The NF-kB Signaling Protein Bcl10
Regulates Actin Dynamics
by Controlling AP1 and OCRL-Bearing Vesicles
Sabrina Marion,1,2,3Julie Mazzolini,1,2,3,9Floriane Herit,1,2,3,9Pierre Bourdoncle,1,2,3Nade `ge Kambou-Pene,1,2,3
Stephan Hailfinger,4Martin Sachse,5Ju ¨rgen Ruland,6Alexandre Benmerah,1,2,3Arnaud Echard,7,8Margot Thome,4
and Florence Niedergang1,2,3,*
1Inserm, U1016, Institut Cochin, Paris, France
2CNRS, UMR 8104, Paris, France
3Universite ´ Paris Descartes, Sorbonne Paris Cite ´, Paris, France
4Department of Biochemistry, University of Lausanne, CH-1066 Epalinges, Switzerland
5Institut Pasteur, Imagopole, Plate-Forme de microscopie ultrastructurale, 25-28 rue du Dr Roux, 75724 Paris Cedex 15, France
6Institut fu ¨r Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universita ¨t Mu ¨nchen, 81675 Mu ¨nchen, Germany
7Institut Pasteur, Membrane Traffic and Cell Division Lab, 25-28 rue du Dr Roux, 75724 Paris Cedex 15, France
8CNRS, URA2582, Paris, France
9These authors contributed equally to the work
The protein Bcl10 contributes to adaptive and innate
immunity through the assembly of a signaling
complex that plays a key role in antigen receptor
and FcR-induced NF-kB activation. Here we demon-
strate that Bcl10 has an NF-kB-independent role in
actin and membrane remodeling downstream of
FcR in human macrophages. Depletion of Bcl10
impaired Rac1 and PI3K activation and led to an
abortive phagocytic cup rich in PI(4,5)P2, Cdc42,
and F-actin, which could be rescued with low doses
of F-actin depolymerizing drugs. Unexpectedly, we
found Bcl10 in a complex with the clathrin adaptors
AP1 and EpsinR. In particular, Bcl10 was required
to locally deliver the vesicular OCRL phosphatase
that regulates PI(4,5)P2and F-actin turnover, both
crucial for the completion of phagosome closure.
Thus, we identify Bcl10 as an early coordinator of
NF-kB-mediated immune response with endosomal
trafficking and signaling to F-actin remodeling.
The protein B cell lymphoma/leukemia-10 (Bcl10) plays a central
role inthe stimulationof immuneresponses triggered byimmune
recognition receptors, such as the B and T cell antigen recep-
tors, as well as phagocytic receptors that bind to the Fc portion
of immunoglobulins (FcRs) (Rawlings et al., 2006; Thome and
Weil, 2007). Indeed, Bcl10-deficient mice are severely immuno-
deficient because of impaired antigen (Ag) receptor- and FcR-
induced NF-kB activation and cytokine production (Chen et al.,
2007; Klemm et al., 2006; Ruland et al., 2001; Xue et al., 2003).
Bcl10 and its interaction partners Malt1 (MALT lymphoma trans-
location protein-1) and Carma1 (CARD-containing MAGUK
protein-1) are essential signaling components in Ag receptor-
triggered lymphocytes but also in cells derived from the
activated B cell subtype of diffuse large B cell lymphomas
(ABC-DLBCL), which critically depend on abnormal constitutive
NF-kB activation for survival (Ngo et al., 2006). The appearance
of lymphomas of the mucosa-associated lymphoid tissue (MALT
lymphomas) often correlates with a persistent stimulation of the
immune system by an infectious agent, such as Helicobacter
pylori (Du, 2007). Aggressive forms of MALT lymphomas are
characterized by chromosomal translocations of the genes en-
proliferation by constitutive NF-kB activation (Du, 2007). Despite
these recent insights into the biological relevance of Bcl10, the
molecular function of Bcl10 is still poorly understood. Biochem-
ical studies that were performed mainly in T cells have shown
that Ag receptor stimulation triggers the recruitment and activa-
tion of Carma1 (also known as CARD11), which stimulates the
subsequent recruitment of preformed Bcl10-Malt1 complexes.
The assembled Carma1-Bcl10-Malt1 (CBM) complex activates
NF-kB by promoting the physical recruitment and activation of
the NF-kB-regulating IkB kinase (IKK) complex (Thome et al.,
2010). The IKKb kinase subunit phosphorylates IkBa, targeting
it for ubiquitinylation and degradation by the proteasome. Freed
NF-kB dimers (predominantly p65 (RelA)/p50) then translocate
to the nucleus to activate gene transcription (Ha ¨cker and Karin,
2006). A related CBM complex containing the Carma1 homolog
CARD9, Bcl10, and Malt1 is thought to play a similar role in
promoting NF-kB activation by Dectin-1, Dectin-2, and FcR in
monocytes/macrophages and dendritic cells (Bi et al., 2010;
Gross et al., 2006; Hara et al., 2007). Myeloı ¨d cells have a high
expression of CARD9, which binds to Bcl10 via its CARD motif
(Gross et al., 2006; Hara et al., 2008; Hara and Saito, 2009).
During both fungal and bacterial internalization, CARD9 is
actively recruited to phagosomes and required for efficient path-
ogen killing but not necessary for the entry process itself
(Goodridge et al., 2009; Strasser et al., 2012; Wu et al., 2009).
In contrast, we have previously shown that Bcl10 is required
954 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
for efficient entry of particles downstream of FcR signaling in
human monocytes (Rueda et al., 2007). Upon TCR and FcR
stimulation, silencing of Bcl10 impaired actin cytoskeleton reor-
ganization required for immune synapse and phagocytic cup
formation, respectively. However, the molecular mechanisms
involved were not identified.
Phagocytosis strictly requires actin polymerization that, in the
case of FcR, is driven by the small GTPases Rac1, Rac2, and
Cdc42. Downstream effectors, such as the Wiskott-Aldrich
syndrome protein (WASP) activate the Actin-Related Protein
2/3complex(Arp2/3), which nucleates actinfilaments. Phospha-
polymerization that drives pseudopod formation, and its conver-
sion to PI(3,4,5)P3 is required for pseudopod extension and
phagosomal closure (Flannagan et al., 2012; Swanson, 2008).
In addition, efficient phagocytosis relies on focal exocytosis of
intracellular compartments that are thought to contribute to the
release of membrane tension, allowing efficient phagosome
formation around large particles (Braun and Niedergang, 2006;
Swanson, 2008). Although the major part of the membrane form-
ing a phagosome is of plasmalemmal origin (Touret et al., 2005),
several intracellular compartments, including recycling endo-
somes containing the SNARE protein VAMP3/Cellubrevin (Bajno
et al., 2000; Niedergang et al., 2003) and a subpopulation of late
endosomes bearing the SNARE protein VAMP7/TI-VAMP (Braun
et al., 2004), have been shown to be recruited and undergo focal
exocytosis at the site of phagocytosis. In addition, we have
shown that the clathrin-associated adaptor complex AP1 plays
animportant roleinorganizing focal exocytosisunder thecontrol
of the ADP-ribosylation factor ARF1 (Braun et al., 2007).
