Variable actin dynamics requirement for the exit of different cargo from the 𝘵𝘳𝘢𝘯𝘴-Golgi network

Abstract

Efficient post-Golgi trafficking depends on microtubules, but actin filaments and actin-associated proteins are also postulated. Here we examined, by inverse fluorescence recovery after photobleaching, the role of actin dynamics in the exit from the TGN of fluorescent-tagged apical or basolateral and raft or non-raft-associated cargoes. Either the actin-stabilizing jasplakinolide or the actin-depolymerising latrunculin B variably but significantly inhibited post-Golgi traffic of non-raft associated apical p75NTR and basolateral VSV-G cargoes. The TGN-exit of the apical-destined VSV-G mutant was impaired only by latrunculin B. Strikingly, the raft-associated GPI-anchor protein was not affected by either actin toxin. Results indicate that actin dynamics participates in the TGN egress of both apical- and basolateral-targeted proteins but is not needed for apical raft-associated cargo.

Full text

Variable actin dynamics requirement for the exit of different cargo
from the trans-Golgi network
Francisco La
´zaro-Die
´guez
a,b,c,1
, Cecilia Colonna
a,1
, Miguel Cortegano
a
, Marı
´a Calvo
d
,
Susana E. Martı
´nez
a,b
, Gustavo Egea
a,b,c,*
a
Departament de Biologia CelÆlular i Anatomia Patolo
`gica, Facultat de Medicina, Universitat de Barcelona, C/Casanova 143, E-08036 Barcelona, Spain
b
Institut d’Investigacions Biome
`diques August Pi i Sunyer (IDIBAPS), Universitat de Barcelona, C/Casanova 143, E-08036 Barcelona, Spain
c
Institut de Nanocie
`ncia i Nanotecnologia (IN
2
UB), Universitat de Barcelona, 08036 Barcelona, Spain
d
Serveis Cientifico-Te
`cnics (SCT-UB), Universitat de Barcelona, 08036 Barcelona, Spain
Received 25 April 2007; revised 6 July 2007; accepted 8 July 2007
Available online 17 July 2007
Edited by Felix Wieland
Abstract Efficient post-Golgi trafficking depends on microtu-
bules, but actin filaments and actin-associated proteins are also
postulated. Here we examined, by inverse fluorescence recovery
after photobleaching, the role of actin dynamics in the exit from
the TGN of fluorescent-tagged apical or basolateral and raft or
non-raft-associated cargoes. Either the actin-stabilizing jasplak-
inolide or the actin-depolymerising latrunculin B variably but
significantly inhibited post-Golgi traffic of non-raft associated
apical p75NTR and basolateral VSV-G cargoes. The TGN-exit
of the apical-destined VSV-G mutant was impaired only by
latrunculin B. Strikingly, the raft-associated GPI-anchor protein
was not affected by either actin toxin. Results indicate that actin
dynamics participates in the TGN egress of both apical- and
basolateral-targeted proteins but is not needed for apical raft-
associated cargo.

2007 Federation of European Biochemical Societies. Published
by Elsevier B.V. All rights reserved.
Keywords: Actin; Cytoskeleton; Golgi apparatus; Raft;
Polarized transport; iFRAP
1. Introduction
The trans-Golgi network (TGN) is defined as the major sort-
ing compartment of the secretory pathway of synthesized pro-
teins upon recognition of specific apical or basolateral sorting
signals [1–3]. Tightly associated with the TGN is the cytoskel-
eton, in particular microtubules [4], whereas actin filaments
have also recently attracted increasing attention [5,6]. Thus,
a variety of actin-regulatory/binding proteins have been impli-
cated in post-Golgi trafficking such as Arp2/3, spir1, myosins
II and VI, cortactin, Cdc42, huntingtin related protein 1 (see
[5] and references herein), and huntingtin [7]. In addition, the
role of actin filaments has also been tested in the transport
from the Golgi to the plasma membrane [8–10] and to early/
late endosomes [11–13] with a variety of results in some cases.
This might be attributable to the use of only one actin-dep-
olymerizing agent (usually cytochalasins) or to utilize different
cargoes or both. In any case, the substantial homology be-
tween the protein-sequestration and sorting machinery acting
at the plasma membrane and the TGN [6,14] points to a signif-
icant role of actin filaments or actin dynamics (actin depoly-
merization/polymerization cycle) at the TGN as occurs at the
plasma membrane, but it remains to be established whether ac-
tin acts at the level of cargo sorting, in membrane budding/
scission, or in the subsequent locomotion of TGN-derived
transport carriers or both [5]. As a first approach to this end,
we here explore how actin dynamics participates in the exit
from the TGN of a variety of cargo proteins using in vivo in-
verse fluorescence recovery after photobleaching (iFRAP)
analysis. This technique allows analysis of the kinetic parame-
ters of the exit of a fluorescently-tagged cargo protein from a
subcellular compartment (the Golgi stack/TGN in our case).
