1, 326-341.
•
Chacko, G. W., Brandt, J. T., Coggeshall, K. M., and Anderson, C. L. (1996).
Phosphoinositide 3-kinase and p72syk noncovalently associate with the
low affinity Fc gamma receptor on human platelets through an
immunoreceptor tyrosine-based activation motif. Reconstitution with
synthetic phosphopeptides, J Biol Chem 271, 10775-81.
•
Chapman, R. E., and Munro, S. (1994). Retrieval of TGN proteins from the
cell surface requires endosomal acidification, Embo J 13, 2305-12.
•
Chen, H., Fre, S., Slepnev, V. I., Capua, M. R., Takei, K., Butler, M. H., Di
Fiore, P. P., and De Camilli, P. (1998). Epsin is an EH-domain-binding
protein implicated in clathrin-mediated endocytosis, Nature 394, 793-7.
•
Chen, Y., and Norkin, L. C. (1999). Extracellular simian virus 40 transmits
a signal that promotes virus enclosure within caveolae, Experimental Cell
Research 246, 83-90.
•
Christiansen, G., Landers, T., Griffith, J., and Berg, P. (1977).
Characterization of components released by alkali disruption of simian
virus 40, J Virol 21, 1079-84.
•
Christoforidis, S., McBride, H. M., Burgoyne, R. D., and Zerial, M. (1999a).
The Rab5 effector EEA1 is a core component of endosome docking, Nature
397, 621-5.
•
Christoforidis, S., Miaczynska, M., Ashman, K., Wilm, M., Zhao, L., Yip, S.
C., Waterfield, M. D., Backer, J. M., and Zerial, M. (1999b).
Phosphatidylinositol-3-OH kinases are Rab5 effectors, Nat Cell Biol 1, 24952.
•
Clever, J., Yamada, M., and Kasamatsu, H. (1991). Import of simian virus
40 virions through nuclear pore complexes, Proc Natl Acad Sci U S A 88,
7333-7.
•
Cohen, P., Holmes, C. F., and Tsukitani, Y. (1990). Okadaic acid: a new
probe for the study of cellular regulation, Trends Biochem Sci 15, 98-102.
130
•
Caveolae-mediated endocytosis of SV40
Cole, C. N., Landers, T., Goff, S. P., Manteuil-Brutlag, S., and Berg, P.
(1977). Physical and genetic characterization of deletion mutants of simian
virus 40 constructed in vitro, J Virol 24, 277-294.
•
Conner, S. D., and Schmid, S. L. (2002). Identification of an adaptorassociated kinase, AAK1, as a regulator of clathrin-mediated endocytosis, J
Cell Biol 156, 921-9.
•
Coue, M., Brenner, S. L., Spector, I., and Korn, E. D. (1987). Inhibition of
actin polymerization by latrunculin A, FEBS Lett 213, 316-8.
•
Cox, D., Tseng, C. C., Bjekic, G., and Greenberg, S. (1999). A requirement
for phosphatidylinositol 3-kinase in pseudopod extension, J Biol Chem
274, 1240-7.
•
Cremona, O., Di Paolo, G., Wenk, M. R., Luthi, A., Kim, W. T., Takei, K.,
Daniell, L., Nemoto, Y., Shears, S. B., Flavell, R. A., et al. (1999). Essential
role of phosphoinositide metabolism in synaptic vesicle recycling, Cell 99,
179-88.
•
Damke, H., Baba, T., van der Bliek, A. M., Schmid, S. L. (1995). Clathrinindependent endocytosis is induced in cells overexpressing a temperaturesenstive mutant of dynamin, J Cell Biol 131, 69-80
•
D'Souza-Schorey, C., Boshans, R. L., McDonough, M., Stahl, P. D., and Van
Aelst, L. (1997). A role for POR1, a Rac1-interacting protein, in ARF6mediated cytoskeletal rearrangements, Embo J 16, 5445-54.
•
Dangoria, N. S., Breau, W. C., Anderson, H. A., Cishek, D. M., and Norkin,
L. C. (1996). Extracellular simian virus 40 induces an ERK/MAP kinaseindependent signalling pathway that activates primary response genes
and promotes virus entry, J Gen Virol 77, 2173-82.
•
Darby, C., Geahlen, R. L., and Schreiber, A. D. (1994). Stimulation of
macrophage Fc gamma RIIIA activates the receptor- associated protein
tyrosine kinase Syk and induces phosphorylation of multiple proteins
including p95Vav and p62/GAP-associated protein, J Immunol 152, 542937.
•
Dautry-Varsat, A. (2000). Clathrin-independent endocytosis. In
Endocytosis, M. Marsh, ed. (Oxford, Oxford University Press), pp. 26-57.
•
de Beer, T., Carter, R. E., Lobel-Rice, K. E., Sorkin, A., and Overduin, M.
(1998). Structure and Asn-Pro-Phe binding pocket of the Eps15 homology
domain, Science 281, 1357-60.
Bibliography
•
131
De Camilli, P., Takei, K., and McPherson, P. S. (1995). The function of
dynamin in endocytosis, Curr Opin Neurobiol 5, 559-65.
•
De Renzis, S., Sonnichsen, B., and Zerial, M. (2002). Divalent Rab effectors
regulate the sub-compartmental organization and sorting of early
endosomes, Nat Cell Biol 4, 124-33.
•
Defacque, H., Egeberg, M., Habermann, A., Diakonova, M., Roy, C.,
Mangeat, P., Voelter, W., Marriott, G., Pfannstiel, J., Faulstich, H., and
Griffiths, G. (2000). Involvement of ezrin/moesin in de novo actin
assembly on phagosomal membranes, Embo J 19, 199-212.
•
Desjardins, M., Celis, J. E., van Meer, G., Dieplinger, H., Jahraus, A.,
Griffiths, G., and Huber, L. A. (1994). Molecular characterization of
phagosomes, J Biol Chem 269, 32194-200.
•
Dessy, C., Kelly, R. A., Balligand, J. L., and Feron, O. (2000). Dynamin
mediates caveolar sequestration of muscarinic cholinergic receptors and
alteration in NO signaling, Embo J 19, 4272-80.
•
Dietzen, D. J., Hastings, W. R., and Lublin, D. M. (1995). Caveolin is
palmitoylated on multiple cysteine residues. Palmitoylation is not
necessary for localization of caveolin to caveolae, J Biol Chem 270, 6838-42.
•
Donaldson, J. G., and Jackson, C. L. (2000). Regulators and effectors of the
ARF GTPases, Curr Opin Cell Biol 12, 475-82.
•
Drab, M., Verkade, P., Elger, M., Kasper, M., Lohn, M., Lauterbach, B.,
Menne, J., Lindschau, C., Mende, F., Luft, F. C., et al. (2001). Loss of
caveolae, vascular dysfunction, and pulmonary defects in caveolin-1 genedisrupted mice, Science 293, 2449-52.
•
Dramsi, S., and Cossart, P. (1998). Intracellular pathogens and the actin
cytoskeleton, Annu Rev Cell Dev Biol 14, 137-66.