In this study, we unraveled an unexpected function for the
NF-kB signaling protein Bcl10 in promoting phagocytic cup
extension and closure. Our data indicate that Bcl10 modulates
Rac1 activity and PI3K activation and controls the delivery of
AP1/EpsinR-positive vesicular compartments containing the
PI(4,5)P2phosphatase oculocerebrorenal syndrome of Lowe 1
(OCRL1) in nascent phagosomes, thereby regulating F-actin
and activation of NF-kB to the nascent phagosome. Thus, Bcl10
plays a crucial role in coordinating the activation of the NF-kB-
mediated immune response and local signaling events regulat-
ing actin and membrane remodeling at the site of phagocytosis.
Bcl10 Controls Phagocytic Cup Extension and Closure
To better characterize the defect in phagocytosis observed after
depletion of Bcl10 (Rueda et al., 2007), we generated THP1
human monocytic cells stably transduced with two distinct
shRNA constructs directed against Bcl10 (Bcl10 KD). Cells
transduced with a nonrelevant shRNA were used as control
(Ctrl). Both shRNAs efficiently reduced Bcl10 expression,
whereas the expression of the two other CBM proteins, Malt1
and CARD9, was not affected (Figure 1A). We also generated
two stable THP1 cell lines overexpressing FLAG-tagged
versions of either the wild-type Bcl10 protein (FLAG-Bcl10) or
the Bcl10 serine 138 phosphorylation-deficient mutant (FLAG-
S138A) (Rueda et al., 2007) (Figure 1A). FcR-mediated phagocy-
tosis efficiency was then measured after incubation of the cells
with IgG-coated red blood cells (IgG-RBC). Silencing of Bcl10
impaired phagocytosis, whereas its overexpression led to higher
phagocytosis efficiency than in control cells (Figure 1D). The
overexpression of FLAG-S138A, previously shown to inhibit
TCR-mediated actin remodeling (Rueda et al., 2007), strongly in-
hibited FcR-induced phagocytosis. In addition, the absence of
Bcl10 in primary human monocyte-derived macrophages
(MDM) treated with siRNA (Figures 1B and 1E) and in macro-
phages derived from bone marrows (BMDM) of Bcl10 knockout
(Bcl10?/?) (Ruland et al., 2001) mice (Figures 1C and 1F) also led
to inhibition of FcR-mediated phagocytosis, whereas initial
binding of the particles was not affected.
These results demonstrate that Bcl10 contributes to FcR-
mediated phagocytosis in human and mouse monocytes/
We next analyzed in more detail the phagocytosis defect
observed. Bcl10 KD cells or control cells were incubated with
IgG-RBC for different time points, then fixed, and stained for
external and internal RBC and F-actin. The number of F-actin-
enriched phagocytic cups present at the surface of the cells
and the number of internalized RBC were quantified for each
time point. Control cells exhibited a peak of actin cup formation
at 2 min that progressively decreased until 10 min, correspond-
ing to the internalization of the RBC (Figures 1G and 1H). Bcl10
KD cells displayed similar numbers of F-actin cups at 2 min,
but 56% ± 9% of the originally formed cups stayed opened
and were still observable at the cell surface at 10 min. Those
cups finally collapsed, resulting in a strong defect in internaliza-
tion efficiency at 30 min. A similar defect was observed in
Bcl10?/?BMDM, which was even more pronounced that actin
cup formation was initiated as early as 30 s after contact in
wild-type BMDM (Figures 1I and S1 available online).
Quantification of the fluorescence intensity associated with
F-actin staining in phagocytic cups compared to the cortical
area of the cell indicated similar amounts of F-actin in cups
formed at 5 min in Bcl10 KD and control cells (Figures 1J and
1K). However, the aborted cups that remained for 10 min at
the surface of the Bcl10 KD cells showed an increase in F-actin
content compared to control cups or cups formed at 5 min
in Bcl10 KD cells, suggesting that F-actin polymerization acti-
vity persisted and actin filaments accumulated (Figure 1K).
Moreover, these blocked cups exhibited a different morphology
by fluorescence microscopy (Figure 1H), which was further
examined by scanning electron microscopy (Figure 1L). Unlike
the cups observed in control cells at 5 min, which displayed
thin membrane extensions tightly apposed against the particle,
the Bcl10 KD cups exhibited disorganized and shorter exten-
sions around the RBC both at 5 and 10 min. As observed by fluo-
rescence microscopy, particles were internalized in control cells
at 10 min (Figure 1L).
Together, these results indicate that Bcl10 is not required for
the onset of actin polymerization but is rather critical for regu-
lating phagocytic cup extension and closure.
The NFkB Complex Is Recruited but Not Required for
Phagocytic Cup Formation
To further examine the link between the CBM signaling pathway
and the role of Bcl10 in phagocytic cup formation, we analyzed
the formation of the CBM complex upon FcR stimulation. We
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 955
Figure 1. Bcl10 Promotes Phagocytic Cup Extension and Closure
(A)ImmunoblotoflysatesfromTHP1cells stablytransduced,either withcontrol ortwodifferentshRNAsequencestargeting Bcl10(shRNA#1and shRNA#2)(left),
expression was analyzed using specific antibodies. Tubulin was used as a loading control.
(B and C) Immunoblots of lysates from human monocyte-derived macrophages (MDM) treated with control siRNA or siRNA-targeting Bcl10 (B) and bone-
marrow-derived macrophages (BMDM) from WT mice or mice knockout for Bcl10 (Bcl10?/?) (C). Bcl10 expression was analyzed, and tubulin was used as
a loading control.
(D) Bcl10 knockdown cells (Bcl10 KD) with shRNA#1 or shRNA#2, as well as THP1 overexpressing FLAG-Bcl10 or FLAG-S138A were incubated with
IgG-opsonized RBC (IgG-RBC) for 10 min and the phagocytosis efficiency was quantified in at least 50 cells per experiment. The results are expressed as the
percentage of control cells ± SEM (n = 3 experiments, p < 0.05).
(E and F) Control MDM or MDM silenced for Bcl10 (E) and WT or Bcl10?/?BMDM (F) were incubated with IgG-RBC for 60 min and the RBC association efficiency
(gray bars) and phagocytic activity (black bars) was quantified in at least 25 cells per experiment. The results are expressed as the percentage of control cells ±
SEM (n = 3 experiments, p < 0.05).
(G) Bcl10 KD and control THP1 cells were incubated with IgG-RBC for the indicated times, and the number of F-actin-positive phagocytic cups per cell (left) and
the corresponding number of internalized RBC (right) were quantified for at least 50 cells per experiment (data show the means ± SEM, n = 3).
(H) Control and Bcl10 KD cells were incubated with IgG-RBC for 5 or 10 min, fixed and labeled for F-actin (green) and IgG-RBC (red). Asterisks indicate inter-
(I) WT or Bcl10?/?BMDM were incubated with IgG-RBC for the indicated times and the mean number of F-actin-positive phagocytic cups per cell was quantified
for at least 25 cells per experiment (data show the means ± SEM, n = 3).