Furthermore, our study is focused on the potential role of
actin dynamics in post-Golgi egress regardless of the intrinsic
molecular mechanisms that determine the sorting and subse-
quent polarized transport of these cargoes. Thus, basolateral
sorting is mediated by peptide sequences containing tyrosine-
and dileucine motifs [15]. In contrast, apical sorting is facili-
tated by one of two mechanisms: either by lipid-lipid and
lipid-protein interactions within the transmembrane or luminal
domains, given by N- and O-glycans [16], or by such interac-
tions within the glycosylphosphatidylinositol (GPI) anchor do-
main, which is invariably associated with lipid rafts [17,18].
Recent lines of evidence indicate that oligomerization of mem-
brane proteins at the TGN is an important sorting determinant
for apical transport of raft-associated proteins [19]. Therefore,
apical sorting and transport can be mediated by raft-depen-
dent and raft-independent mechanisms. It is important to high-
light that both polarized and non-polarized epithelial cells
share molecular mechanisms in the sorting of secretory cargo,
the formation of transport carriers and their subsequent trans-
port to the plasma membrane. This validates the utilization of
non-polarized cell lines, for instance COS-1 or non-polarized
MDCK cells [20,21] to address the aforementioned questions.
In this respect, non-epithelial cells are capable of packaging
and transporting secretory cargo selectively in distinct types
of Golgi-derived carriers like fully-polarized cells do [22].
In this study we examined how actin toxins that depolyme-
rize (latrunculin B/LtB) or stabilize (jasplakinolide/Jpk) actin
filaments impair the TGN-exit of non-raft-associated apical
*
Corresponding author. Address: Departament de Biologia CelÆlular i
Anatomia Patolo
`gica, Facultat de Medicina, Universitat de Barcelona,
C/Casanova 143, E-08036 Barcelona, Spain. Fax: +34 93 4021907.
E-mail address: gegea@ub.edu (G. Egea).
1
These authors equally contributed to this study.
0014-5793/$32.00 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.febslet.2007.07.015
FEBS Letters 581 (2007) 3875–3881

(p75NTR)- or basolateral (VSV-G)-targeted proteins. At the
same time, we also examined whether actin dynamics is needed
for the exit of raft-associated GPI from the TGN. Lipid rafts
are membrane platforms enriched in sphingomyelin, choles-
terol and phosphatidylinositol 4,5-biphosphate (PIP
2
). The lat-
ter has a tight relationship with the actin cytoskeleton since it
promotes actin assembly into filaments [23]. We observed that
actin dynamics participates in the exit of both non-raft-associ-
ated apical and basolateral-targeted cargo, but it was not
needed for the TGN exit of raft-anchored GPI. This indicates
that actin does not functionally associate with raft-associated
GPI anchored proteins at the TGN.
2. Materials and methods
2.1. Cell culture and reagents
COS-1 cells were grown at 37 C in complete DMEM culture med-
ium, which consisted of DMEM (Gibco BRL, Eggestein, Germany)
supplemented with penicillin (100 i.u./ml), streptomycin (100 mg/ml),
LL
-glutamine (2 mM), MEM sodium pyruvate (1 mM) and fetal calf
serum (10%). Latrunculin B was purchased from Calbiochem (EMD
Biosiences, Inc., darnstadt, Germany) and Jasplakinolide from Molec-
ular probes (Eugene, OR, USA). Monoclonal antibodies to giantin were
kindly provided by H.P. Hauri (Biozentrum, Basel University) and
those to TGN46 were purchased from Serotec (Oxford, UK). Second-
ary Alexa Fluor 546 donkey anti-sheep IgG (H + L) and Alexa Fluor
647 goat anti-mouse was from Invitrogen (Carlsbad, CA, USA).
2.2. Plasmids
GFP-tagged wild type and apical-targeted GFP-VSV-G mutant
cDNAS, as well as the CFP-tagged VSV-G wild type cDNA were
kindly provided by Kai Simons (Max Planck Institute of Molecular
Cell Biology and Genetics, Dresden, Germany), and GFP-p75NTR
and YFP-GPI cDNAs were kindly supplied by Roman Polishchuk
(CMNS, Chieti, Italy).
2.3. Transient transfection
COS-1 cells were grown on coverslips in a 35 mm Petri dish and
transiently transfected with the corresponding cDNA using Effectene,
in accordance with the manufacture’s instructions (Qiagen) for 16 h at
37 C in complete DMEM.
2.4. Actin toxins treatments and inversal FRAP experiments
To synchronize post-Golgi traffic at the TGN, transfected cells were
incubated at 19.5 C for 2 h in the presence of cycloheximide (100 lm/ml).