•
Dupree, P., Parton, R. G., Raposo, G., Kurzchalia, T. V., and Simons, K.
(1993). Caveolae and sorting in the trans-Golgi network of epithelial cells,
Embo J 12, 1597-605.
•
Engqvist-Goldstein, A. E., Kessels, M. M., Chopra, V. S., Hayden, M. R.,
and Drubin, D. G. (1999). An actin-binding protein of the Sla2/Huntingtin
interacting protein 1 family is a novel component of clathrin-coated pits
and vesicles, J Cell Biol 147, 1503-18.
•
Falnes, P. O., and Sandvig, K. (2000). Penetration of protein toxins into
cells, Curr Opin Cell Biol 12, 407-13.
132
•
Caveolae-mediated endocytosis of SV40
Ferguson, S. S. (2001). Evolving concepts in G protein-coupled receptor
endocytosis: the role in receptor desensitization and signaling, Pharmacol
Rev 53, 1-24.
•
Fielding, C. J., Bist, A., and Fielding, P. E. (1997). Caveolin mRNA levels
are up-regulated by free cholesterol and down- regulated by oxysterols in
fibroblast monolayers, Proc Natl Acad Sci U S A 94, 3753-8.
•
Fielding, C. J., and Fielding, P. E. (1997). Intracellular cholesterol transport,
J Lipid Res 38, 1503-21.
•
Fish, K. N., Schmid, S. L., and Damke, H. (2000). Evidence that dynamin-2
functions as a signal-transducing GTPase, J Cell Biol 150, 145-54.
•
Fra, A. M., Williamson, E., Simons, K., and Parton, R. G. (1995). De novo
formation of caveolae in lymphocytes by expression of VIP21-caveolin,
Proc Natl Acad Sci U S A 92, 8655-9.
•
Frischknecht, F., Cudmore, S., Moreau, V., Reckmann, I., Rottger, S., and
Way, M. (1999a). Tyrosine phosphorylation is required for actin-based
motility of vaccinia but not Listeria or Shigella, Curr Biol 9, 89-92.
•
Frischknecht, F., Moreau, V., Rottger, S., Gonfloni, S., Reckmann, I.,
Superti-Furga, G., and Way, M. (1999b). Actin-based motility of vaccinia
virus mimics receptor tyrosine kinase signalling, Nature 401, 926-9.
•
Frischknecht, F., and Way, M. (2001). Surfing pathogens and the lessons
learned for actin polymerization, Trends Cell Biol 11, 30-38.
•
Fujimoto, L. M., Roth, R., Heuser, J. E., and Schmid, S. L. (2000). Actin
assembly plays a variable, but not obligatory role in receptor- mediated
endocytosis in mammalian cells, Traffic 1, 161-71.
•
Fujimoto, T., Kogo, H., Ishiguro, K., Tauchi, K., and Nomura, R. (2001).
Caveolin-2 is targeted to lipid droplets, a new "membrane domain" in the
cell, J Cell Biol 152, 1079-85.
•
Gagnon, E., Duclos, S., Rondeau, C., Chevet, E., Cameron, P. H., SteeleMortimer, O., Paiement, J., Bergeron, J. J., and Desjardins, M. (2002).
Endoplasmic Reticulum-mediated phagocytosis is a mechanism of entry
into macrophages, Cell 110, 119-131.
•
Gaidarov, I., Santini, F., Warren, R. A., and Keen, J. H. (1999). Spatial
control of coated-pit dynamics in living cells, Nat Cell Biol 1, 1-7.
•
Galbiati, F., Engelman, J. A., Volonte, D., Zhang, X. L., Minetti, C., Li, M.,
Hou, H., Jr., Kneitz, B., Edelmann, W., and Lisanti, M. P. (2001). Caveolin-3
Bibliography
133
null mice show a loss of caveolae, changes in the microdomain
distribution of the dystrophin-glycoprotein complex, and t- tubule
abnormalities, J Biol Chem 276, 21425-33.
•
Geuze, H. J., Slot, J. W., and Schwartz, A. L. (1987). Membranes of sorting
organelles display lateral heterogeneity in receptor distribution, J Cell Biol
104, 1715-23.
•
Geuze, H. J., Slot, J. W., Strous, G. J., Lodish, H. F., and Schwartz, A. L.
(1983). Intracellular site of asialoglycoprotein receptor-ligand uncoupling:
double-label immunoelectron microscopy during receptor-mediated
endocytosis, Cell 32, 277-87.
•
Gilbert, J. M., and Benjamin, T. L. (2000). Early steps of polyomavirus
entry into cells, J Virol 74, 8582-8.
•
Glenney, J. R., Jr. (1989). Tyrosine phosphorylation of a 22-kDa protein is
correlated with transformation by Rous sarcoma virus, J Biol Chem 264,
20163-6.
•
Gold, E. S., Morrissette, N. S., Underhill, D. M., Guo, J., Bassetti, M., and
Aderem, A. (2000). Amphiphysin IIm, a novel amphiphysin II isoform, is
required for macrophage phagocytosis, Immunity 12, 285-92.
•
Gold, E. S., Underhill, D. M., Morrissette, N. S., Guo, J., McNiven, M. A.,
and Aderem, A. (1999). Dynamin 2 is required for phagocytosis in
macrophages, J Exp Med 190, 1849-56.
•
Goode, B. L., Wong, J. J., Butty, A. C., Peter, M., McCormack, A. L., Yates,
J. R., Drubin, D. G., and Barnes, G. (1999). Coronin promotes the rapid
assembly and cross-linking of actin filaments and may link the actin and
microtubule cytoskeletons in yeast, J Cell Biol 144, 83-98.
•
Greber, U., and Kasamatsu, H. (1996). Nuclear targeting of SV40 and
adenovirus, TICB 6, 189-195.
•
Greenberg, S., Burridge, K., and Silverstein, S. C. (1990). Colocalization of
F-actin and talin during Fc receptor-mediated phagocytosis in mouse
macrophages, J Exp Med 172, 1853-6.
•
Greenberg, S., Chang, P., and Silverstein, S. C. (1994). Tyrosine
phosphorylation of the gamma subunit of Fc gamma receptors, p72syk,
and paxillin during Fc receptor-mediated phagocytosis in macrophages, J
Biol Chem 269, 3897-902.
134
•
Caveolae-mediated endocytosis of SV40
Greenberg, S., Chang, P., Wang, D. C., Xavier, R., and Seed, B. (1996).
Clustered syk tyrosine kinase domains trigger phagocytosis, Proc Natl
Acad Sci U S A 93, 1103-7.
•
Greenberg, S., el Khoury, J., di Virgilio, F., Kaplan, E. M., and Silverstein,
S. C. (1991). Ca(2+)-independent F-actin assembly and disassembly during
Fc receptor- mediated phagocytosis in mouse macrophages, J Cell Biol 113,
757-67.
•
Griffith, G. R., and Consigli, R. A. (1984). Isolation and characterization of
monopinocytotic vesicles containing polyomavirus from the cytoplasm of
infected mouse kidney cells, J Virol 50, 77-85.