Bcl10 Controls Endosome Traffic and Actin Dynamics
956 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
incubated FLAG-Bcl10-expressing THP1 cells with IgG-RBC (or
nonopsonized RBC as control) for 5 min in order to trigger FcR
signaling and then immunoprecipitated Bcl10 (Figure 2). Malt1
was coimmunoprecipitated with FLAG-Bcl10 independently of
FcR stimulation, confirming a previously described constitutive
association between the two proteins (Uren et al., 2000),
although a slight increase in binding was triggered by phagocy-
tosis. In contrast, CARD9 association with Bcl10 was greatly
enhanced upon phagocytosis (Figure 2A). The Ser138A mutant
of Bcl10 showed a strong constitutive association with Malt1
and CARD9, which was not modified by phagocytosis. We
next analyzed the localization of the CBM proteins during FcR-
mediated phagocytosis in THP1 monocytes and in human
primary macrophages by immunofluorescence. Bcl10, MALT1,
and CARD9 were all recruited in phagocytic cups defined by
the accumulation of F-actin (Figure 2B). In immune cells, the
formation andactivationofthe CBMcomplex triggers theactiva-
tion of the IKK complex, leading to the phosphorylation and
degradation of IkBa, thus activating NFkB RelA/p50 dimers.
We therefore analyzed whether the phagocytic cup could serve
as a local and transient signaling platform that initiates the
activation of the NF-kB pathway during the early steps of phago-
cytosis. Indeed, although nuclear translocation of RelA was
observed only upon prolonged stimulation as expected (Fig-
ure 2E), both IkBa and RelA were found highly enriched at the
site of phagocytosis after FcR stimulation in THP1 monocytes
(Figure 2C), in primary mouse macrophages (Figure 2D), and in
primary human macrophages (Figure 2E). In contrast, in the
blocked cups formed at the surface of cells silenced or knockout
for Bcl10, both RelA and IkBa exhibited a diffuse localization,
which was not concentrated in the cortical area beneath the
bound particle (Figures 2C and 2D). Thus, Bcl10 is critical to
recruit the IKK complex at sites of phagosome formation, where
the CBM complex is assembled and activated.
To further decipher the role of Bcl10 in activating IKK and
NF-kB during phagocytosis, we analyzed the phosphorylation
status of the IkBa regulatory subunit. Induction of phagocytosis
led to an early and transient phosphorylation of IkBa correlating
with the phagocytic cup closure step (5 and 10 min). Bcl10
silencing completely impaired IkBa phosphorylation, whereas
overexpression of Bcl10 accelerated and increased its phos-
phorylation (Figure 2F). The overexpression of the S138A mutant
of Bcl10, which perturbs actin polymerization but not IKK activa-
ure 1D), led to the phosphorylation of IkBa as efficiently as the
wild-type Bcl10 (Figure 2F), confirming that signaling to actin
polymerization is not required for IKK activation. Inversely, to
further analyze whether IKK activation is necessary for phago-
cytic cup formation, THP1 cells were transiently transfected to
express a dominant negative mutant of IkBa (DN-IkBa), which
NF-kB activation. We found that DN-IkBa-transfected cells dis-
played a normal FcR-mediated phagocytic activity (Figure 2G).
Together, thesedata reveal adual role forBcl10. On one hand,
Bcl10 regulates the recruitment and activation of the IKK
complex and NF-kB at phagocytic sites. On the other hand,
Bcl10 is able to modulate actin cytoskeleton dynamics, allowing
closure of the phagocytic cup, which we decided to characterize
in more details.
Silencing of Bcl10 Impairs Rac1 Activation
Because the stalled phagocytic cups present in Bcl10 KD cells
accumulated F-actin, we examined the localization of some
upstream factors known to stimulate actin assembly/disas-
sembly during phagosome formation, including the Rho-family
GTPases Cdc42 and Rac1 (Niedergang and Chavrier, 2005;
Swanson, 2008). Previous studies had shown that Cdc42 is
activated early during pseudopod extension, recruiting the
N-WASP/WASP actin nucleation promoting factors, which in
turn stimulate the Arp2/3 complex to promote actin nucleation
and polymerization (Flannagan et al., 2012; Swanson, 2008).
Here, we found that Arp3, N-WASP, and Cdc42 accumulated
in higher amounts in the blocked cups formed at 10 min in
Bcl10 KD cells as compared to normal phagocytic cups in
control cells (Figures 3A and 3B). In contrast, we did not monitor
significant changes in Rac1 recruitment (data not shown). We
then analyzed the Rac1 activation status during FcR-induced
phagocytosis by GTP pull-down assays. In control cells, Rac1
showed a transient activation peak at 5 and 10 min, correlating
with phagocytic cup closure. In contrast, Bcl10 silencing re-
sulted in both a higher basal activation of Rac1 and a defect in
Rac1 activation (Figures 3C and 3D).
We next investigated the effects of overexpressing constitu-
tive active (Rac1V12-GFP) or inactive (Rac1N17-GFP) mutants
of Rac1 on phagosome formation in control and Bcl10 KD cells.
First, we observed that wild-type Rac1 (Rac1WT-GFP) was en-
riched at the base of the phagocytic cup in control cells (Fig-
ure 3E). As previously described (Caron and Hall, 1998; Massol
et al., 1998), expression of Rac1N17 impaired phagosome
formation in control cells (Figure 3F). No additive defect was
observed in Bcl10 KD cells transfected with Rac1N17. Transfec-
tion of the constitutively active mutant Rac1V12 increased
phagocytic efficiency monitored 10 min after IgG-RBC binding
in control cells, but most importantly, fully rescued the phago-
cytic defect displayed by Bcl10 KD cells (Figures 3E and 3F).
Taken together, these data indicate that Rac1 is a downstream
to regulate phagosome extension and closure.
The stalled cups formed in Bcl10-silenced cells displayed
a phenotype similar to the one observed upon PI3K inhibition,
(J) Control and Bcl10 KD THP1 cells were incubated with IgG-RBC for 5 or 10 min and the amount of F-actin present at the phagocytic cup was quantified by
fluorescence microscopy. TheprofilesofF-actinfluorescenceintensitiesalongthelinesdrawnatthephagocyticsite(redline)andatthecellcortex(whiteline)are
shown (lower graph).
(K) The fluorescence intensities measured in the phagocytic cups were divided by the fluorescence intensities measured for cortical actin. This ratio defined the
index of recruitment. The means ± SEM of three independent experiments are plotted (n = 60 actin cups per condition, p < 0.05).
(L)Control and Bcl10KDcells wereincubatedwithIgG-RBCfor 5and10minandthenfixed and processed forscanning EM.TheRBC are pseudo-colored inred.
Arrows indicate sites of membrane deformation due to potentially internalized particles. Scale bars, 5 mm.
See also Figure S1.
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 957
Figure 2. Bcl10 Triggers the Phagosome Recruitment and the Activation of the IkBa/NF-kB Pathway during Phagocytosis
(A) FLAG-Ctrl, FLAG-Bcl10,and FLAG-S138Acells wereincubated withnonopsonizedRBC (lane ?)or IgG-RBC (lane+)for 5min.FLAG-Bcl10and FLAG-S138A
were immunoprecipitated with anti-FLAG antibodies and coimmunoprecipitated endogenous CARD9 and Malt1 as revealed by specific antibodies (upper). The
amounts of total proteins in lysates (5% of total lysate; lower panels) are shown in all conditions. FLAG-Ctrl cells were used as negative controls.
(B)FLAG-Bcl10-expressing THP1cellsor primarymacrophageswereincubatedwithIgG-RBCfor5min,fixed, andlabeled withanti-FLAG,anti-CARD9,orMalt1
(red) antibodies. F-actin in phagocytic cups was detected by incubation with phalloidin-Alexa488 (green). IgG-RBC were detected with secondary antibodies
(blue). Scale bars, 5 mm.
show details of the phagocytic cup marked by an arrow. Scale bars, 5 mm.
(D) Detection of RelA (red) and F-actin (green) in WT or Bcl10?/?BMDM incubated with IgG-RBC for 5 min. Arrows indicate F-actin-positive phagocytic cups.
Scale bars, 5 mm.
(E) Detection of RelA (red) in control MDM incubated with IgG-RBC for 5 min (upper panel) or 60 min (lower panel). F-actin (green) and RBC (blue) were also
labeled. Arrows indicate RelA-positive phagocytic cups in the upper panel and RelA-positive nuclei in the lower panel. Bars: 5 mm.
(F) Ctrl, Bcl10 KD, FLAG-Bcl10, and FLAG-S138A cells were incubated with IgG-RBC for the indicated times. Cells were then lysed, and an equal amount of
lysate was analyzed by western blot (WB) for phosphorylated IkBa (pIkBa), as well as total IkBa and Bcl10. Tubulin was used as a loading control.
(G) Control THP1 cells were transiently transfected for GFP as a negative control or dominant-negative IkBa (IkBa-DN), and phagocytosis efficiency was
assessedbyincubating cellswithIgG-RBCfor30min.Theresultsareexpressedasapercentage ofcontrolcells(mean±SEMofthreeindependent experiments,
n = 30 cells per experiments).
Bcl10 Controls Endosome Traffic and Actin Dynamics
958 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
with accumulated Cdc42 and F-actin at the base of the cup
and a defect in Rac1 activation (Araki et al., 1996). Therefore,
PI3-K activation during phagocytosis in control and Bcl10 KD
cells was examined by analyzing the phosphorylation status of
Akt, a direct downstream effector of PI3-K. In control cells,
Akt was transiently phosphorylated during phagosome forma-
tion with a maximal activation peak at 5 min. In Bcl10-silenced
cells, Akt activation upon FcR stimulation was impaired (Figures
3C and 3D).
Altogether, the accumulation of F-actin and upstream effec-
tors in the open Bcl10-deficient cups, and the correlating inhibi-
tion of Rac1 and PI3K in the absence of Bcl10, suggest that
a specific transition signal involving Bcl10 is required to downre-
gulate actin assembly, a step necessary for phagosome closure.
Figure 3. Silencing of Bcl10 Impairs Rac1 Activation
(A) Detection of Arp3, N-WASP, Cdc42 (red), and F-actin (green) in Bcl10 KD cells incubated with IgG-RBC for 10 min. Images show one z-plane of the cell
analyzed by fluorescence microscopy. Scale bars, 5 mm.
(B) The recruitment of N-WASP and Cdc42 in phagocytic cups was quantified in Ctrl and Bcl10 KD cells by measuring the fluorescence intensities in the cups
compared to the cortical region of the cell as described in Figures 1J and 1K. The means ± SEM of three independent experiments are plotted (n = 45 actin cups
per condition, p < 0.05).
(C) Ctrl and Bcl10KD cells were incubated with IgG-RBC for the indicated times and lysed and an equal amount of total lysate was used to pull-down activated
Rac1. The amount of Rac1-GTP in pull-down samples, and the amount of total Rac1 and Bcl10 in the lysates were analyzed by WB. Phosphorylated AKT (pAKT)
and total AKT in the lysates were analyzed by WB (one representative experiment of three is shown).
(D) The graph indicates the fold activation of Rac1 (left) and AKT (right) induced upon phagocytosis, corresponding to the densitometric quantification of
immunoblots as described in (C). Data show mean ± SEM from at least three independent experiments.
(E) Rac1WT-GFP, Rac1N17-GFP, and Rac1-V12 were transiently transfected in control THP1 cells, and their localization was analyzed by fluorescence
microscopy (red) after incubation with IgG-RBC for 5 min (blue or phase contrast). F-actin was labeled in green. Scale bar, 5 mm.
(F) Quantification of the phagocytic efficiency (expressed as the mean number of RBC per cell ± SEM of three independent experiments, n = 40 cells per
experiments) in Ctrl and Bcl10 KD cells transiently transfected with Rac1WT-GFP, Rac1N17-GFP, and Rac1-V12.
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 959
Actin Disassembly Is Sufficient to Rescue the
Phagocytosis Defect in Bcl10-Deficient Cells
Because Cdc42 inactivation and actin disassembly are depen-
dent on PI(4,5)P2hydrolysis (Scott et al., 2005), we analyzed
whether the accumulation of F-actin in the stalled phagocytic
cups formed in Bcl10-silenced cells was correlated with an
accumulation of PI(4,5)P2. For this, the PH domain of PLCd
fused to GFP (PHPLCd-GFP) was transiently expressed in control
and Bcl10 KD cells. In control cells, we observed that PI(4,5)P2
was preferentially enriched in the tips of pseudopods extending
around the particle and almost absent from the base of the
cup, as previously described (Scott et al., 2005). In contrast, in
Bcl10 KD cells, PI(4,5)P2was found accumulated all around
the blocked phagocytic cups, together with F-actin (Figures 4A
and 4B). These results suggest that the absence of Bcl10
led to a defective PI(4,5)P2hydrolysis at the base of the phago-
To directly investigate whether abnormal F-actin accumula-
tion, likely caused by PI(4,5)P2accumulation, was the cause of
the phagocytic defect, we treated Bcl10-depleted cells with
very low amounts (5–20 nM) of the F-actin depolymerizing drug
Latrunculin-A (LatA). In control cells, the addition of 5 nM LatA
did not modify phagocytosis efficiency, whereas higher doses
(10 and 20 nM) caused a decrease in phagocytosis (Figure 4C).
Remarkably, phagocytic defects associated with Bcl10 deple-
tion were completely rescued by 5 nM LatA treatment (Fig-
ure 4C). These observations show that abnormal F-actin levels
are responsible for the defects in phagocytosis observed in the
absence of Bcl10.
Finally, we examined by transmission electron microscopy the
phagocytic cups formed in control and Bcl10 KD cells. Phago-
cytic cups in control cells displayed an actin-rich zone in the
extending pseudopodia at 5 min, which were devoid of intracel-
lular compartments. Intracellular vesicles were detected in the
close vicinity of the plasma membrane in the region localized
at the base of the nascent phagosome (Figure 4D). In contrast,
blocked cups of Bcl10 KD cells exhibited a thick dense region
all around the extending cup at 5 and 10 min, which was devoid
of intracellular organelles (Figure 4D). This result suggests that
the accumulation of F-actin in Bcl10-deficient phagocytic cups
could be accompanied by a defect in focal delivery of vesicles,
thereby impairing efficient phagosome formation.
Depolymerization of F-Actin and Focal Delivery
of Intracellular Vesicles at the Base of the Forming
the delivery of intracellular compartments during phagosome
formation has not been precisely investigated so far. To gain
insight into this issue, we used the ‘‘frustrated phagocytosis’’
experimental setup, in which cells are allowed to spread on
IgG-coated coverslips, combined with total internal reflexion
Figure 4. Actin Disassembly Is Sufficient to Rescue the Phagocytosis Defect in Bcl10 KD Cells
(A) The localization of PI(4,5)P2(red) was detected after transient transfection of Ctrl and Bcl10 KD THP1 cells to express the PH domain of PLCd coupled to GFP.
Cellswereincubated withIgG-RBCfor5and10min,respectively,fixed andstainedforF-actin (green)and RBC(blue)and imageswereanalyzedbyfluorescence
microscopy. Scale bars, 5 mm.
(B) Quantification of PI(4,5)P2recruitment in phagocytic cups in Ctrl and Bcl10 KD cells from images acquired as described in A (p < 0.05).
(C) Ctrl and Bcl10 KD cells were preincubated with the indicated amount of Latrunculin A for 30 min and phagocytosis was induced for 30 min. Phagocytic
efficiency was quantified and expressed as the percentage of internalized RBC per cell/total amount of RBC detected. Data show the means ± SEM of three
independent experiments (n = 45 cells per experiment).
(D) Transmission EM images of Ctrl and Bcl10KD cells incubated with IgG-RBC for 5 and 10 min, respectively, showing the distribution of intracellular vesicles
around the phagocytic cups in each cell type. Scale bars, 2 mm.
Bcl10 Controls Endosome Traffic and Actin Dynamics
960 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
fluorescence microscopy (TIRFM) imaging to analyze the dy-
namic events taking place close to the plasma membrane (Fig-
ure 5A). RAW264.7 macrophages were transfected to transiently
express VAMP3-GFP to label the endocytic compartments and
Lifeact-mCherry to detect F-actin. The actin polymerization
signal was clearly detected as an intense ring in the peripheral
region of the spreading cell (Figure 5B; Movie S1). The lack of
signal in the cell center was not due to a loss of cell adhesion
membrane, was present in that region (data not shown). In
in the central region of the cell, which was devoid of F-actin
(Figure 5B). We designed a custom analysis with the ImageJ
software and quantified that 67.9% ± 2.1% of the total VAMP3
vesicles were detected in the region (R2) defined as the cell
center, whereas 32.1% were detected in the region R1, corre-
sponding to the cell periphery, at sites of more intense F-actin
staining (n = 5, Figures 5C and 5D). Cell spreading occurred in
a completely disorganized manner on naked glass or on cover-
slips coated with polylysine (Figure S2). In that case, F-actin
and VAMP3 staining were evenly distributed close to the plasma
membrane. Therefore, the differential localization of F-actin and
VAMP3 was specifically induced upon FcR triggering.
To get better insight into the three-dimensional (3D) spatio-
polymerization during phagosome formation, we designed a
specific ‘‘phagosome closure assay’’ to follow phagocytic cup
extension and closure based on TIRFM. We coated glass cover-
slips with opsonized particles and observed the closure of the
plasma membrane around the particles. This experimental setup
enabled us to acquire TIRFM images of the tips of the pseudo-
pods that are apposed to the glass coverslip and to concomi-
tantly acquire the epifluorescence images at 3 mm distance
above the coverslip (Figure 5E). We observed a clear recruitment
of F-actin at the very tips of the closing pseudopods, whereas no
VAMP3-positive vesicles could be detected in this closing zone
of the phagocytic cup. VAMP3 was only detected as a diffuse
signal corresponding to the plasma membrane (Figure 5F). In
contrast, VAMP3-positive vesicles were clearly detected at the
cence images (Figure 5F).
Together, our results clearly show that F-actin polymerization
occurs in the tips of the extending pseudopods and that vesic-
ular trafficking is localized at the base of the phagocytic cup in
a zone where F-actin is depolymerized.
Bcl10 Controls Vesicular Trafficking at the Site of
To get further insight into the molecular links between Bcl10 and
the defective actin reorganization at sites of phagocytosis, we
performed coimmunoprecipitation experiments with an anti-
Bcl10 antibody and mass spectrometry analysis on lysates
from THP1 monocytes. Unexpectedly, we found that AP1 and
EpsinR were part of the proteins coimmunoprecipitated with
Bcl10. We confirmed this result by western blot (Figure 6A).
The same proteins were found immunoprecipitated with FLAG-
Bcl10 using an anti-FLAG antibody in the FLAG-Bcl10-overex-
pressing cells (Figure 6B). Interestingly, AP1 and EpsinR were
shown to interact and regulate transport between early endo-
somes and the trans-Golgi network (TGN) (Hirst et al., 2003;
Popoff et al., 2007; Shiba et al., 2010). Furthermore, we previ-
ously described that AP1 is recruited to nascent phagosomes
and required for efficient phagocytic cup formation in macro-
phages (Braun et al., 2007). Therefore, we first investigated
whether EpsinR, like AP1, is functionally involved in FcR phago-
cytosis (Figure 6C). EpsinR depletion in control THP1 cells
caused a 66% ± 18% decrease in phagocytic efficiency (Fig-
ure 6D), without any effect on particle binding, indicating a role
of EpsinR in phagosome formation. We next analyzed the
recruitment of AP1 and EpsinR at sites of phagosome formation.
Although both proteins were found enriched in nascent phago-
somes in control cells, their recruitment was strongly inhibited
in the absence of Bcl10 (Figures 6E and 6F).
Our findings suggest that Bcl10 is required for the recruitment
of AP1/EpsinR-positive endosomal compartments at sites of
The OCRL PI(4,5)P2Phosphatase Is Recruited
Downstream of Bcl10
Because Bcl10 was associated with the endocytic machinery
and because PI(4,5)P2and F-actin accumulated in high amounts
P2and PI(3,4,5)P35-phosphatase OCRL, which is localized in
cells in the TGN and peripheral endosomes (Choudhury et al.,
2005; Erdmann et al., 2007), as it has been described for the
proteins AP1 and EpsinR (Hirst et al., 2003). We therefore exam-
ined the colocalization of OCRL and AP1 in the whole cell and
during phagocytosis in THP1 cells. As expected, we observed
that OCRL and AP1 displayed a partial colocalization at the
TGN level and in peripheral endosomes in nonstimulated cells
(Figure S3). Most importantly, upon FcR stimulation, the two
proteins were found colocalized in vesicles recruited at the
phagocytic cup (Figure 6G, arrows). This observation suggests
that OCRL could be delivered at sites of phagocytosis via the
recruitment of AP1-positive compartments, which might play
an essential role to regulate actin depolymerization, a step
necessary for the completion of phagosome closure.
It has been recently shown that OCRL depletion causes
a PI(4,5)P2and F-actin accumulation at the cleavage furrow of
dividing cells, thereby impairing cytokinesis (Dambournet et al.,
2011). In addition, it was shown that OCRL is required for
efficient phagocytosis in Dictyostelium discoideum and in
macrophages (Bohdanowicz et al., 2012; Loovers et al., 2007).
We observed that OCRL was clearly enriched in nascent phago-
somes in control THP1 monocytes and in primary human macro-
phages (Figures 7A and 7C). Importantly, this recruitment was
strongly impaired in Bcl10 KD cells (Figures 7A and 7B). More-
over, OCRL depletion by siRNA in THP1 cells (Figure 7D)
impaired FcR-mediated phagocytosis (Figure 7E). Importantly,
treatment of the OCRL-depleted cells with 5 nM of LatA fully
rescued the phagocytic defect (Figure 7E), as previously
observed in Bcl10 KD monocytes (Figure 4C). To assess the
functional role of OCRL relative to Bcl10, we silenced OCRL
expression in FLAG-Bcl10-overexpressing cells (Figure 7F).
The depletion of OCRL abolished the increase of phagocytosis
normally observed in these cells (Figure 7G), suggesting that
OCRL acts downstream of Bcl10 and participates in a common
signaling pathway that regulates phagosome formation. We
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 961
Figure 5. Spatiotemporal Organization of F-Actin and VAMP3 during Phagocytosis by Macrophages
(A) Schematic representation of the frustrated phagocytosis assay.
(B) Frustrated phagocytosis was performed using RAW264.7 murine macrophages transiently expressing Lifeact-mCherry and VAMP3-GFP. VAMP3-GFP (top)
and Lifeact-mCherry (bottom) imaging by TIRFM were performed alternatively at 37?C every 100 ms during 120 s. Scale bar, 10 mm.
(C) Sequences of images illustrating the procedure used by the designed macrocommand in ImageJ software (see the Experimental Procedures); (1) total cell
area (R1+R2); (2) R2 region obtained after 12 erosions; (3) TopHat filter detection of dotty particles in region R2.
(D) The number of VAMP3-positive vesicles was quantified and plotted over time in areas R2 and R1 corresponding respectively to the central region of the cell
and to the cell periphery. The histogram on the right shows the mean percentage ± SEM of vesicles detected in the area R1 and R2 (n = 6 experiments).
(E) Schematic illustration of the phagosome closure assay.
Bcl10 Controls Endosome Traffic and Actin Dynamics
962 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
finallyanalyzedbyTIRFMthe recruitmentof OCRL-GFP-positive
structures and actin polymerization events in the cortical region
during frustrated phagocytosis (Figure 7H). In the majority of
control cells, F-actin appeared enriched at the cell periphery
and highly dynamic OCRL-positive structures were detected in
the central region (9 out of 11 independent experiments). By
contrast, in Bcl10 KD cells, polymerized actin was not localized
in the cell periphery and OCRL-positive compartments were
either very weakly detected in the TIRF region or appeared as
immobile structures (in 5 out of 11 cases) (Figure 7H).
Together, our results suggest that the phagocytic defect
of AP1 and OCRL-positive structures, which in turn causes local
defects in actin remodeling.
Bcl10 is well known to contribute to the immune response by
stimulating the activation of the IKK complex and NF-kB-
mediated transcription of inflammatory genes downstream of
antigen receptors in lymphocytes, and we found a similar func-
tion in macrophages. Unexpectedly, we observed that Bcl10
also mediated the active recruitment of IkBa and NF-kB to the
membrane of the forming phagosome. This could serve to
sequester the transcription factor before nuclear translocation,
therefore allowing precise temporal and spatial control of the
immune response. Interestingly, IkBa phosphorylation was not
required for phagosome formation, suggesting that NF-kB
activationand F-actincytoskeleton remodeling aretwo indepen-
dent functions exerted by Bcl10 during FcR phagocytosis in
Here, we unraveled an original mechanism by which Bcl10
regulates phagocytic cup extension and closure. Using a dedi-
cated assay based on TIRF microscopy, we analyzed the local-
ization of F-actin and vesicular trafficking during phagocytic cup
formation in 3D. We clearly observed that vesicular trafficking
events were concentrated at the base of the phagocytic cup
and transient polymerization activities of the actin meshwork
play a positive role in the fine-tuning of vesicles docking and
exocytosis (Malacombe et al., 2006), but in the phagocytic
cup, the mutual exclusion of actin and vesicles argues for
a model in which actin filaments act as a barrier that blocks
vesicle docking at the plasma membrane.
The phagocytic defect monitored in Bcl10 KD, as well as
OCRL KD cells could be rescued by low doses of actin depoly-
merizing drugs, suggesting that actin depolymerization is indeed
a prerequisite to complete phagosome closure. It was shown
that Cdc42 inactivation and F-actin disappearance at the basis
of the cup are directly correlated with and dependent on
PI(4,5)P2 hydrolysis (Scott et al., 2005). During phagosome
formation, several pathways contribute to the disappearance
of PI(4,5)P2. Besides the arrest of its synthesis, PLCg hydrolyzes
PI(4,5)P2. PI(4,5)P2 is also phosphorylated and thereby con-
sumed by PI3K into PI(3,4,5)P3. Previous studies demonstrate
that the accumulated 30PIs provide a negative feedback
response to deactivate Cdc42 that allows actin disassembly
necessary for phagosome closure (Beemiller et al., 2010). In
other systems, Rac1 and PI3K have been reported to signal in
a positive feedback loop to sustain PI3K activity (Welch et al.,
2003). Therefore, the defect in PI3K activity that we observed
in Bcl10 silenced cells might be the consequence of a defect in
Rac1 activation and, most importantly, could participate in the
phate-5-phosphatase (Inpp5) and/or OCRL (Bohdanowicz et al.,
2012). OCRL has been shown to be localized in the TGN and in
peripheral endosomal structures, together with AP1 (Choudhury
et al., 2005). We observed that Bcl10 silencing impaired AP1 and
OCRL delivery to the phagocytic cup. Based on these findings,
and because OCRL was reported to prefer PI(4,5)P2(Zhang
et al., 1995), we propose that OCRL could locally participate to
the hydrolysis of PI(4,5)P2and/or PI(3,4,5)P3to complete phag-
osome closure. Interestingly, recent data demonstrate that
during Yersinia entry in macrophages, the fusion of Rab5-posi-
tive vesicles with the forming vacuole requires PI3K activity
and that this event is essential for the recruitment of OCRL and
Inpp5b and the subsequent hydrolysis of PI(4,5)P2 (Sarantis
et al., 2012). Consistent with these findings, our data indicate
that Bcl10 is required for the stimulation of Rac1 and PI3K
activity, as well as the recruitment of OCRL, which also binds
to Rac1 via a Rho GAPlike domain, although the GAP activity
was not shown in vivo (Faucherre et al., 2003). Both activities
allow a rapid and efficient clearance of PI(4,5)P2in order to
complete phagosome closure.
Here we found that Bcl10 interacts with AP1 and EpsinR.
Importantly, we also performed Bcl10 pull-down experiments
with lysates of T cells after TCR activation. By mass spectros-
copy analysis, AP1 and AGAP2, a GAP factor for ARF1, were
found specifically associated with Bcl10 after T cell stimulation
(data not shown). We previously demonstrated that AP1 is re-
cruited at phagocytic sites under the control of ARF1 (Braun
et al., 2007). ARF1 activation is PI3K-dependent, and its inhibi-
tion impaired phagosome closure but not the initial step of
pseudopod extension (Beemiller et al., 2006). Furthermore,
it has been shown that AGAP2 is a PI(4,5)P2-dependent
ARF1GAP that associates with a fast recycling compartment
containing AP1, Rab4, and the transferrin receptor (Nie et al.,
2005). Therefore, it is conceivable that Bcl10 controls the
activation status of ARF1, and thus the membrane association
of AP1 and the endosomal dynamics that in turn regulate
PI(4,5)P2hydrolysis and actin depolymerization. Presumably,
factors other than OCRL are delivered to the phagocytic cup
via the AP1-positive vesicles and a dynamic association/disso-
ciation of signaling components is required for efficient actin
(bottom), and VAMP3-GFP (bottom) in TIRFM (top) and Epifluorescence (bottom) imaging were performed alternatively at 37?C every 50 ms by switching from
TIRF plan to RBC region (DZ = 3 mm) using a piezo. One region of interest is shown (arrows and inserts), corresponding to the phagocytic area for each epi-
fluorescence picture. Scale bar, 10 mm.
See also Figure S2.
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 963
polymerization/depolymerization cycle. Indeed, the Ser138A
mutant of Bcl10, which acts as a dominant negative mutant
for phagocytosis, was found to be constitutively associated
with CARD9 and AP1 (Figure S4). This appeared to efficiently
block actin dynamics and pseudopod closure efficiently.
Thus, our data argue for a model in which the vesicular
delivery of intracellular compartments is not only a way to
release plasma membrane tension but also allows to deliver
signaling cargos that regulate F-actin dynamics and phago-
some closure. We propose that intracellular compartments
(AP1, EpsinR positive) are required to provide signaling com-
ponents, such as OCRL, promoting PI(4,5)P2 consumption
and depolymerization of actin at the base of phagocytic
cups. This would then allow subdomains of the compartments
Figure 6. Bcl10 Regulates the Recruitment of AP1/EpsinR-Positive Compartments at the Site of Phagocytosis
(A)THP1celllysateswereincubatedwithpolyclonalanti-Bcl10antibodiesorirrelevantpolyclonalIgG, andcoimmunoprecipitated proteinswereanalyzedbyWB.
The amount of total proteins in lysates (5% of total lysate; right) are shown in all conditions.
(B) Lysates from FLAG or FLAG-Bcl10-expressing cells were incubated with monoclonal anti-FLAG antibodies, and coimmunoprecipitated proteins were
analyzed by WB using anti-AP1, EpsinR, Bcl10, and Malt1 antibodies.
(C)THP1 cells were treated with siRNA control or siRNA-targeting EpsinR. Equal amount of lysates were analyzed by WB. Bcl10 was used as a loading control.
(D) Quantification of the phagocytic efficiency in EpsinR-silenced cells. The data are expressed as the percentage of control cells and show the mean ± SEM of
three independent experiments.
(E and F) Immunofluorescencedetection of AP1 (E) and EpsinR (F) in Ctrl and Bcl10 KD cells incubated with IgG-RBC (blue) for 5 and 10 min, respectively. F-actin
is shown in green. Images represent one z-plane of the cell analyzed by fluorescence microscopy and deconvolution. The histogram on the right of each image
panel indicates the quantification of AP1 (E) and EpsinR (F) recruitment in the phagocytic cups in Ctrl and Bcl10 KD cells, as described previously. Data show the
means ± SEM of three independent experiments (n = 75 phagosomes, p < 0.05). Scale bars, 5 mm.
(G) Ctrl THP1 cells were transiently transfected with OCRL-GFP (green) and incubated with IgG-RBC (blue) for 5 min and colocalization with AP1 (red) was
analyzed by immunofluorescence. Arrows indicate two different phagocytic cups. Inserts on the right represent a different z-plane of the cell and show details of
the lower phagocytic cup. Arrows indicate vesicles detected at the cup, positive for OCRL (green) and AP1 (red). Scale bars, 5 mm.
See also Figures S3 and S4.
Bcl10 Controls Endosome Traffic and Actin Dynamics
964 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
equipped with fusion machineries to bud and fuse with the
plasma membrane (Braun et al., 2007; Niedergang et al.,
2003), for further pseudopod extension.
In conclusion, we have unraveled an NF-kB-independent
function for Bcl10 to promote intracellular trafficking and F-actin
dynamics at the internalization step of phagocytosis. Emerging
evidence indicates that Bcl10-containing signaling complexes
are implicated downstream of a growing number of receptors,
including not only immune receptors, but also EGFR or
G-protein-coupled receptors (Rosebeck et al., 2011). Therefore,
the Bcl10-dependent regulation of F-actin cytoskeleton and
membrane trafficking that we observed in macrophages
might be a mechanism involved in different functions in various
Information on antibodies, plasmids, cell culture, transfection and transduc-
tion, cell stimulation, immunoprecipitation and GST pull-down assays, trans-
mission and scanning electron microscopy, and mass spectrometry are
included in the Supplemental Experimental Procedures.
Figure 7. OCRL Is Recruited in Phagosomes Downstream of Bcl10
represent one z-plane of the cell analyzed by WF microscopy and deconvolved. Scale bars, 5 mm.
(B) The histogram indicates the quantification of OCRL recruitment in the phagocytic cups in Ctrl and Bcl10 KD cells, as described previously. Data show the
means ± SEM of three independent experiments (n = 30 phagosomes per experiment, p < 0.05).
(C) Detection of endogenous OCRL (red) inprimary human macrophagesincubated withIgG-RBC (blue) for 5 min. F-actin was labeled ingreen. Scale bars, 2mm.
(D) Immunoblots detecting OCRL from lysates of control THP1 cells transfected with control siRNA or targeting OCRL. Tubulin was used as a loading control.
(E) Quantification of phagocytic efficiency in THP1 cells treated with siRNA ctrl or siRNA OCRL, after incubation with IgG-RBC for 60 min, in absence (?Lat) or
presence (+Lat) of 5 nM Latrunculin A. The results are expressed as the percentage of control cells ± SEM (n = 3 experiments, p < 0.05).
(F) Immunoblots detecting OCRL from lysates of FLAG-Bcl10 cells transfected with siRNA control (lane ?) or targeting OCRL (lane +). Tubulin was used as
a loading control.
(G) Quantification of phagocytic efficiency in FLAG-Bcl10 cells incubated with siRNA ctrl or OCRL, after incubation with IgG-RBC for 10 min. The data are
expressed as the mean percentage of internalized RBC/total RBC from three independent experiments.
(H) Frustrated phagocytosis experiments were performed using transiently transfected Control (top) or Bcl10 KD (bottom) THP1 cells expressing both Lifeact-
mCherry and OCRLA-GFP. OCRLA-GFP and Lifeact-mCherry imaging by TIRFM were performed alternatively at 37?C every 480 ms during 120 s. Sequences of
images show the TIRF plan. Scale bar, 5 mm.
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 965
Phagocytosis assays were performed with adherent cells plated on glass
coverslips or with THP-1 cells plated onto poly-L-lysine-coated coverslips
before incubation with IgG-opsonized SRBC (IgG-RBC) (Braun et al., 2004).
For this, RBC were washed in PBS1X, incubated with anti-RBC antibodies
for 30 minutes at room temperature (RT), then washed, and resuspended in
serum-free medium. After internalization of the IgG-RBC for the indicated
times, cells were fixed in 4% PFA/4% sucrose for 10 minutes, and external
RBC were labeled for 10 minutes with Cy5-labeled F(ab’)2 anti-mouse or
anti-rabbit IgG in PBS/BSA 1%. Cells were then permeabilized with 0.05%
saponine before labeling of the intracellular RBC with AMCA-labeled F(ab’)2
anti-mouse or anti-rabbit IgG in PBS/saponine 0.05%/BSA 1%. To quantify
phagocytosis, the ratio: number of internalized RBC/number of [internalized+
bound] RBC was calculated in at least 50 cells randomly chosen on the cover-
slips, corresponding to the phagocytic index. The index obtained was divided
by the index obtained for control cells and expressed as a percentage of
control cells. A minimum of three independent experiments was performed.
We checked that control shRNA-depleted cells and parental nontransduced
THP-1 cells showed similar phagocytosis efficiencies (data not shown). Immu-
microscope (Leica DMI6000, Leica Microsystems, Wetzlar, Germany) with
a 1003 (1.4 NA) objective and a MicroMAX Princeton Instruments). Z-series
of images were taken at0.2 mm increments,and deconvolution wasperformed
with the software Huygens (Scientific Volume Imaging, Hilversum, The
disk confocal microscope equipped with a CoolSnap HQ2 (Photometrics,
Tucson, AZ, USA) camera.
Frustrated Phagocytosis Assay
Glass-bottom dishes of 35 mm in diameter (MatTek Corporation, El Segundo,
CA, USA) were coated with anti-RBC IgG in PBS overnight at 4?C and then
washed twice with PBS. Macrophages were resuspended in serum-free
microscopy medium (red phenol-free RPMI 1640, 2 mM L-Glutamine,
10 mM HEPES, 1 mM sodium pyruvate, and 50 mM b-mercaptoethanol) and
then allowed to sediment on the antibody-coated dishes at 37?C.
Phagosome Closure Assay
IgG-RBC was centrifuged onto 35 mm glass bottom dishes (MatTek Corpora-
tion) pretreated with 0.01% poly-L-Lysine in PBS for 30 minutes at RT.
The dishes were then washed once with a 10% BSA in PBS solution and incu-
bated with prewarmed serum-free microscopy medium. Macrophages were
resuspended and allowed to sediment onto opsonized SRBC-coated dishes
The statistical significance of the data was tested with an unpaired Student’s
t test. Differenceswereconsideredsignificant ifpvalues wereless than0.05 (*)
and 0.005 (**).
Supplemental information includes four figures, one movie, and Supplemental
Experimental Procedures and can be found with this article online at http://dx.
We thank Andre ´s Alcover for his support; Benedicte Capron and Catherine
Fabre (EFS Saint Vincent de Paul) for buffy coat supply; Jean-Franc ¸ois Alkom-
bre and his team (INRA, Center de Jouy-en-Josas) for collecting samples of
sheep blood; the 3P5 Proteomic facility of Paris Descartes University for
mass spectrometry;Ce ´line Loussertand Antonio Mucciolo from the EM facility
at the University of Lausanne; and Ste ´phanie Guadaginini from the EM facility
at Pasteur Institute, Paris. This work was supported by grants from Fondation
pour la Recherche Me ´dicale (FRM, INE20041102865), CNRS (ATIP Program),
Ville de Paris and Agence Nationale de la Recherche (ANR2011-BSVSE3-025)
to F.N. A.E. was supported by grants from Institut Pasteur, the CNRS, the
Agence Nationale de la Recherche (ANR ANR07-JCJC-0089), and the
Schlumberger Foundation for Education and Research. M.T. acknowledges
support from the Swiss National Science Foundation and the Swiss Cancer
League (Oncosuisse), as well as the foundations Leenaards, Pierre Mercier,
and Novartis Consumer Health. S.M. was supported by postdoctoral fellow-
ships from FRM and EMBO. J.M. was supported by a doctoral fellowship
Received: March 28, 2012
Revised: August 18, 2012
Accepted: September 21, 2012
Published online: November 12, 2012
3-kinase in the completion of macropinocytosis and phagocytosis by macro-
phages. J. Cell Biol. 135, 1249–1260.
Bajno, L., Peng, X.-R., Schreiber, A.D., Moore, H.-P., Trimble, W.S., and
Grinstein, S. (2000). Focal exocytosis of VAMP3-containing vesicles at sites
of phagosome formation. J. Cell Biol. 149, 697–706.
Beemiller, P., Hoppe, A.D., and Swanson, J.A. (2006). A phosphatidylinositol-
3-kinase-dependent signal transition regulates ARF1 and ARF6 during
Fcgamma receptor-mediated phagocytosis. PLoS Biol. 4, e162.
Beemiller, P., Zhang, Y., Mohan, S., Levinsohn, E., Gaeta, I., Hoppe, A.D., and
Swanson, J.A. (2010). A Cdc42 activation cycle coordinated by PI 3-kinase
during Fc receptor-mediated phagocytosis. Mol. Biol. Cell 21, 470–480.
Bi, L., Gojestani, S., Wu, W., Hsu, Y.M., Zhu, J., Ariizumi, K., and Lin, X. (2010).
CARD9 mediates dectin-2-induced IkappaBalpha kinase ubiquitination
leading to activation of NF-kappaB in response to stimulation by the hyphal
form of Candida albicans. J. Biol. Chem. 285, 25969–25977.
Bohdanowicz, M., Balkin, D.M., De Camilli, P., and Grinstein, S. (2012).
Recruitment of OCRL and Inpp5B to phagosomes by Rab5 and APPL1
depletes phosphoinositides and attenuates Akt signaling. Mol. Biol. Cell 23,
Braun, V., and Niedergang, F. (2006). Linking exocytosis and endocytosis
during phagocytosis. Biol. Cell 98, 195–201.
Braun, V., Fraisier, V., Raposo, G., Hurbain, I., Sibarita, J.B., Chavrier, P., Galli,
T., and Niedergang, F. (2004). TI-VAMP/VAMP7 is required for optimal phago-
cytosis of opsonised particles in macrophages. EMBO J. 23, 4166–4176.
Braun, V., Deschamps, C., Raposo, G., Benaroch, P., Benmerah, A., Chavrier,
P., and Niedergang, F. (2007). AP-1 and ARF1 control endosomal dynamics at
sites of FcR mediated phagocytosis. Mol. Biol. Cell 18, 4921–4931.
Caron, E., and Hall, A. (1998). Identification of two distinct mechanisms of
phagocytosis controlled by different Rho GTPases. Science 282, 1717–1721.
Chen, Y., Pappu, B.P., Zeng, H., Xue, L., Morris, S.W., Lin, X., Wen, R., and
Wang, D. (2007). B cell lymphoma 10 is essential for FcepsilonR-mediated
degranulation and IL-6 production in mast cells. J. Immunol. 178, 49–57.
Choudhury, R., Diao, A., Zhang, F., Eisenberg, E., Saint-Pol, A., Williams, C.,
Konstantakopoulos, A., Lucocq, J., Johannes, L., Rabouille, C., et al. (2005).
Lowe syndrome protein OCRL1 interacts with clathrin and regulates protein
trafficking between endosomes and the trans-Golgi network. Mol. Biol. