Content uploaded by Alexander Bershadsky
Author content
All content in this area was uploaded by Alexander Bershadsky
Content may be subject to copyright.
Proc.
Natl.
Acad.
Sci.
USA
Vol.
91,
pp.
5686-5689,
June
1994
Cell
Biology
Disruption
of
the
Golgi
apparatus
by
brefeldin
A
blocks
cell
polarization
and
inhibits
directed
cell
migration
ALEXANDER
D.
BERSHADSKY*t
AND
ANTHONY
H.
FUTERMAN*
Departments
of
*Chemical
Immunology
and
tMembrane
Research
and
Biophysics,
Weizmann
Institute
of
Science,
Rehovot
76100,
Israel
Communicated
by
Israel
Gelfand,
March
14,
1994
(received
for
review
January
5,
1994)
ABSTRACT
The
role
of
the
Golgi
apparatus
in
the
motile
activity
of
fibroblasts
was
examined
with
brefeldin
A
(BFA),
which
disrupts
the
Golgi
apparatus
in
a
variety
of
cells.
Upon
incubation
with
BFA,
Swiss
mouse
3T3
fibroblasts
lost
their
typical
polarized
morphology,
in
which
the
leading
edge
is
characterized
by
intensive
lamellipodia
formation.
BFA
af-
fected
cell
asymmetry
as
demonstrated
by
a
decrease
in
the
morphometric
indices,
dispersion,
and
elongation.
After
BFA
treatment,
cells
showed
little
protrusional
activity
and
did
not
form
a
dense
actin
network
at
the
leading
edge,
and
conse-
quently
the
rate
of
cell
migration
into
an
experimental
wound
was
significantly
reduced.
In
addition,
BFA
prevented
an
increase
in
pseudopodial
activity
and
prevented
the
formation
of
long
processes
induced
by
phorbol
12-myristate
13-acetate.
The
effects
of
BFA
on
cell
shape
and
protrusional
activity
were
quantitatively
similar
to
those
observed
with
the
microtubule-
disrupting
agent
nocodazole,
although
BFA
had
no
effect
on
microtubule
integrity.
These
results
suggest
that
the
integrity
of
both
the
Golgi
apparatus
and
microtubules
is
necessary
for
the
generation
and
maintenance
of
fibroblast
asymmetry,
which
is
a
prerequisite
for
directed
cell
migration.
Directed
migration
of
a
variety
of
cell
types
depends
on
the
polarized
distribution
of
protrusions
(pseudopods)
at
the
cell
periphery
(1).
For
example,
in
directionally
migrating
fibro-
blasts,
most
pseudopodial
activity
is
located
at
the
leading
edge
of
the
cell,
whereas
no
protrusional
activity
is
observed
at
the
sides
and
trailing
edge,
resulting
in
the
typical
fan-like
appearance
of
fibroblasts.
The
asymmetric
distribution
of
protrusional
activity
is
a
general
characteristic
of
directional
motility
and/or
directional
growth.
Fish
keratocytes,
which
are
highly
motile,
display
intensive
pseudopodial
activity
at
the
edge
of
the
cell
which
is
oriented
in
the
direction
of
movement
(2),
and
in
neurons,
protrusional
activity
is
re-
stricted
to
the
growth
cone
(3).
In
all
of
these
cases,
pseudopodial
activity
requires
local
polymerization
and
crosslinking
of
actin
filaments
at
the
leading
edge
(4).
Polarization
of
pseudopodial
activity
also
depends
on
other
cytoskeletal
elements.
Vasiliev,
Gelfand,
and
collaborators
(5-7)
showed
that
in
cultured
fibroblasts,
disruption
of
mi-
crotubules
by
drugs
such
as
colchicine
or
nocodazole
re-
sulted
in
the
loss
of
asymmetry
of
cell
shape
and
a
decrease
in
directed
cell
migration.
Two
potential
mechanisms
can
explain
the
role
of
microtubules
in
maintaining
cell
asymme-
try.
First,
microtubules
could
play
a
structural
role
by
mechanically
restricting
the
formation
of
pseudopods
at
the
sides
and
trailing
edge
of
the
cell.
Second,
since
microtubules
provide
the
framework
upon
which
the
molecular
motors
kinesin
and
dynein
vectorially
transport
organelles
and
ves-
icles
(8),
they
may
be
involved
in
transport
of
the
cell
components
necessary
for
formation of
the
leading
edge.
The
observation
that
microinjection
of
fibroblasts
with
anti-
kinesin
antibodies
results
in
loss
of
cell
polarity
(5,
9)
supports
the
latter
possibility.
We
have
examined
the
role
of
the
Golgi
apparatus
in
the
motile
activity
of
cells.
For
this,
we
used
brefeldin
A
(BFA)
(10-12),
which
disrupts
the
Golgi
apparatus,
resulting
in
subsequent
inhibition
of
vesicle
transport
to
the
cell
surface
without
affecting
microtubule
organization
(13).
We
report
that
disruption
of
the
Golgi
apparatus
leads
to
changes
in
cell
morphology
and
motility
that
are
indistinguishable
from
those
caused
by
disruption
of
microtubules.
These
results
suggest
that
the
generation
and
maintenance
of
fibroblast
asymmetry,
which
is
a
prerequisite
for
directed
migration,
depends
on
maintaining
both
the
supply
of
Golgi
apparatus-
derived
vesicles
and
the
integrity
of
microtubule-based
ves-
icle
transport.
MATERIALS
AND
METHODS
Cell
Culture.
Swiss
mouse
3T3
cells
were
cultured
in
Dulbecco's
modified
Eagle's
medium
containing
10o
calf
serum
and
maintained
in
a
water-saturated
atmosphere
of
7%
CO2.
Cells
were
dissociated
with
trypsin/EDTA
and
plated
at
-5
x
103
per
cm2
on
glass
coverslips.
The
effect
of
BFA
on
cell
shape
was
measured
1
or
2
days
after
plating;
BFA
was
prepared
as
a
stock
solution
(10
mg/ml)
in
dimethyl
sulfoxide.
Analysis
of
Cell
Shape
and
Motility.
Dispersion
and
elon-
gation
indices
of
cell
outlines
were
calculated
according
to
Dunn
and
Brown
(14).
Cells
were
labeled
with
tetramethyl-
rhodamine
isothiocyanate
(TRITC)-conjugated
phalloidin
and
microscopic
images
were
captured
with
a
charge-
coupled-device
camera.
Cell
outlines
were
identified
and
analyzed
with
software
provided
by
Z.
Kam,
Department
of
Chemical
Immunology,
Weizmann
Institute
of
Science.
Ap-
proximately
40-50
cells
were
analyzed
for
each
experimental
point.
A
confluent
cell
monolayer
obtained
1
week
after
plating
was
used
for
experiments
on
cell
migration
into
an
experi-
mental
wound.
Experimental
wounds
were
made
in
the
middle
of
the
coverslip
by
using
the
plunger
of
a
disposable
Eppendorf
syringe.
Video
microscopy
of
cell
migration
was
performed
with
a
Plan
Neofluar
x
100/1.3
n.a.
objective
of
a
Zeiss
Axiophot
microscope,
equipped
with
Nomarski
optics,
a
JVC
video
camera,
and
a
Panasonic
video
cassette
re-
corder.
In
experiments
examining
the
effect
of
BFA
pretreatment
on
process
formation,
cells
were
incubated
with
BFA
for
18
hr
prior
to
addition
of
phorbol
12-myristate
13-acetate
(PMA,
100
ng/ml)
for
various
times;
BFA
was
present
throughout
the
incubation
with
PMA.
To
quantify
the
effect
of
BFA
on
the
formation
of
long-processes,
cells
were
labeled
with
Abbreviations:
BFA,
brefeldin
A;
PMA,
phorbol
12-myristate
13-
acetate;
TRITC,
tetramethylrhodamine
isothiocyanate;
C6-NBD-
ceramide,
N-[N-(7-nitro-2,1,3-benzoxadiazol-4-yl)-e-aminohex-
anoyl]-D-erythro-sphingosine.
tTo
whom
reprint
requests
should
be
addressed.
5686
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.
Proc.
Natl.
Acad.
Sci.
USA
91
(1994)
5687
TRITC-conjugated
phalloidin
to
visualize
cell
outlines
and
with
4',6-diamidino-2-phenylindole
(DAPI)
to
visualize
the
nucleus.
A
cell
was
considered
to
have
a
long
process
("tail";
see
Fig.
5C)
if
the
length
of
the
process
was
>10
times
the
diameter
of
the
nucleus.
Immunofluorescence
and
Fluorescence
Microscopy.
Micro-
tubule
distribution
was
examined
by
using
a
monoclonal
anti-a-tubulin
antibody
(clone
DM1A,
Sigma)
and
actin
dis-
tribution
was
examined
by
using
TRITC-conjugated
phalloi-
din.
The
Golgi
apparatus
was
visualized
by
N-[N-(7-nitro-
2,1,
3-benzoxadiazol-4-yl)-
E-aminohexanoyl]-D-erythro-
sphingosine.
(C6-NBD-ceramide)
(15)
after
fixation
in
3%
paraformaldehyde.
RESULTS
BFA
and
Cell
Morphology.
Initial
experiments
demon-
strated
that
incubation
with
low
concentrations
of
BFA
(0.2-1
AM)
resulted
in
dispersion
of
the
Golgi
apparatus,
as
visualized
by
C6-NBD-ceramide,
in
Swiss
3T3
fibroblasts.
After
several
hours
of
incubation
with
BFA,
cells
lost
their
typical
fan-like
shape
with
prominent
lamellipodial
activity
at
the
leading
edge
(Fig.
1A)
and
became
discoid
with
a
random
distribution
of
small
lamellipodia
over
the
entire
cell
periph-
ery
(Fig.
1C).
In
control
cells,
the
Golgi
apparatus
was
prominently
labeled
with
C6-NBD-ceramide
(Fig.
1B)
and,
in
many
cells,
was
positioned
forward
of
the
nucleus
in
the
direction
of
the
cell
edge
which
displayed
pseudopodial
activity
(16).
Disruption
of
the
Golgi
apparatus
by
BFA
(Fig.
1
B
and
D)
occurred
within
<1
hr
of
incubation;
by
this
stage
not
all
cells
had
lost
their
polarized
morphology
(data
not
shown).
However,
no
discernible
Golgi
apparatus
was
ob-
served
in
discoid
cells
after
BFA
treatment
(Fig.
1D).
These
results
indicate
that
disruption
of
the
Golgi
apparatus
by
BFA
preceded
changes
in
cell
morphology.
The
effects
of
BFA
on
cell
morphology
were
compared
with
those
induced
by
the
microtubule-disrupting
agent
no-
codazole
and
quantified
by
using
morphometric
indices
of
dispersion
and
elongation.
These
indices
reflect
the
degree
of
C
FIG.
1.
Effect
of
BFA
on
cell
morphology
and
on
the
integrity
of
the
Golgi
apparatus.
(Left)
Nomarski
images
of
control
(A)
and
BFA-treated
(C)
cells.
(Right)
C6-NBD-ceramide
labeling
of
the
Golgi
apparatus
of
control
(B)
and
BFA-treated
(D)
cells.
(Bar
=
20
gm-)
Removal
of
drug
0.6
-
0.5
0.4
0)
0.
0.1
0.2
0.1
0
0
1
2
4
8
24
Time
(h)
0
0)
0)
c
c
LU
Time
(h)
FIG.
2.
Quantification
of
the
time
course
of
the
morphological
effects
induced
by
BFA
and
nocodazole.
Filled
bars,
0.5
AM
BFA;
hatched
bars,
10
,M
nocodazole.
Values
represent
means
±
SEM.
cell
asymmetry
(14).
Dispersion
and
elongation
indices
de-
creased
during
incubation
with
either
0.5
,uM
BFA
or
10
tLM
nocodazole
(Fig.
2),
indicating
loss
of
polarization
of
cell
shape.
Asymmetry
was
restored
after
removal
of
BFA
or
nocodazole
(Fig.
2),
demonstrating
that
the
effects
of
these
two
drugs
were
reversible.
The
loss
of
asymmetry
was
not
due
to
"rounding"
of
the
cells
(i.e.,
their
partial
detachment
from
the
substratum),
since
BFA-treated
cells,
which
occu-
pied
a
large
area
of
the
substrate,
also
had
a
low
dispersion
index
(data
not
shown)
and,
moreover,
little
change
was
observed
in
the
average
projected
area
of
the
cell.
Thus,
after
8
hr,
the
area
occupied
by
BFA-treated
cells
was
1382
±
104
,um2,
that
occupied
by
nocodazole-treated
cells
was
1265
±
88
1=m2,
and
that
occupied
by
control
cells
was
1539
±
114
Am2.
Prolonged
incubation
with
a
high
concentration
of
BFA
(5
AM)
led
to
a
more
pronounced
decrease
in
cell
area
and
ultimately
to
complete
rounding
of
the
cells.
BFA
and
the
Cytoskeleton.
BFA
had
no
apparent
effect
on
the
radial
distribution
of
microtubules
(Fig.
3).
In
contrast,
significant
changes
in
the
actin
cytoskeleton
were
observed
after
BFA
treatment
(Fig.
4).
In
control
cells,
actin
was
organized
into
two
types
of
structures:
(i)
actin-rich
regions
located
directly
beneath
the
membrane
at
the
leading
edge
and
(ii)
parallel
bundles
of
actin
filaments
oriented
along
the
long
axis
of
the
cell
(Fig.
4A).
After
BFA
treatment,
a
less
pronounced
accumulation
of
actin-rich
areas
at
the
cell
periphery
(Fig.
4
B-D)
was
always
observed,
but
the
distri-
bution
of
actin
bundles
varied,
being
organized
either
cir-
cumferentially
(Fig.
4B)
or
radially
(Fig.
4C)
or
stretching
from
adhesion
points
without
obvious
orientation
to
each
other
(Fig.
4D).
The
effects
of
nocodazole
on
actin
reorga-
nization
(data
not
shown)
were
identical
to
those
of
BFA.
rA
-_
FIG.
3.
Distribution
of
microtubules
in
control
cells
(A)
and
cells
treated
with
BFA
(0.5
AM)
for
18
hr
(B).
BFA
treatment
has
no
effect
on
the
radial
distribution
of
microtubules.
(Bar
=
20
um.)
Cell
Biology:
Bershadsky
and
Futerman
5688
Cell
Biology:
Bershadsky
and
Futerman
cesses
(Fig.
5
B
and
C),
similar
to
the
effect
of
pretreatment
with
Colcemid
(19).
Second,
since
fibroblast
motility
depends
on
pseudopodial
activity,
we
examined
whether
BFA-treated
cells
were
able
to
migrate
into
an
experimental
wound.
The
leading
edge
of
cells
at
the
border
of
an
experimental
wound
is
characterized
by
extension
and
retraction
of
broad
and
flat
lamellipodia
(Fig.
6A),
with
ruffles
containing
a
dense
actin
network
(Fig.
6C).
This
protrusional
and
retractional
activity
results
in
directed
cell
migration
(Fig.
6E).
After
BFA
treatment,
cells
showed
little
protrusional
and
retractional
activity
(Fig.
6
B
BI
i)
NE.
.
A
I
I
lo
FIG.
4.
BFA
causes
reorganization
of
the
actin
cytoskeleton.
(A)
Control
cells.
(B-D)
Cells
treated
with
BFA
(0.5
pM)
for
18
hr.
Note
the
changes
in
the
organization
of
actin
bundles
and
the
loss
of
polarized
distribution
of
actin-rich
lamellipodia
after
BFA
treatment.
(Bar
=
20
am.)
BFA
Inhibits
Protrusional
Activity
and
Cell
Motility.
Two
experimental
paradigms
were
used
to
examine
the
effects
of
BFA
on
pseudopodial
activity.
First,
it
is
known
that
the
pseudopodial
activity
of
migrating
fibroblasts
can
be
en-
hanced
by
activation
of
protein
kinase
C.
Thus,
phorbol
esters
induce
changes
in
morphology
and
in
organization
of
the
actin
cytoskeleton
(17,
18),
resulting
in
accumulation
of
actin
at
the
leading
edge,
destruction
of
large
actin
bundles,
and
formation
of
long
processes
(Fig.
SA).
Pretreatment
with
BFA
prevented
the
PMA-induced
formation
of
long
pro-
60
50'
40
UT
-T
ki)
CL
C
I
I
30
20
10;
0
Non-
PMA
PMA
treated
r1i
hi
i5
ii
FiG.
5.
BFA
prevents
the
ability
of
PMA
to
stimulate
lamellipo-
dial
activity
and
formation
of
long
processes.
After
PMA
treatment
(1.5
hr),
Swiss
3T3
cells
extend
long
processes
(A),
but
preincubation
(18
hr)
with
BFA
(1
tM)
(B)
prevents
this
effect.
(Bar
=
50
PM.)
(C)
Quantification
of
the
effect
of
pretreatment
with
BFA
(hatched
bars)
on
PMA-induced
formation
of
long
processes.
Filled
bars,
control
cells.
7
f
4
4¢J
1'
F
.I
I
.0
a
-
.
2
FIG.
6.
Effect
of
BFA
on
directional
cell
migration.
(A
and
B)
Lamellipodial
activity
at
the
edge
of
an
experimental
wound
was
examined
by
video
microscopy.
Outlines
of
the
leading
edge
were
traced
from
five
successive
images
captured
every
25
sec.
The
difference
in
superimposed
outlines
demonstrates
the
extent
of
protrusional
and
retractional
activity.
The
amplitude
of
these
activ-
ities
is
much
higher
in
control
cells
than
in
BFA-treated
cells.
In
addition,
pseudopodial
activity
is
restricted
to
a
small
area
of
the
cell
periphery
in
BFA-treated
cells
but
is
more
evenly
distributed
in
control
cells.
(Bar
=
20
pm.)
(C
and
D)
Actin
distribution
in
cells
at
the
edge
of
an
experimental
wound.
Arrows
indicate
accumulation
of
actin
at
the
leading
edge
of
control
cells.
After
BFA
treatment,
there
are
few
actin-rich
regions
at
the
cell
periphery,
and
actin
is
concen-
trated
mainly
in
stress
fibers.
(Bar
=
20
pm.)
(E
and
F)
An
experimental
wound
of
0.5
mm
is
completely
filled
with
cells
in
control
cultures
8
hr
after
wounding
but
devoid
of
cells
after
BFA
treatment.
(Bar
=
500
gm.)
Proc.
Natl.
Acad
Sci.
USA
91
(1994)
Proc.
Natl.
Acad.
Sci.
USA
91
(1994)
5689
and
D),
and
consequently
the
rate
of
cell
migration
into
the
experimental
wound
was
significantly
reduced
(Fig.
6F).
The
effects
of
BFA
on
cell
motility
were
similar
to
those
observed
with
microtubule-depolymerizing
agents
(20).
DISCUSSION
In
the
current
report,
we
demonstrate
that
incubation
of
Swiss
3T3
fibroblasts
with
low
concentrations
of
BFA
(0.2-1
,uM)
leads
to loss
of
polarization
of
cell
shape
and
pseudopo-
dial
activity.
As
a
result,
pronounced
reorganization
of
the
actin
cytoskeleton
occurs,
and
consequently
cells
are
unable
to
form
long
processes
and
to
migrate
into
an
experimental
wound.
The
effects
of
BFA
on
cell
motility
are
strikingly
similar
to
those
obtained
using
microtubule-disrupting
agents
(5-7,
19,
20)
(see
also
Fig.
2).
The
necessity
of
a
centralized
microtubule
system
in
maintaining
cell
polarity
appears
to
depend
on
cell
type
(21).
In
fibroblasts,
disruption
of
microtubules
by
drugs
such
as
colchicine
or
nocodazole
results
in
loss
of
asymmetry
of
cell
shape
and
a
decrease
in
directed
cell
migration
(5-7).
In
addition,
blocking
vesicle
transport
along
microtubules
with
anti-kinesin
antibodies
(9)
results
in
loss
of
cell
polarity.
Even
though
the
effects
of
BFA
on
cell
morphology
and
motility
are
almost
identical
to
those
observed
with
micro-
tubule-disrupting
agents,
the
cellular
target
of
action
of
these
drugs
is
completely
different.
Thus,
we
observed
no
effect
of
BFA
on
microtubule
integrity,
consistent
with
published
studies
(13),
and
preliminary
observations
from
video
mi-
croscopy
suggest
that
no
change
occurs
in
the
motility
of
mitochondria,
supporting
the
idea
that
BFA
has
no
effect
on
microtubule-based
motor
activity.
In
contrast,
it
is
now
well
established
that
BFA
interacts
with
the
cellular
machinery
responsible
for
vesicle
budding
(22,
23),
disrupting
both
the
integrity
of
organelles
of
the
central
vacuolar
system,
and
the
transport
of
vesicles
between
these
organelles
(10-12).
In
particular,
BFA
disrupts
the
Golgi
apparatus
and
inhibits
vesicle
transport
from
distal
Golgi
apparatus
compartments
to
the
cell
surface
(24).
The
similarities
in
the
effects
on
cell
organization
and
motility
of
BFA,
of
microtubule-disrupting
agents,
and
of
inhibition
of
the
molecular
motor
kinesin
(9,
25)
may
therefore
result
from
disruption
of
the
supply
of
intracellular
vesicles
containing
the
components
necessary
to
form
and
maintain
leading-edge
activity.
Thus,
BFA
affects
the
availability
of
vesicles,
inhibition
of
the
activity
of
microtubule
motors
affects
the
transport
of
vesicles,
and
disruption
of
microtubules
affects
the
molecular
framework
upon
which
vesicles
are
transported.
It
has
been
proposed
that
the
polarized
delivery
of
vesicles
derived
from
the
Golgi
apparatus
and
their
subsequent
in-
sertion
at
the
plasma
membrane
provide
a
basis
for
polarized
lamellipodial
activity
at
the
leading
edge
(26).
This
suggestion
was
based
on
experiments
in
which
the
newly
synthesized
G
protein
of
vesicular
stomatitis
virus
was
observed
to
be
inserted
initially
at
the
leading
edge
of
motile
fibroblasts
(27,
28).
However,
direct
proof
of
this
proposal
has
been
lacking
due
to
the
unavailability
of
experimental
data
directly
char-
acterizing
the
role
of
the
Golgi
apparatus
and
vesicle
trans-
port
in
the
regulation
of
cell
shape
and
motility.
Our
current
data
support
this
proposal
by
demonstrating
that
disruption
of
the
Golgi
apparatus
inhibits
lamellipodial
activity;
simi-
larly,
cultured
gonocytes
(29)
are
unable
to
develop
processes
after
BFA
treatment,
and
axonal
growth
is
inhibited
in
cultured
hippocampal
neurons
(30).
Together,
these
results
demonstrate
multiple
ways
of
regulating
pseudopodial
activ-
ity
and
cell
migration,
either
by
modifying
the
cytoskeletal
components
or
the
molecular
motors
involved
in
vesicle
transport
or
by
modifying
the
supply
of
the
vesicles
them-
selves.
Modifying
and
regulating
the
cellular
machinery
that
is
responsible
for
vesicle
fusion
and/or
budding
may
provide
an
alternative
means
for
controlling
cell
motility
and
cyto-
skeletal
structure.
We
thank
Benny
Geiger
for
support
and
encouragement,
Natalie
Prigozhina
for
help
in
measurement
of
cell
shape,
and
Zvi
Kam
for
providing
software.
This
work
was
supported
by
the
Raschi
Foun-
dation
(to
A.D.B.).
A.H.F.
is
the
incumbent
of
the
Recanati
Career
Development
Chair
in
Cancer
Research.
1.
Bray,
D.
(1992)
Cell
Movements
(Garland,
New
York).
2.
Lee,
J.,
Ishihara,
A.
&
Jacobson,
K.
(1993)
Trends
Cell.
Biol.
3,
366-370.
3.
Bray,
D.
(1992)
in
The
Nerve
Growth
Cone,
eds.
Letourneau,
P.
C.,
Kater,
S.
B.
&
Macagno,
E.
R.
(Raven,
New
York),
pp.
7-17.
4.
Stossel,
T.
P.
(1993)
Science
260,
1086-1094.
5.
Bershadsky,
A.
D.
&
Vasiliev,
J.
M.
(1993)
in
Cell
Behavior:
Adhesion
and
Motility,
eds.
Jones,
G.,
Wigley,
C.
&
Warn,
R.
(Company
Biol.,
Cambridge,
U.K.),
pp.
353-373.
6.
Vasiliev,
J.
M.,
Gelfand,
I.
M.,
Domnina,
L.
V.,
Ivanova,
0.
Y.,
Komm,
S.
G.
&
Olshevskaja,
L.
V.
(1970)
J.
Embryol.
Exp.
Morphol.
24,
625-640.
7.
Vasiliev,
J.
M.
&
Gelfand,
1.
M.
(1976)
in
Cell
Motility,
eds.
Goldman,
R.,
Pollard,
T.
&
Rosenbaum,
J.
(Cold
Spring
Harbor
Lab.
Press,
Plainview,
NY),
pp.
279-304.
8.
Walker,
R.
A.
&
Sheetz,
M.
P.
(1993)
Annu.
Rev.
Biochem.
62,
429-451.
9.
Rodionov,
V.
I.,
Gyoeva,
F.
K.,
Tanaka,
E.,
Bershadsky,
A.
D.,
Vasiliev,
J.
M.
&
Gelfand,
V.
I.
(1993)
J.
Cell
Biol.
123,
1811-1820.
10.
Hurtley,
S.
M.
(1992)
Trends
Biochem.
Sci.
17,
325-327.
11.
Klausner,
R.
D.,
Donaldson,
J.
G.
&
Lippincott-Schwartz,
J.
(1992)
J.
Cell
Biol.
116,
1071-1080.
12.
Pelham,
H.
(1991)
Cell
67,
449-451.
13.
Burgess,
T.
L.,
Skoufias,
D.
A.
&
Wilson,
L.
(1991)
Cell
Motil.
Cytoskel.
20,
289-300.
14.
Dunn,
G.
A.
&
Brown,
F.
(1986)
J.
Cell
Sci.
83,
313-340.
15.
Pagano,
R.
E.,
Sepanski,
M.
&
Martin,
0.
C.
(1989)
J.
Cell
Biol.
109,
2067-2079.
16.
Kupfer,
A.,
Louvard,
D.
&
Singer,
S.
J.
(1982)
Proc.
Natl.
Acad.
Sci.
USA
79,
2603-2607.
17.
Schliwa,
M.,
Nakamura,
T.,
Porter,
K.
R.
&
Euteneuer,
U.
(1984)
J.
Cell
Biol.
99,
1045-1059.
18.
Dugina,
V.
B.,
Svitkina,
T.
M.,
Vasiliev,
J.
M.
&
Gelfand,
I.
M.
(1987)
Proc.
Natl.
Acad.
Sci.
USA
84,
4122-4125.
19.
Lyass,
L.
A.,
Bershadsky,
A.
D.,
Vasiliev,
J.
M.
&
Gelfand,
I.
M.
(1988)
Proc.
Natl.
Acad.
Sci.
USA
85,
9538-9541.
20.
Bershadsky,
A.
D.,
Vaisberg,
E.
A.
&
Vasiliev,
J.
M.
(1991)
Cell
Motil.
Cytoskel.
19,
152-158.
21.
Schliwa,
M.
&
Honer,
B.
(1993)
Trends
Cell.
Biol.
3,
377-380.
22.
Helms,
J.
B.
&
Rothman,
J.
E.
(1992)
Nature
(London)
360,
352-354.
23.
Donaldson,
J.
G.,
Finazzi,
D.
&
Klausner,
R.
D.
(1992)
Nature
(London)
360,
350-352.
24.
Miller,
S.
G.,
Carnell,
L.
&
Moore,
H.
H.
(1992)
J.
Cell
Biol.
118,
267-283.
25.
Ferreira,
A.,
Niclas,
J.,
Vale,
R.
D.,
Banker,
G.
&
Kosik,
K.
S.
(1992)
J.
Cell
Biol.
117,
595-606.
26.
Singer,
S.
J.
&
Kupfer,
A.
(1986)
Annu.
Rev.
Cell
Biol.
2,
337-365.
27.
Bergmann,
J.
E.,
Kupfer,
A.
&
Singer,
S.
J.
(1983)
Proc.
Natl.
Acad.
Sci.
USA
80,
1367-1371.
28.
Kupfer,
A.,
Kronebusch,
P.
J.,
Rose,
J.
K.
&
Singer,
S.
J.
(1987)
Cell.
Motil.
Cytoskel.
8,
182-189.
29.
McGuinness,
M.
P.
&
Orth,
J.
M.
(1992)
Eur.
J.
Cell
Biol.
59,
196-210.
30.
Jareb,
M.
&
Banker,
G.
(1991)
Soc.
Neurosci.
Abstr.
17,
730.
Cell
Biology:
Bershadsky
and
Futerman