Cells were subsequently treated for 20 min at 19.5 C with LtB or Jpk
(500 nM each) (see the experimental scheme in Fig. 1A).
Inverse fluorescence recovery after photobleaching (iFRAP) experi-
ments were carried out using a Leica TCS SL laser scanning confocal
spectral microscope (Leica Microsystems Heidelberg GmbH, Man-
heim, Germany) with Argon and HeNe lasers attached to a Leica
DMIRE2 inverted microscope equipped with an incubation system
with temperature and CO
2
control. All experiments were performed
at 32 C and 5% CO
2
. For visualization of GFP/YFP, images were ac-
quired using a 63·oil immersion objective lens (NA 1.32), 488 nm laser
line, excitation beam splitter RSP 500, emission range detection: 500–
600 nm with the confocal pinhole set at 4.94 Airy units to minimize
changes in fluorescence due to protein-GFP moving away from the
plane of focus. The whole cytoplasm, except the Golgi, of GFP-fusion
protein transfected cell was photobleached using 50–80 scans with the
488 nm laser line at full power. Postbleach images were monitored at
5 s intervals for 15 min. The excitation intensity was attenuated to
approximately 5% of the half laser power to avoid significant photoble-
aching. Longer recording times were also performed to check the arri-
val of cargo at the plasma membrane and the subsequent absence of
fluorescence in the Golgi.
2.5. Inverse FRAP analysis
To evaluate the results of photobleaching experiments, the observed
fluorescence equilibration in the unbleached region (the Golgi com-
plex) was quantified using the Image Processing Leica Confocal Soft-
ware. Background fluorescence was measured in a random field
outside of cells. All experiments were background substracted, cor-
rected and normalized using the equation described below. For each
time point the relative loss of total fluorescent intensity in the
unbleached region of interest was calculated as:
Irel ¼ðItÞ=ðIoÞðTmean =TtÞ
where I
t
is the average intensity of the unbleached region of interest at
time point t,I
o
is the average pre-bleach intensity of the region of inter-
est and T
mean
and T
t
are the total mean cell intensity of the whole post-
bleach period and average total cell intensity at each time of
postbleach, respectively. Fitting of iFRAP curves was performed with
Graphpad Prism Software v.3.0 (Graphpad Software, San Diego, CA)
and modelled assuming two-phase exponential decay iFRAP, whereas
they were equally well modelled with the one-phase exponential decay
equation:
Yðfluorescence decayÞ¼Span expðKXÞþPlateau
where Ystarted at span + plateau and decayed to plateau with a rate
constant K. Half-time was calculated as: 0.69/K. Afterwards, data plot-
ted as fluorescence intensity that remained in the Golgi vs. time.
Mobile fraction (MF) was calculated as:
MF ¼ðFpre FendÞ=Fpre
where F
pre
was the initial fluorescence intensity and F
end
the final
recovered fluorescence intensity. Statistical analysis was performed
by non-parametric ANOVA post-test Bonferroni.
2.6. Immunofluorescence
Indirect immunofluorescence was carried out as previously described
[7] with the following antibody dilutions: anti-giantin, 1:500 and anti-
TGN46, 1:100. Secondary antibodies were used at 1:250.
3. Results and discussion
We analyzed the fluorescence decay curves of a variety of
green or yellow fluorescent protein (GFP/YFP)-tagged cargoes
from the trans-Golgi network (TGN) by iFRAPs in single liv-
ing mammalian cells. This technique allows us to measure the
time of association of the different GFP/YFP-tagged cargoes
within the TGN, and their dissociation is then reflected by
the loss of fluorescence from this region over time monitored
by confocal microscopy.
We performed experiments for three types of GFP/YFP-
tagged proteins: either raft-associated and apical-targeted such
as GPI, or non-raft-associated and basolateral-targeted like
VSV-G protein, or apical-targeted such as p75NTR and
VSV-G mutant. All these proteins have been reported in api-
cal- and basolateral-like transport carriers in non-polarized
[24] and fully-polarized MDCK cells [2,25–28], Vero, BHK
and CHO cell lines [22,29]. Taking this into account, we uti-
lized COS-1 cells (a non-polarized cell line), which are highly
suitable to be transfected, in contrast to MDCK cells which
usually require microinjection. For quantitative analysis, we
chose those cells that express relatively low levels of GFP/
YFP-tagged proteins to overcome potential toxicity of overex-
pression and, furthermore, those with identical Golgi-associ-
ated basal fluorescence to render the data comparable.
We analyzed the kinetics of the TGN-exit of membrane car-
go by plotting Golgi fluorescence intensity over time and by
determining, for each condition, the mobile fraction, which is
defined as the amount of cargo that has left the Golgi after
15 min at 32 C and the halftime point (T
1/2
), which is defined
as the time at which 50% of the mobile fraction has been lost
(Fig. 1A). COS-1 cells were then transfected with the corre-
3876 F. La
´zaro-Die
´guez et al. / FEBS Letters 581 (2007) 3875–3881

sponding plasmids and left overnight at 37 C. Before the
in vivo recording, cargo was allowed to accumulate at the Gol-
gi/TGN by culturing cells at 19.5 C for 2 h in the presence of
cycloheximide. Cells were treated with the corresponding actin
toxin (LtB or Jpk) for the last 20 min at 19.5 C. We confirmed
that LtB and Jpk significantly disrupted the actin cytoskeleton
at this time and temperature (data not shown). Subsequently,
cells were placed in the confocal chamber at 32 C and iFRAP
was carried out (Fig. 1A). When transfected cells were cultured
at 19.5 C, GFP/YFP-tagged proteins invariably accumu-
lated in the Golgi, which was distinguishable because of the
high fluorescence accumulated in a peri-nuclear structure
(Fig. 1B) and by double-labeling experiments using well-estab-
lished Golgi stack and TGN markers such as giantin and
Fig. 1. (A) Scheme of the experimental design used in this study. (B) Representative frames obtained from time-lapse iFRAP confocal microscopy
experiments in transiently transfected COS-1 cells expressing wild-type basolateral GFP-VSV-G. Note that in control cells, the Golgi-associated
fluorescence of GFP-VSV-G decays over time, whereas in cells treated with the actin-stabilizing toxin jasplakinolide (Jpk; 500 mM) the fluorescence
remains virtually unaltered. (C) Colocalization experiments in COS-1 cells of GFP-tagged cargoes used in this study with the Golgi stack marker
giantin and the TGN marker TGN-46 in COS-1 cells incubated at 19.5 C in the presence of cycloheximide. Bar, 10 lm.
F. La
´zaro-Die
´guez et al. / FEBS Letters 581 (2007) 3875–3881 3877

TGN46, respectively (Fig. 1C). Moreover, reticular and punc-
tate cytoplasmic fluorescence pools were also observed, which
corresponded to proteins located in the endoplasmic reticulum
and transport carriers in transit to the Golgi or the plasma
membrane, respectively. A characteristic morphological visual-
ization of an iFRAP experiment for untreated (control) and
Jpk-treated COS-1 cells expressing GFP-VSV-G is shown in
Fig. 1B. In contrast to the control cell, the Golgi-associated
fluorescence corresponding to VSV-G was virtually unaltered
over time as a consequence of Jpk treatment. The kinetics of
Golgi release followed by VSV-G showed a rapid decrease
(Figs. 1B and 2A, red), indicating that protein was released
up to 56% with a T
1/2
of 3.0 min in untreated cells (Fig. 2A,
red; Table 1). In the presence of Jpk, VSV-G release from
the Golgi was robustly inhibited (Fig. 1B), resulting in a fall
of the mobile fraction after 15 min which was only of 10%
(Fig. 2A, blue; Table 1). When cells were treated with the ac-
tin-depolymerizing toxin LtB (Fig. 2A, green), the mobile frac-
tion was also significantly reduced to 27% (T
1/2
of 5.7 min)
but to a lesser extent than Jpk treatment (Fig. 2A; Table 1).
These results indicate that VSV-G exit from the TGN is signif-
icantly reduced by either the stabilization or depolymerization
of actin filaments, which suggests that actin dynamics (for
example, a rapid actin turnover coupled to the formation of
the transport carrier [5]) is necessary for its post-Golgi trans-
port.
In order to test whether this actin dynamics-dependence was
in turn dependent on the intrinsic basolateral destination of
the VSV-G, we used an apical-destined VSV-G mutant in
which the tyrosine-based motif is masked [30]. We reasoned
that since the post-Golgi exit of VSV-G involving actin
dynamics is also signal-mediated, we would expect differences
in the TGN exit kinetic parameters in the apical-targeted mu-
tant in comparison to the basolateral wild-type form. We ob-
served that the apical VSV-G mutant (Fig. 2B, red) showed
a lower mobile fraction than that of the basolateral VSV-G
(36% and 56%, respectively; Table 1) whereas the T
1/2
was practically the same (3.2 min and 3.0 min, respectively;
Table 1). Unexpectedly, the apical VSV-G mutant was only
sensitive to LtB (Fig. 2B, green), giving rise to a significant
reduction of the mobile fraction (25%; Table 1) and a larger
T
1/2
(6.2 min). This indicates that cargo transport was slowed-
down but, as occurred in control cells, the T
1/2
of LtB-treated
cells expressing either wild-type or mutant VSV-G displayed
no significant difference (5.7 min and 6.2 min, respectively;
Table 1). In contrast, Jpk-treated cells showed no changes in
the shape of the decay curve, the mobile fraction, or the T
1/2
in comparison to control cells (Fig. 2B, blue; Table 1).
The delay in the VSVG-GFP egress from the Golgi in LtB-
treated cells implies the involvement of actin in protein export
from the TGN. This result is consistent with previous observa-
tions in which VSV-G exit from the TGN was lower in cells
treated with cytochalasin-B than in untreated cells [8]. These
authors also reported morphological changes in cargo-contain-
ing tubules, which suggested that actin and actin-based cyto-
skeleton components are involved in the detachment of
TGN-derived transport carriers. The arrival of VSV-G at the
basolateral plasma membrane also decreased in LtB-treated
non-polarized MDCK cells [31]. Several groups have found
that interfering with interactions between dynamin II and syn-
dapin II in living cells caused an inhibition of VSV-G exit from
the TGN [32], and the overexpression of a dominant-interfer-
ing mutant of actin nucleator Spir-1 also strongly inhibited
post-Golgi VSV-G transport to the plasma membrane [33].
Not only actin but also actin regulators such as Cdc42 partic-
ipate in the appropriate sorting at the TGN. Thus, VSV-G and
LDL receptor were missorted to the apical membrane in cells
expressing mutants of Cdc42 [31]. Therefore, our results indi-
cate that actin depolymerization/polymerization cycle controls
the exit of basolateral VSV-G protein from the TGN rather
than the presence of filamentous actin alone, whose involve-
ment in apical VSV-G mutant transport seems to be more di-
rect according to the inhibitory effects caused only by LtB.
Next, we tested the effect of both actin toxins on post-Golgi
traffic of membrane protein p75 neurotrophin receptor
(p75NTR) coupled to GFP [9,31]. In stably transfected polar-
ized and non-polarized MDCK cells, p75NTR is a non-raft-
associated protein that is exclusively transported to the apical
plasma membrane via apical sorting signals contained within
the O-glycosylated stalk [9,27]. As shown in Fig. 3A, in control
cells the decay curve shape for GFP-p75NTR was similar to
those observed for both VSV-Gs. However, the mobile frac-
tion (46%) was greater than in the apical VSV-G mutant
Fig. 2. Inverse FRAP decay curves of basolateral GFP-VSV-G (A)
and apical GFP-VSV-G mutant (B) in COS-1 cells. Results are the
means ± S.E.M. from ncells indicated in Table 1.
3878 F. La
´zaro-Die
´guez et al. / FEBS Letters 581 (2007) 3875–3881

(36%) but lower than that obtained for the wild-type basolat-
eral VSV-G (56%) (Table 1). The exit of p75NTR from the
Golgi was significantly inhibited when cells were treated with
Jpk or LtB (Fig. 3A), which resulted in a significant decrease
in their respective protein mobile fractions (28% and
27%, respectively) (Table 1). These results are inconsistent
with those obtained with the apical VSV-G mutant protein,
which is also non-raft-associated, since their respective sensi-
tivity to Jpk appears to be different. In contrast, both are
equally sensitive to LtB, whose use and sensitivity for
p75NTR was previously reported [9]. Note that the decay
curve for p75NTR during the first 3 min at 32 C was the same
in both control and treated cells, and that the differences
caused by actin toxins appeared later (Fig. 3A). This behavior
indicates that actin dynamics is not required for molecular pro-
cesses that lead to the immediate egress of cargo from the
TGN (for instance, the formation of the transport carrier).
In other words, actin toxins do not appear to alter the initial
mobile fraction. In contrast, actin dynamics might be neces-
sary for earlier molecular processes (for instance, cargo sort-
ing) and any alterations would then affect the immobile
fraction. Previous results showed that p75NTR transport to
the membrane is altered by expression of Cdc42 mutants,
but both constitutive active and inactive mutants differentially
altered the exit of apical and basolateral proteins from the
TGN in vitro [31]. The sensitivity to Jpk treatment could be
a particularity of p75NTR, suggesting that actin cytoskeleton
dynamics would mediate different effects depending on the api-
cally-destined cargo examined. p75NTR contains three dileu-
cine motifs that might represent a type of basolateral sorting
signal [34], but that are present on naturally occurring mem-
brane proteins targeted basolaterally because they lack apical
sorting information.
Finally, an unconventional (non-sorting-receptor mediated)
mechanism for apical sorting is the association with lipid rafts
[35]. A role for rafts in protein sorting was first described for
apical sorting of GPI-anchored proteins in polarized epithelial
cells, where these proteins associate with detergent-resistant
microdomains (DRMs) during their passage through the Golgi
apparatus [36]. Therefore, to explore whether the actin dynam-
ics actively participates in post-Golgi traffic of raft-associated
proteins, we examined the raft-associated glycosylphosphat-
idylinositol (GPI)-anchored protein tagged to YFP, which is
sorted to the apical surface [26,29,37]. The release of YFP-
GPI from the Golgi in untreated COS cells showed a similar
decay curve to those observed for other cargoes but, notably,
after the release from the 19.5 C blockage (Fig. 3B) and con-
versely to the other examined apical-targeted proteins, GPI
was totally unaffected by actin toxins (Fig. 3B; Table 1). More-
over, the final YFP-GPI subcellular destination was the plas-
ma membrane and equally occurred both in treated and
untreated cells (not shown). In fact, all this was unexpected
taking into account the attributed role of lipid rafts and actin
cytoskeleton polymerization occurring in the plasma mem-
brane [23]. It could be postulated that this occurs since GPI,
unlike the other examined cargoes, accumulates in a different
perinuclear compartment upon 20 C incubation and, there-
fore, we were monitoring its exit from another subcellular
compartment in which actin dynamics is irrelevant. This possi-
bility was discarded since cells co-expressing YFP-GPI and
Table 1
Mobile fraction (MF; in %) and T
1/2
(in min) values for each GFP/YFP-tagged cargo exiting from the TGN examined by iFRAP in control and actin
toxin-treated cells
Basolateral VSV-G Apical VSV-G p75NTR GPI
MF T
1/2
MF T
1/2
MF T
1/2
MF T
1/2
Control 56 ± 1.9 (n= 6) 3.0 36.6 ± 3.4 (n= 10) 3.2 46.1 ± 4.5 (n= 8) 7.1 44.8 ± 2.9 (n= 11) 4.1
Jpk 9.5 ± 4.2
***
(n= 6) >10 41.2 ± 3.5 (n= 6) 4.6 28.4 ± 6.0
*
(n= 7) 3.5 41.8 ± 4.5 (n= 6) 3.1
LtB 26.8 ± 8.9
**
(n= 6) 5.7 25.2 ± 3.7
*
(n= 7) 6.2 26.9 ± 3.6
**
(n= 9) 2.1 38.7 ± 4.9 (n= 8) 3.0
Results are the means ± S.E.M. Statistical significance:
*
P60.05,
**
P60.01,
***
P60.001.
Fig. 3. Inverse FRAP decay curves of non-raft-associated apical-
targeted GFP-p75NTR (A) and raft-associated apical YFP-GPI-
anchored protein (B). Results are the means ± S.E.M. from ncells
indicated in Table 1.
F. La
´zaro-Die
´guez et al. / FEBS Letters 581 (2007) 3875–3881 3879

CFP-VSV-G showed that both proteins largely colocalized in
the same compartment upon 19.5 C accumulation (Fig. 4).
An alternative and more probable explanation is that Golgi-
associated lipid rafts are not enriched with PIP
2
like those in
the plasma membrane. In this respect, it is known that PIP
2
levels at the Golgi are very low [38]. Recent data have provided
additional clues on how rafts might participate in apical sort-
ing, regardless of the putative participation of actin filaments.
These authors demonstrated that protein oligomerization is a
specific requirement for apical sorting of GPI-tagged apical
proteins in MDCK and Fischer rat thyroid (FRT) cells, and
is not used by transmembrane, non-raft-associated apical pro-
teins [19]. Association with rafts in the Golgi seems to be insuf-
ficient for this process and an additional step that concentrates
the proteins into large complexes is required, as suggested by
the fact that this behavior is not observed for basolateral
GPI-proteins [39].
The results reported here indicate that transport carriers that
contain raft-associated GPI anchored proteins are primarily
sorted or formed at the TGN independently of actin. This is
not the case for those non-raft-associated cargoes destined to
apical or basolateral plasma membrane domains. Moreover,
data suggest that actin dynamics is particularly relevant to car-
go targeted to the basolateral plasma membrane domain.
Acknowledgements: We thank Ine
´s Ferna
´ndez-Ulibarri and Emma
Martı
´nez-Alonso for critical reading of the manuscript, Maite Mun˜oz
and Anna Bosch for technical support and Robin Rycroft for editorial
assistance. This work has been supported by Grants BFU2006-876/
BMC and European Commission RTN2002-00252 to G.E.
References
[1] Griffiths, G. and Simons, K. (1986) The trans-Golgi network:
sorting at the exit site of the Golgi complex. Science 234, 438–443.
[2] Wandinger-Ness, A., Bennett, M.K., Antony, C. and Simons, K.
(1990) Distinct transport vesicles mediate the delivery of plasma
membrane proteins to the apical and basolateral domains of
MDCK cells. J. Cell Biol. 11, 987–1000.
[3] Keller, P., Toomre, D., Diaz, E., White, J. and Simons, K. (2001)
Multicolour imaging of post-Golgi sorting and trafficking in live
cells. Nat. Cell Biol. 3, 40–149.
[4] Musch, A. (2004) Microtubule organization and function in
epithelial cells. Traffic 5, 1–9.
[5] Egea, G., La
´zaro-Die
´guez, F. and Vilella, M. (2006) Actin
dynamics at the Golgi complex in mammalian cells. Curr. Opin.
Cell Biol. 18, 168–178.
[6] McNiven, M.A. and Thompson, H.M. (2006) Vesicle formation
at the plasma membrane and trans-Golgi network: the same but
different. Science 313, 1591–1594.
[7] del Toro, D., Canals, J.M., Gine
´s, S., Kojima, M., Egea, G. and
Alberch, J. (2006) Mutant huntingtin impairs the post-Golgi
trafficking of brain-derived neurotrophic factor but not its
Val66Met polymorphism. J. Neurosci. 26, 12748–12757.
[8] Hirschberg, K., Miller, C.M., Ellenberg, J., Presley, J.F., Siggia,
E.D., Phair, R.D. and Lippincott-Schwartz, J. (1998) Kinetic
analysis of secretory protein traffic and characterization of Golgi
to plasma membrane transport intermediates in living cells. J. Cell
Biol. 143, 1485–1503.
[9] Musch, A., Cohen, D., Kreitzer, G. and Rodriguez-Boulan, E.
(2001) Cdc42 regulates the exit of apical and basolateral proteins
from the TGN. EMBO J. 20, 2171–2179.
[10] Cao, H., Weller, S., Orth, J.D., Chen, J., Huang, B., Chen, J.L.,
Stamnes, M. and McNiven, M.A. (2005) Actin and Arf1-
dependent recruitment of a cortactin-dynamin complex to the
Golgi regulates post-Golgi transport. Nat. Cell Biol. 7, 483–492.
[11] Apodaca, G. (2001) Endocytic traffic in polarized epithelial cells:
role of the actin and microtubule cytoskeleton. Traffic 2, 149–159.
[12] Brown, B.K. and Song, W. (2001) The actin cytoskeleton is
required for the trafficking of the B cell antigen receptor to the late
endosomes. Traffic 2, 414–427.
[13] Mundy, D.I., Machleidt, T., Ying, Y.S., Anderson, R.G. and
Bloom, G.S. (2002) Dual control of caveolar membrane traffic by
microtubules and the actin cytoskeleton. J. Cell Sci. 115, 4327–
4339.
[14] Traub, L.M. (2005) Common principles in clathrin-mediated
sorting at the Golgi and the plasma membrane. Biochim. Biophys.
Acta 1744, 415–437.
[15] Bonifacino, J.S. and Traub, L.M. (2003) Signals for sorting of
transmembrane proteins to endosomes and lysosomes. Annu.
Rev. Biochem. 72, 395–447.
[16] Potter, B.A., Ihrke, G., Bruns, J.R., Weixel, K.M. and Weisz,
O.A. (2004) Specific N-glycans direct apical delivery of trans-
membrane, but not soluble or glycosylphosphatidylinositol-
anchored forms of endolyn in Madin-Darby canine kidney cells.
Mol. Biol. Cell 15, 1407–1416.
[17] Mun˜ iz, M. and Riezman, H. (2000) Intracellular transport of
GPI-anchored proteins. EMBO J. 19, 10–15.
[18] Schuck, S. and Simons, K. (2004) Polarized sorting in epithelial
cells: raft clustering and the biogenesis of the apical membrane. J.
Cell Sci. 117, 5955–5964.
[19] Paladino, S., Sarnataro, D., Tivodar, S. and Zurzolo, C. (2007)
Oligomerization is a specific requirement for apical sorting of
glycosyl-phosphatidylinositol-anchored proteins but not for non-
raft-associated apical proteins. Traffic 8, 251–258.
[20] Keller, P. and Simons, K. (1997) Post-Golgi biosynthetic
trafficking. J. Cell Sci. 110, 3001–3009.
[21] Rodrı
´guez-Boula
´n, E., Kreitzer, G. and Musch, A. (2005)
Organization of vesicular trafficking in epithelia. Nat. Rev. Mol.
Cell. Biol. 6, 233–247.
[22] Rustom, A., Bajohrs, M., Kaether, C., Keller, P., Toomre, D.,
Corbeil, D. and Gerdes, H.H. (2002) Selective delivery of
secretory cargo in Golgi-derived carriers of nonepithelial cells.
Traffic 3, 279–288.
[23] Caroni, P. (2001) New EMBO members’ review: actin cytoskel-
eton regulation through modulation of PI(4,5)P
2
rafts. EMBO J.
20, 4332–4336.
Fig. 4. Colocalization of YFP-GPI and the CFP-VSV-G wild-type form in the Golgi complex of COS-1 cells incubated at 19.5 C in the presence of
cycloheximide. Bar, 10 lm.
3880 F. La
´zaro-Die
´guez et al. / FEBS Letters 581 (2007) 3875–3881

[24] Musch, A., Xu, H., Shields, D. and Rodrı
´guez-Boula
´n, E. (1996)
Transport of vesicular stomatitis virus G protein to the cell
surface is signal mediated in polarized and nonpolarized cells. J.
Cell Biol. 133, 543–558.
[25] Rindler, M.J., Ivanov, I.E., Plesken, H., Rodrı
´guez-Boula
´n, E.
and Sabatini, D.D. (1984) Viral glycoproteins destined for
apical or basolateral plsma membrane domains traverse the
same Golgi apparatus during their intracellular transport in
doubly infected Madin-Darby canine kidney cells. J. Cell Biol. 98,
1304–1319.
[26] Lisanti, M.P., Caras, I.W., Davitz, M.A. and Rodrı
´guez-Boula
´n,
E. (1989) A glycophospholipid membrane anchor acts as an apical
targeting signal in polarized epithelial cells. J. Cell Biol. 109,
2145–2156.
[27] Yeaman, C., Le Gall, A.H., Baldwin, A.N., Monlauzeur, L.,
Le Bivic, A. and Rodriguez-Boulan, E. (1997) The O-glycosylated
stalk domain is required for apical sorting of neurotro-
phin receptors in polarized MDCK cells. J. Cell Biol. 139,
929–940.
[28] Keller, P., Toomre, D., Diaz, E., White, J. and Simons, K. (2001)
Multicolour imaging of post-Golgi sorting and trafficking in live
cells. Nat. Cell Biol. 3, 140–149.
[29] Yoshimori, T., Keller, P., Roth, M.G. and Simons, K. (1996)
Different biosynthetic transport routes to the plasma membrane in
BHK and CHO cells. J. Cell Biol. 133, 247–256.
[30] Schuck, S., Gerl, M.J., Ang, A., Manninen, A., Keller, P.,
Mellman, I. and Simons, K. (2007) Rab10 is involved in
basolateral transport in polarized Madin-Darby Canine Kidney
cells. Traffic 8, 47–60.
[31] Cohen, D., Musch, A. and Rodrı
´guez-Boula
´n, E. (2001) Selective
control of basolateral membrane protein polarity by Cdc42.
Traffic 2, 556–564.
[32] Kessels, M.M., Dong, J., Leibig, W., Westermann, P. and
Qualmann, B. (2006) Complexes of syndapin II with dynamin II
promote vesicle formation at the trans-Golgi network. J. Cell Sci.
119, 1504–1516.
[33] Kerkhoff, E., Simpson, J.C., Leberfinger, C.B., Otto, I.M.,
Doerks, T., Bork, P., Rapp, U.R., Raabe, T. and Pepperkok,
R. (2001) The Spir actin organizers are involved in vesicle
transport processes. Curr. Biol. 11, 1963–1968.
[34] Johnson, D., Lanahan, A., Buck, C.R., Sehgal, A., Morgan, C.,
Mercer, E., Bothwell, M. and Chao, M. (1986) Expression and
structure of the human NGF receptor. Cell 47, 545–554.
[35] Helms, J.B. and Zurzolo, C. (2004) Lipids as targeting signals:
lipid rafts and intracellular trafficking. Traffic 5, 247–254.
[36] Brown, D.A. and Rose, J.K. (1992) Sorting of GPI-anchored
proteins to glycolipid-enriched membrane subdomains during
transport to the apical cell surface. Cell 68, 533–544.
[37] Paladino, S., Pocard, T., Catino, M.A. and Zurzolo, C. (2006)
GPI-anchored proteins are directly targeted to the apical surface
in fully polarized MDCK cells. J. Cell Biol. 172, 1023–1034.
[38] De Matteis, M.A. and Godi, A. (2004) Protein-lipid interactions
in membrane trafficking at the Golgi complex. Biochim. Biophys.
Acta 1666, 264–274.
[39] Sarnataro, D., Paladino, S., Campana, V., Grassi, J., Nitsch, L.
and Zurzolo, C. (2002) PrPC is sorted to the basolateral
membrane of epithelial cells independently of its association with
rafts. Traffic 3, 810–821.
F. La
´zaro-Die
´guez et al. / FEBS Letters 581 (2007) 3875–3881 3881