•
Griffith, G. R., Marriot, S. J., Rintoul, D. A., and Consilgi, R. A. (1988).
Early events in polyomavirus infection: Fusion of monopinocytotic
vesicles with mouse kidney cells, Virus Res, 41-51.
•
Griffiths, G., Hoflack, B., Simons, K., Mellman, I., and Kornfeld, S. (1988).
The mannose 6-phosphate receptor and the biogenesis of lysosomes, Cell
52, 329-41.
•
Gruenberg, J. (2001). The endocytic pathway: a mosaic of domains, Nat
Rev Mol Cell Biol 2, 721-30.
•
Haffner, C., Di Paolo, G., Rosenthal, J. A., and de Camilli, P. (2000). Direct
interaction of the 170 kDa isoform of synaptojanin 1 with clathrin and with
the clathrin adaptor AP-2, Curr Biol 10, 471-4.
•
Haigler, H.T., McKanna, J.A., and Cohen, S. (1979). Rapid stimulation of
pinocytosis in human carcinoma cells A-431 by epidermal growth factor. J.
Cell Biol. 83, 82–90.
•
Harder, T., and Simons, K. (1999). Clusters of glycolipid and
glycosylphosphatidylinositol-anchored proteins in lymphoid cells:
accumulation of actin regulated by local tyrosine phosphorylation, Eur J
Immunol 29, 556-62.
•
Harris, H. E., and Weeds, A. G. (1984). Plasma gelsolin caps and severs
actin filaments, FEBS Lett 177, 184-8.
•
Harris, T. W., Hartwieg, E., Horvitz, H. R., and Jorgensen, E. M. (2000).
Mutations in synaptojanin disrupt synaptic vesicle recycling, J Cell Biol
150, 589-600.
Bibliography
•
135
Hartwig, J. H., Thelen, M., Rosen, A., Janmey, P. A., Nairn, A. C., and
Aderem, A. (1992). MARCKS is an actin filament crosslinking protein
regulated by protein kinase C and calcium-calmodulin, Nature 356, 618-22.
•
Haucke, V., and De Camilli, P. (1999). AP-2 recruitment to synaptotagmin
stimulated by tyrosine-based endocytic motifs, Science 285, 1268-71.
•
Helenius, A., Kartenbeck, J., Simons, K., and Fries, E. (1980). On the entry
of semliki forest virus into BHK-21 cells, J Cell Biol 84, 404 - 420.
•
Henley, J. R., Krueger, E. W., Oswald, B. J., and McNiven, M. A. (1998).
Dynamin-mediated internalization of caveolae, J Cell Biol 141, 85-99.
•
Herreros, J., Ng, T., and Schiavo, G. (2001). Lipid rafts act as specialized
domains for tetanus toxin binding and internalization into neurons, Mol
Biol Cell 12, 2947-60.
•
Hicke, L. (2001a). A new ticket for entry into budding vesicles-ubiquitin,
Cell 106, 527-30.
•
Hicke, L. (2001b). Protein regulation by monoubiquitin, Nat Rev Mol Cell
Biol 2, 195-201.
•
Higgs, H. N., and Pollard, T. D. (2001). Regulation of actin filament
network formation through ARP2/3 complex: activation by a diverse
array of proteins, Annu Rev Biochem 70, 649-76.
•
Hinshaw, J. E. (2000). Dynamin and its role in membrane fission, Annu
Rev Cell Dev Biol 16, 483-519.
•
Honda, A., Nogami, M., Yokozeki, T., Yamazaki, M., Nakamura, H.,
Watanabe, H., Kawamoto, K., Nakayama, K., Morris, A. J., Frohman, M.
A., and Kanaho, Y. (1999). Phosphatidylinositol 4-phosphate 5-kinase
alpha is a downstream effector of the small G protein ARF6 in membrane
ruffle formation, Cell 99, 521-32.
•
Horiuchi, H., Lippe, R., McBride, H. M., Rubino, M., Woodman, P.,
Stenmark, H., Rybin, V., Wilm, M., Ashman, K., Mann, M., and Zerial, M.
(1997). A novel Rab5 GDP/GTP exchange factor complexed to Rabaptin-5
links nucleotide exchange to effector recruitment and function, Cell 90,
1149-59.
•
Hummeler, K., Tomassini, N., and Sokol, F. (1970). Morphological aspects
of the uptake of Simian Virus 40 by permissive cells, J Virol 6, 87-93.
•
Huttner, W. B., and Schmidt, A. (2000). Lipids, lipid modification and
lipid-protein interaction in membrane budding and fission--insights from
136
Caveolae-mediated endocytosis of SV40
the roles of endophilin A1 and synaptophysin in synaptic vesicle
endocytosis, Curr Opin Neurobiol 10, 543-51.
•
Ikonen, E., and Parton, R. G. (2000). Caveolins and cellular cholesterol
balance, Traffic 1, 212-7.
•
Izadi, K. D., Erdreich-Epstein, A., Liu, Y., and Durden, D. L. (1998).
Characterization of Cbl-Nck and Nck-Pak1 interactions in myeloid
FcgammaRII signaling, Exp Cell Res 245, 330-42.
•
Kang, Y. S., Ko, Y. G., and Seo, J. S. (2000). Caveolin internalization by heat
shock or hyperosmotic shock, Exp Cell Res 255, 221-8.
•
Kartenbeck, J., Stukenbrok, H., and Helenius, A. (1989). Endocytosis of
simian virus 40 into the endoplasmic reticulum, J Cell Biol 109, 2721-9.
•
Kasamatsu, H., and Nakanishi, A. (1998). How do animal DNA viruses get
to the nucleus?, Annu Rev Microbiol 52, 627-86.
•
Kaufman, M., Leto, T., and Levy, R. (1996). Translocation of annexin I to
plasma membranes and phagosomes in human neutrophils upon
stimulation with opsonized zymosan: possible role in phagosome
function, Biochem J 316, 35-42.
•
Keely, P. J., Westwick, J. K., Whitehead, I. P., Der, C. J., and Parise, L. V.
(1997). Cdc42 and Rac1 induce integrin-mediated cell motility and
invasiveness through PI(3)K, Nature 390, 632-6.
•
Khoury, G., and Lai, C. J. (1979). Preparation of simian virus 40 and its
DNA, Methods Enzymol 58, 404-12.
•
Kiefer, F., Brumell, J., Al-Alawi, N., Latour, S., Cheng, A., Veillette, A.,
Grinstein, S., and Pawson, T. (1998). The Syk protein tyrosine kinase is
essential for Fcgamma receptor signaling in macrophages and neutrophils,
Mol Cell Biol 18, 4209-20.
•
Kirchhausen, T. (1999). Adaptors for clathrin-mediated traffic, Annu Rev
Cell Dev Biol 15, 705-32.
•
Kirchhausen, T. (2000). Clathrin, Annu Rev Biochem 69, 699-727.
•
Kornfeld, S., and Mellman, I. (1989). The biogenesis of lysosomes, Annu
Rev Cell Biol 5, 483-525.
•
Kurzchalia, T. V., Dupree, P., Parton, R. G., Kellner, R., Virta, H., Lehnert,
M., and Simons, K. (1992). VIP21, a 21-kD membrane protein is an integral
component of trans-Golgi- network-derived transport vesicles, J Cell Biol
118, 1003-14.
Bibliography
•
137
Kurzchalia, T. V., and Parton, R. G. (1999). Membrane microdomains and
caveolae, Curr Opin Cell Biol 11, 424-31.
•
Lalli, G., and Schiavo, G. (2002). Analysis of retrograde transport in motor
neurons reveals common endocytic carriers for tetanus toxin and
neurotrophin receptor p75NTR, J Cell Biol 156, 233-40.
•
Lamaze, C., Dujeancourt, A., Baba, T., Lo, C. G., Benmerah, A., and
Dautry-Varsat, A. (2001). Interleukin 2 receptors and detergent-resistant
membrane domains define a clathrin-independent endocytic pathway,
Mol Cell 7, 661-71.
•
Lange, Y., Ye, J., Rigney, M., and Steck, T. L. (1999). Regulation of
endoplasmic reticulum cholesterol by plasma membrane cholesterol, J
Lipid Res 40, 2264-70.
•
Laux, T., Fukami, K., Thelen, M., Golub, T., Frey, D., and Caroni, P. (2000).
GAP43, MARCKS, and CAP23 modulate PI(4,5)P(2) at plasmalemmal
rafts, and regulate cell cortex actin dynamics through a common
mechanism, J Cell Biol 149, 1455-72.
•
Le, P. U., Benlimame, N., Lagana, A., Raz, A., and Nabi, I. R. (2000).
Clathrin-mediated endocytosis and recycling of autocrine motility factor
receptor to fibronectin fibrils is a limiting factor for NIH-3T3 cell motility, J
Cell Sci 113, 3227-40.
•
Le, P. U., Guay, G., Altschuler, Y., and Nabi, I. R. (2001). Caveolin-1 is a
negative regulator of caveolae-mediated endocytosis to the endoplasmic
reticulum, J Biol Chem 27, 27.
•
Lee, E., and De Camilli, P. (2002). Dynamin at actin tails, Proc Natl Acad
Sci U S A 99, 161-6.
•
Lemmon, S. K. (2001). Clathrin uncoating: Auxilin comes to life, Curr Biol
11, R49-52.
•
Lencer, W. I., Hirst, T. R., and Holmes, R. K. (1999). Membrane traffic and
the cellular uptake of cholera toxin, Biochim Biophys Acta 1450, 177-90.
•
Liddington, R. C., Yan, Y., Moulai, J., Sahli, R., Benjamin, T. L., and
Harrison, S. C. (1991). Structure of Simian virus 40 at 3.8-Å resolution,
Nature 354, 278-284.
•
Lisanti, M. P., Tang, Z. L., and Sargiacomo, M. (1993). Caveolin forms a
hetero-oligomeric protein complex that interacts with an apical GPI-linked
protein: implications for the biogenesis of caveolae, J Cell Biol 123, 595-604.
138
•
Caveolae-mediated endocytosis of SV40
Lorenzi, R., Brickell, P. M., Katz, D. R., Kinnon, C., and Thrasher, A. J.
(2000). Wiskott-Aldrich syndrome protein is necessary for efficient IgGmediated phagocytosis, Blood 95, 2943-6.
•
Machesky, L. M., and Insall, R. H. (1998). Scar1 and the related WiskottAldrich syndrome protein, WASP, regulate the actin cytoskeleton through
the Arp2/3 complex, Curr Biol 8, 1347-56.
•
Machesky, L. M., Mullins, R. D., Higgs, H. N., Kaiser, D. A., Blanchoin, L.,
May, R. C., Hall, M. E., and Pollard, T. D. (1999). Scar, a WASp-related
protein, activates nucleation of actin filaments by the Arp2/3 complex,
Proc Natl Acad Sci U S A 96, 3739-44.
•
Mackay, R. L., and Consiligi, R. A. (1976). Early events in polyoma virus
infection: Attachment, penetration, and nuclear entry, J Virol 19, 620-636.
•
Marechal, V., Prevost, M.C., Petit, C., Perret, E., Heard, J.M. and Schwartz,
O. (2001). Human immunodeficiency virus type 1 entry into macrophages
mediated by macropinocytosis, J Virol 75, 11166-11177.
•
Marjomaki, V., Pietiainen, V., Matilainen, H., Upla, P., Ivaska, J., Nissinen,
L., Reunanen, H., Huttunen, P., Hyypia, T., and Heino, J. (2002).
Internalization of Echovirus 1 in Caveolae, J Virol 76, 1856-1865.
•
Marsh, M., Bolzau, E., and Helenius, A. (1983). Penetration of Semliki
Forest virus from acidic prelysosomal vacuoles., Cell 32, 931-940.
•
Marsh, M., and Helenius, A. (1989). Virus entry into animal cells, Adv
Virus Res 36, 107-151.
•
Marsh, M., and McMahon, H. T. (1999). The structural era of endocytosis,
Science 285, 215-20.
•
Martin, T. F. (2001). PI(4,5)P(2) regulation of surface membrane traffic,
Curr Opin Cell Biol 13, 493-9.
•
Maul, G. G., Rovera, G., Vorbrodt, A., and Abramczuk, J. (1978).
Membrane fusion as a mechanism of Simian Virus 40 entry into different
cellular compartments, J Virol 28, 936-944.
•
Maxfield, F. R. (1982). Weak bases and ionophores rapidly and reversibly
raise the pH of endocytic vesicles in cultured mouse fibroblasts, J Cell Biol
95, 676-81.
•
Maxfield, F. R. (1985). Acidification of endocytic vesicles and lysosomes.
(New York, Plenum Publishing Corp.).
Bibliography
•
139
May, R. C., and Machesky, L. M. (2001). Phagocytosis and the actin
cytoskeleton, J Cell Sci 114, 1061-77.
•
Mayor, S., Rothberg, K. G., and Maxfield, F. R. (1994). Sequestration of
GPI-anchored proteins in caveolae triggered by cross-linking, Science 264,
1948-51.
•
McBride, H. M., Rybin, V., Murphy, C., Giner, A., Teasdale, R., and Zerial,
M. (1999). Oligomeric complexes link Rab5 effectors with NSF and drive
membrane fusion via interactions between EEA1 and syntaxin 13, Cell 98,
377-86.
•
McMahon, H. T. (1999). Endocytosis: an assembly protein for clathrin
cages, Curr Biol 9, R332-5.
•
Meier, O., Boucke, K., Hammer, S.V., Keller, S., Stidwill, R.P., Hemmi, S.,
Greber, U.F. (2002). Adenovirus triggers macropinocytosis and endosomal
leakage together with its clathrin-mediated uptake, J Cell Biol 158 11191131.
•
Mellman, I. (1996). Endocytosis and molecular sorting, Annu Rev Cell Dev
Biol 12, 575-625.
•
Mellman, I., R. Fuchs, and A. Helenius (1985). Acidification of endocytic
and exocytic pathways, Ann Rev Biochem 55, 663 - 700.
•
Merrifield, C. J., Moss, S. E., Ballestrem, C., Imhof, B. A., Giese, G.,
Wunderlich, I., and Almers, W. (1999). Endocytic vesicles move at the tips
of actin tails in cultured mast cells, Nat Cell Biol 1, 72-4.
•
Metchnikoff, E. (1887). Sur la lutte des cellules de l'organisme cintre
l'invasion des microbes, Ann Inst Pasteur 1, 321-326.
•
Minshall, R. D., Tiruppathi, C., Vogel, S. M., Niles, W. D., Gilchrist, A.,
Hamm, H. E., and Malik, A. B. (2000). Endothelial cell-surface gp60
activates vesicle formation and trafficking via G(i)-coupled Src kinase
signaling pathway, J Cell Biol 150, 1057-70.
•
Monier, S., Parton, R. G., Vogel, F., Behlke, J., Henske, A., and Kurzchalia,
T. V. (1995). VIP21-caveolin, a membrane protein constituent of the
caveolar coat, oligomerizes in vivo and in vitro, Mol Biol Cell 6, 911-27.
•
Montesano, R., Roth, J., Robert, A., and Orci, L. (1982). Non-coated
membrane invaginations are involved in binding and internalization of
cholera and tetanus toxins, Nature 296, 651-3.
140
•
Caveolae-mediated endocytosis of SV40
Mullins, R. D., Heuser, J. A., and Pollard, T. D. (1998). The interaction of
Arp2/3 complex with actin: nucleation, high affinity pointed end capping,
and formation of branching networks of filaments, Proc Natl Acad Sci U S
A 95, 6181-6.
•
Munro, P., Kojima, H., Dupont, J. L., Bossu, J. L., Poulain, B., and Boquet,
P. (2001). High sensitivity of mouse neuronal cells to tetanus toxin requires
a GPI-anchored protein, Biochem Biophys Res Commun 289, 623-9.
•
Murata, M., Peranen, J., Schreiner, R., Wieland, F., Kurzchalia, T. V., and
Simons, K. (1995). VIP21/caveolin is a cholesterol-binding protein, Proc
Natl Acad Sci U S A 92, 10339-43.
•
Nakano, M. Y., and Greber, U. F. (2000). Quantitative microscopy of
fluorescent adenovirus entry, J Struct Biol 129, 57-68.
•
Newmyer, S. L., and Schmid, S. L. (2001). Dominant-interfering Hsc70
mutants disrupt multiple stages of the clathrin-coated vesicle cycle in vivo,
J Cell Biol 152, 607-20.
•
Nichols, B. J. (2002). A distinct class of endosome mediates clathrinindependent endocytosis to the Golgi complex, Nat Cell Biol 4, 374-8.
•
Nichols, B. J., Kenworthy, A. K., Polishchuk, R. S., Lodge, R., Roberts, T.
H., Hirschberg, K., Phair, R. D., and Lippincott-Schwartz, J. (2001). Rapid
cycling of lipid raft markers between the cell surface and Golgi complex, J
Cell Biol 153, 529-41.
•
Nichols, B. J., and Lippincott-Schwartz, J. (2001). Endocytosis without
clathrin coats, Trends Cell Biol 11, 406-12.
•
Nielsen, E., Severin, F., Backer, J. M., Hyman, A. A., and Zerial, M. (1999).
Rab5 regulates motility of early endosomes on microtubules, Nat Cell Biol
1, 376-382.
•
Norkin, L. C. (1999). Simian virus 40 infection via MHC class I molecules
and caveolae, Immunol Rev 168, 13-22.
•
Oh, P., McIntosh, D. P., and Schnitzer, J. E. (1998). Dynamin at the neck of
caveolae mediates their budding to form transport vesicles by GTP-driven
fission from the plasma membrane of endothelium, J Cell Biol 141, 101-14.
•
Ohkuma, S., and Poole, B. (1978). Fluorescence probe measurement of the
intralysosomal pH in living cells and the perturbation of pH by various
agents, Proc Natl Aca Sci USA 75, 3327-3331.
141
Bibliography
•
Ostermeyer, A. G., Paci, J. M., Zeng, Y., Lublin, D. M., Munro, S., and
Brown, D. A. (2001). Accumulation of caveolin in the endoplasmic
reticulum redirects the protein to lipid storage droplets, J Cell Biol 152,
1071-8.
•
Palade, G. E. (1953). Fine structure of blood capillaries, J Appl Physiol 24,
1424.
•
Parton, R. G. (1996). Caveolae and caveolins, Curr Op Cell Biol 8, 542-548.
•
Parton, R. G., Joggerst, B., and Simons, K. (1994). Regulated internalization
of caveolae, J Cell Biol 127, 1199-215.
•
Parton, R. G., and Lindsay, M. (1999). Exploitation of major
histocompatibility complex class I molecules and caveole by simian vius
40, Immunol Rev 168, 23-31.
•
Parton, R. G., Way, M., Zorzi, N., and Stang, E. (1997). Caveolin-3
associates with developing T-tubules during muscle differentiation, J Cell
Biol 136, 137-54.
•
Pearse, B. M. (1975). Coated vesicles from pig brain: purification and
biochemical characterization, J Mol Biol 97, 93-8.
•
Pelkmans, L., and Helenius, A. (1999). Expression of antibody interferes
with disulfide bond formation and intracellular transport of antigen in the
secretory pathway, J Biol Chem 274, 14495-9.
•
Pelkmans, L., Kartenbeck, J., and Helenius, A. (2001). Caveolar endocytosis
of simian virus 40 reveals a new two-step vesicular-transport pathway to
the ER, Nat Cell Biol 3, 473-83.
•
Pelkmans, L., Puntener, D., and Helenius, A. (2002). Local Actin
Polymerization
and
Dynamin
Recruitment
in
SV40-Induced
Internalization of Caveolae, Science 296, 535-9.
•
Peters, K. R., Carley, W. W., and Palade, G. E. (1985). Endothelial
plasmalemmal vesicles have a characteristic striped bipolar surface
structure, J Cell Biol 101, 2233-8.
•
Pfeffer, S. R. (2001). Caveolae on the move, Nat Cell Biol 3, E108-10.
•
Pol, A., Luetterforst, R., Lindsay, M., Heino, S., Ikonen, E., and Parton, R.
G. (2001). A caveolin dominant negative mutant associates with lipid
bodies and induces intracellular cholesterol imbalance, J Cell Biol 152,
1057-70.
142
•
Caveolae-mediated endocytosis of SV40
Ponnambalam, S., Girotti, M., Yaspo, M. L., Owen, C. E., Perry, A. C.,
Suganuma, T., Nilsson, T., Fried, M., Banting, G., and Warren, G. (1996).
Primate homologues of rat TGN38: primary structure, expression and
functional implications, J Cell Sci 109, 675-85.
•
Puri, V., Watanabe, R., Singh, R. D., Dominguez, M., Brown, J. C.,
Wheatley, C. L., Marks, D. L., and Pagano, R. E. (2001). Clathrindependent and -independent internalization of plasma membrane
sphingolipids initiates two Golgi targeting pathways, J Cell Biol 154, 53547.
•
Qualmann, B., and Kelly, R. B. (2000). Syndapin isoforms participate in
receptor-mediated endocytosis and actin organization, J Cell Biol 148,
1047-62.
•
Quilliam, L. A., Lambert, Q. T., Mickelson-Young, L. A., Westwick, J. K.,
Sparks, A. B., Kay, B. K., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., and
Der, C. J. (1996). Isolation of a NCK-associated kinase, PRK2, an SH3binding protein and potential effector of Rho protein signaling, J Biol
Chem 271, 28772-6.
•
Raiborg, C., Bache, K. G., Gillooly, D. J., Madshus, I. H., Stang, E., and
Stenmark, H. (2002). Hrs sorts ubiquitinated proteins into clathrin-coated
microdomains of early endosomes, Nat Cell Biol 4, 394-8.
•
Razani, B., Engelman, J. A., Wang, X. B., Schubert, W., Zhang, X. L., Marks,
C. B., Macaluso, F., Russell, R. G., Li, M., Pestell, R. G., et al. (2001).
Caveolin-1 null mice are viable but show evidence of hyperproliferative
and vascular abnormalities, J Biol Chem 276, 38121-38.
•
Reaves, B., Horn, M., and Banting, G. (1993). TGN38/41 recycles between
the cell surface and the TGN: brefeldin A affects its rate of return to the
TGN, Mol Biol Cell 4, 93-105.
•
Richards, A. A., Stang, E., Pepperkok, R., and Parton, R. G. (2002).
Inhibitors of COP-mediated Transport and Cholera Toxin Action Inhibit
Simian Virus 40 Infection, Mol Biol Cell 13, 1750-64.
•
Richterova, Z., Liebl, D., Horak, M., Palkova, Z., Stokrova, J., Hozak, P.,
Korb, J., and Forstova, J. (2001). Caveolae Are Involved in the Trafficking
of Mouse Polyomavirus Virions and Artificial VP1 Pseudocapsids toward
Cell Nuclei, J Virol 75, 10880-91.
Bibliography
•
143
Ricotta, D., Conner, S. D., Schmid, S. L., von Figura, K., and Honing, S.
(2002). Phosphorylation of the AP2 mu subunit by AAK1 mediates high
affinity binding to membrane protein sorting signals, J Cell Biol 156, 791-5.
•
Rijnboutt, S., Jansen, G., Posthuma, G., Hynes, J. B., Schornagel, J. H., and
Strous, G. J. (1996). Endocytosis of GPI-linked membrane folate receptoralpha, J Cell Biol 132, 35-47.
•
Rohatgi, R., Ma, L., Miki, H., Lopez, M., Kirchhausen, T., Takenawa, T.,
and Kirschner, M. W. (1999). The interaction between N-WASP and the
Arp2/3 complex links Cdc42- dependent signals to actin assembly, Cell 97,
221-31.
•
Roth, T. F., and Porter, K. R. (1964). Yolk protein uptake in the oocyte of
the mosquito, Aedes aegypti L J Cell Biol 20, 313.
•
Rothberg, K. G., Heuser, J. E., Donzell, W. C., Ying, Y. S., Glenney, J. R.,
and Anderson, R. G. (1992). Caveolin, a protein component of caveolae
membrane coats, Cell 68, 673-82.
•
Rothberg, K. G., Ying, Y. S., Kolhouse, J. F., Kamen, B. A., and Anderson,
R. G. (1990). The glycophospholipid-linked folate receptor internalizes
folate without entering the clathrin-coated pit endocytic pathway, J Cell
Biol 110, 637-49.
•
Roy, S., Luetterforst, R., Harding, A., Apolloni, A., Etheridge, M., Stang, E.,
Rolls, B., Hancock, J. F., and Parton, R. G. (1999). Dominant-negative
caveolin inhibits H-Ras function by disrupting cholesterol-rich plasma
membrane domains [see comments], Nat Cell Biol 1, 98-105.
•
Rozelle, A. L., Machesky, L. M., Yamamoto, M., Driessens, M. H., Insall, R.
H., Roth, M. G., Luby-Phelps, K., Marriott, G., Hall, A., and Yin, H. L.
(2000). Phosphatidylinositol 4,5-bisphosphate induces actin-based
movement of raft-enriched vesicles through WASP-Arp2/3 [In Process
Citation], Curr Biol 10, 311-20.
•
Rubtsov, A. M., and Lopina, O. D. (2000). Ankyrins, FEBS Lett 482, 1-5.
•
Rybin, V., Ullrich, O., Rubino, M., Alexandrov, K., Simon, I., Seabra, M. C.,
Goody, R., and Zerial, M. (1996). GTPase activity of Rab5 acts as a timer
for endocytic membrane fusion, Nature 383, 266-9.
•
Sahil, R., Freund, R., Dibensky, T., Garcea, R., and Benjamin, T. (1993).
Defect in entry and altered pathogenecity of a polyome virus mutant
blocked in VP2 myristylation., Virology 192, 142-153.
144
•
Caveolae-mediated endocytosis of SV40
Sahli, R., and Beard, P. (1994). Propagation and purification of Polyoma
and Simian Virus 40. In Cell Biology: A laboratory handbook, J. E. Celis,
ed. (San Diego, Academic Press, Inc), pp. 471-478.
•
Sandvig, K., Garred, Ø., Prydz, K., Kozolov, J. V., Hansen, S. H., and van
Deurs, B. (1992). Retrograde transport of endocytosed Shiga toxin to the
endoplasmic reticulum, Nature 358, 510-512.
•
Sargiacomo, M., Sudol, M., Tang, Z., and Lisanti, M. P. (1993). Signal
transducing molecules and glycosyl-phosphatidylinositol-linked proteins
form a caveolin-rich insoluble complex in MDCK cells, J Cell Biol 122, 789807.
•
Schimmoller, F., Simon, I., and Pfeffer, S. R. (1998). Rab GTPases, directors
of vesicle docking, J Biol Chem 273, 22161-4.
•
Schlegel, A., and Lisanti, M. P. (2001). Caveolae and their coat proteins, the
caveolins: from electron microscopic novelty to biological launching pad, J
Cell Physiol 186, 329-37.
•
Schmid, S., Fuchs, R., Kielian, M., Helenius, A., and Mellman, I. (1989).
Acidification of endosome subpopulations in wild-type chinese hamster
ovary cells and temperature-sensitive acification-defective mutants, JCell
Biol 108, 1291-1300.
•
Schmid, S. L., and Carter, L. L. (1990). ATP is required for receptormediated endocytosis in intact cells, J Cell Biol 111, 2307-18.
•
Schnitzer, J. E., and Bravo, J. (1993). High affinity binding, endocytosis,
and degradation of conformationally modified albumins. Potential role of
gp30 and gp18 as novel scavenger receptors, J Biol Chem 268, 7562-70.
•
Schnitzer, J. E., Liu, J., and Oh, P. (1995). Endothelial caveolae have the
molecular transport machinery for vesicle budding, docking, and fusion
including VAMP, NSF, SNAP, annexins, and GTPases, J Biol Chem 270,
14399-404.
•
Schnitzer, J. E., Oh, P., Pinney, E., and Allard, J. (1994). Filipin-sensitive
caveolae-mediated transport in endothelium: reduced transcytosis,
scavenger endocytosis, and capillary permeability of select
macromolecules, J Cell Biol 127, 1217-32.
•
Sever, S., Damke, H., and Schmid, S. L. (2000). Garrotes, springs, ratchets,
and whips: putting dynamin models to the test, Traffic 1, 385-92.
Bibliography
•
145
Shimura, H., Umeno, Y., and Kimura, G. (1987). Effects of inhibitors of the
cytoplasmic structures and functions on the early phase of infection of
cultured cells with simian virus 40, Virology 158, 34-43.
•
Shin, J. S., and Abraham, S. N. (2001). Caveolae as portals of entry for
microbes, Microbes Infect 3, 755-61.
•
Shin, J. S., Gao, Z., and Abraham, S. N. (2000). Involvement of cellular
caveolae in bacterial entry into mast cells [In Process Citation], Science 289,
785-8.
•
Shogomori, H., and Futerman, A. H. (2001). Cholera toxin is found in
detergent-insoluble rafts/domains at the cell surface of hippocampal
neurons but is internalized via a raft- independent mechanism, J Biol
Chem 276, 9182-8.
•
Simons, K., H. Garoff, and A. Helenius (1982). How an animal virus gets
into and out of its host cell, Sci American 246, 58-66.
•
Simons, K., and Ikonen, E. (1997). Functional rafts in cell membranes,
Nature 387, 569-72.
•
Simons, K., and Toomre, D. (2000). Lipid rafts and signal transduction, Nat
Rev Mol Cell Biol 1, 31-9.
•
Simonsen, A., Gaullier, J. M., D'Arrigo, A., and Stenmark, H. (1999). The
Rab5 effector EEA1 interacts directly with syntaxin-6, J Biol Chem 274,
28857-60.
•
Simonsen, A., Wurmser, A. E., Emr, S. D., and Stenmark, H. (2001). The
role of phosphoinositides in membrane transport, Curr Opin Cell Biol 13,
485-92.
•
Smart, E. J., Graf, G. A., McNiven, M. A., Sessa, W. C., Engelman, J. A.,
Scherer, P. E., Okamoto, T., and Lisanti, M. P. (1999). Caveolins, liquidordered domains and signal transduction, Mol Cell Biol 19, 7289-7304.
•
Smart, E. J., Ying, Y. S., Conrad, P. A., and Anderson, R. G. (1994).
Caveolin moves from caveolae to the Golgi apparatus in response to
cholesterol oxidation, J Cell Biol 127, 1185-97.
•
Smart, E. J., Ying, Y. S., Donzell, W. C., and Anderson, R. G. (1996). A role
for caveolin in transport of cholesterol from endoplasic reticulum to
plasma membrane, J Biol Chem 271, 29427-35.
146
•
Caveolae-mediated endocytosis of SV40
Stan, R. V., Ghitescu, L., Jacobson, B. S., and Palade, G. E. (1999). Isolation,
cloning, and localization of rat PV-1, a novel endothelial caveolar protein, J
Cell Biol 145, 1189-98.
•
Stan, R. V., Roberts, W. G., Predescu, D., Ihida, K., Saucan, L., Ghitescu, L.,
and Palade, G. E. (1997). Immunoisolation and partial charaterization of
endothelial plasmalemmal vesicles (caveolae), Mol Biol Cell 8, 595-605.
•
Stang, E., Kartenbeck, J., and Parton, R. G. (1997). Major histocompatibility
complex class I molecules mediate association of SV40 with caveolae, Mol
Biol Cell 8, 47-57.
•
Steegmaier, M., Oorschot, V., Klumperman, J., and Scheller, R. H. (2000).
Syntaxin 17 is abundant in steroidogenic cells and implicated in smooth
endoplasmic reticulum membrane dynamics [In Process Citation], Mol
Biol Cell 11, 2719-31.
•
Steegmaier, M., Yang, B., Yoo, J.-S., Huang, B., Shen, M., Yu, S., Luo, Y.,
and Sheller, R. H. (1998). Three novel proteins of the Syntaxin/SNAP-25
family, J Biol Chem 273, 34171-34179.
•
Stehle, T., Gamblin, S. J., Yan, Y., and Harrison, S. C. (1996). The structure
of simian virus 40 refined at 3.1 A resolution, Structure 4, 165-82.
•
Stehle, T., Yan, Y., Benjamin, T. L., and Harrison, S. C. (1994). Structure of
murine polyomavirus complexed with an oligosaccharide receptor
fragment, Nature 369, 160-3.
•
Steinman, R. M., and Swanson, J. (1995). The endocytic activity of dendritic
cells, J Exp Med 182, 283-8.
•
Taunton, J., Rowning, B. A., Coughlin, M. L., Wu, M., Moon, R. T.,
Mitchison, T. J., and Larabell, C. A. (2000). Actin-dependent propulsion of
endosomes and lysosomes by recruitment of N-WASP, J Cell Biol 148, 51930.
•
Thomas, S. M., and Brugge, J. S. (1997). Cellular functions regulated by Src
family kinases, Annu Rev Cell Dev Biol 13, 513-609.
•
Thomsen, P., Roepstorff, K., Stahlhut, M., and van Deurs, B. (2002).
Caveolae Are Highly Immobile Plasma Membrane Microdomains, Which
Are not Involved in Constitutive Endocytic Trafficking, Mol Biol Cell 13,
238-50.
Bibliography
•
147
Torgersen, M. L., Skretting, G., van Deurs, B., and Sandvig, K. (2001).
Internalization of cholera toxin by different endocytic mechanisms, J Cell
Sci 114, 3737-47.
•
Underhill, D. M., and Ozinsky, A. (2002). Phagocytosis of microbes:
complexity in action, Annu Rev Immunol 20, 825-52.
•
Upcroft, P. (1987). Simian virus 40 infection is not mediated by lysosomal
activation, J Gen Virol 678, 2477-2480.
•
van der Goot, F. G., and Harder, T. (2001). Raft membrane domains: from
a liquid-ordered membrane phase to a site of pathogen attack, Semin
Immunol 13, 89-97.
•
Velasco, A., Henrdicks, L., Moremen, K. W., Tulsiani, D. R. P., Touster, O.,
and Fraquhar, M. G. (1993). Cell type-dependent variations in the
subcellular distribution of α-mannosidase I and II, J Cell Biol 122, 39-51.
•
Verkade, P., Harder, T., Lafont, F., and Simons, K. (2000). Induction of
caveolae in the apical plasma membrane of Madin-Darby canine kidney
cells, J Cell Biol 148, 727-39.
•
Ward, B. M., and Moss, B. (2001). Vaccinia Virus Intracellular Movement Is
Associated with Microtubules and
•
Independent of Actin Tails., J Virol 75, 11651-11663.
•
Wendland, B., and Emr, S. D. (1998). Pan1p, yeast eps15, functions as a
multivalent adaptor that coordinates protein-protein interactions essential
for endocytosis, J Cell Biol 141, 71-84.
•
Wendland, B., Emr, S. D., and Riezman, H. (1998). Protein traffic in the
yeast endocytic and vacuolar protein sorting pathways, Curr Opin Cell
Biol 10, 513-22.
•
Werling, D., Hope, J. C., Chaplin, P., Collins, R. A., Taylor, G., and
Howard, C. J. (1999). Involvement of caveolae in the uptake of respiratory
syncytial virus antigen by dendritic cells, J Leukoc Biol 66, 50-8.
•
Wilde, A., and Brodsky, F. M. (1996). In vivo phosphorylation of adaptors
regulates their interaction with clathrin, J Cell Biol 135, 635-45.
•
Whittaker, G. R., Kann, M., and Helenius, A. (2000). Viral entry into the
nucleus, Annu Rev Cell Dev Biol 16, 627-51.
•
Wigge, P., and McMahon, H. T. (1998). The amphiphysin family of
proteins and their role in endocytosis at the synapse, Trends Neurosci 21,
339-44.
148
•
Caveolae-mediated endocytosis of SV40
Yamada, E. (1955). The fine structure of the gall bladder epithelium of the
mouse, J Biophys Biochem Cytol 1, 445-458.
•
Yamada, M., and Kasamatsu, H. (1993). Role of nuclear pore complex in
simian virus 40 nuclear targeting, J Virol 67, 119-30.
•
Yin, H. L., Albrecht, J. H., and Fattoum, A. (1981). Identification of
gelsolin, a Ca2+-dependent regulatory protein of actin gel-sol
transformation, and its intracellular distribution in a variety of cells and
tissues, J Cell Biol 91, 901-6.
•
Zerial, M., and McBride, H. (2001). Rab proteins as membrane organizers,
Nat Rev Mol Cell Biol 2, 107-17.
Acknowledgements
149
Acknowledgements
This dissertation would not have been completed without the help of many people,
most importantly Ari. I must admit, I did not know anything of your work until
Peter van der Sluijs (thank you Peter) persuaded me to contact you and ask for a
place in your lab to do my second major for my Masters degree. This was without
doubt the best choice I could have made. I realized soon that the specific
environment in your lab, combined with your remarkable skill to ‘see’ interesting
biological questions and your typical way of supervising people, giving them a great
amount of freedom, would be hard to find somewhere else. Although I enjoyed
working on glycoprotein folding during that period, I got the feeling that I wanted to
do something more cell-biological, and maybe something a bit more exploring,
during my Ph.D. training. Fortunately, also this was possible in your lab and in fact,
the approach to use viruses to learn more about uptake pathways in the cell was
something that appealed to me very much. Thank you Ari! My decision to work on
Simian Virus 40 was also influenced by Anna, who had just set up the live
microscopy equipment in the lab. This proved to be extremely important for my
work. Thank you Anna! Karin, thank you very much for your excellent technical
support and phenomenal virus preps. I am sure the EM will be even better.
Benchmates Anthony and Klaus and labmates Mauri and Ivo, thanks a lot for many
very helpful suggestions. To see things sometimes from another perspective would
not have been possible without the many zigi-pauses, as well as the many Fridayevening beers. Thank you Andreas, Kowi, Anna, Marius and Lars. Also, I would like
to thank ‘my students’. Lydia, being an ERASMUS student from Holland, just as I
was, it was good to be able to speak Dutch again in the lab. Thank you for taking up
the difficult task to isolate caveosomes. You have come a long way, and we are
continuing on your work. Daniel, you brought humor, a relaxed attitude, good labskills and a wish to learn the ‘scientific approach’ with you. I couldn’t have asked for
more. Finally your assay worked, thank you very much! Nathalie and Christopher,
thank you for your work. Although not very concrete yet, I am sure your
contributions will soon be. Finally I want to thank the people outside the lab, who,
indirectly, contributed tremendously. ‘De Ranzige Aap’ and ‘Ozewiezewoze’, thanks
for the support from Holland. I know, we should go sailing (Sneekweek), skiing,
snowboarding, hiking, eating, drinking and on jc-weekends more often. Who knows,
maybe I’ll return to Holland one day. Papa, Mama, Simone, Mathijs, Remko and Julie
thanks for your continuous support and curiosity in what I do. Maryse, you were my
best supporter for a long time. Thank you for everything! Doris, for you the most
prominent place! A year ago you came into my life and managed to have more
influence on it than anyone else. Thank you for being there.
150
Caveolae-mediated endocytosis of SV40
Curriculum vitae
The author of this dissertation was born on September 2nd, 1975 in Nijmegen,
The Netherlands. After graduating from ‘Stedelijk Gymnasium Nijmegen‘ in
June 1993, he started studying Medical Biology at the University of Utrecht. In
September 1994, the propadeutic degree was obtained (with honors). From
September 1996 until September 1997 he temporarily stopped his studies to
take seat in the board of the Utrecht Student Society ‘Collegium Studiosorum
Veritas’. In June 1999, the Master of Sciences degree (with honors) was
obtained with two majors: One at the department of Cell Biology of the
University of Utrecht Medical School on the function of small GTPases of the
rab family and the second, being an ERASMUS-scholarship student, at the
institute of Biochemistry of the Swiss Federal Institute of Technology in
Zürich on glycoprotein folding in mammalian cells. Since December 1999 he
was a Ph.D. student at the same institute under the supervision of Prof. Dr.
Ari Helenius, where he took part in studying the entry pathways of
mammalian viruses into their host cell, with emphasis on the entry pathway
of Simian Virus 40.
List of publications
151
List of publications
Pelkmans, L., Helenius, A. (1999). Expression of antibody interferes with
disulfide bond formation and intracellular transport of antigen in the
secretory pathway, Journal of Biological Chemistry 274, 14495-14499.
Nagelkerken, B., van Anken, E., van Raak, M., Gerez, L., Mohrmann, K., van
Uden, N., Holthuizen, J., Pelkmans, L., van der Sluijs, P. (2000). Rabaptin4, a
novel effector of the small GTPase rab4a, is recruited to perinuclear recycling
vesicles, Biochemical Journal 346, 593-601.
Pelkmans, L., Kartenbeck, J., Helenius, A. (2001). Caveolar endocytosis of
simian virus 40 reveals a new two-step vesicular-transport pathway to the ER,
Nature Cell Biology 3, 473-83.
Pelkmans, L., Püntener, D., Helenius, A. (2002). Local actin polymerization
and dynamin recruitment in SV40-induced internalization of caveolae, Science
296, 535-539.
Pelkmans, L., Helenius, A. (2002). Endocytosis via caveolae, Traffic 3, 311-320.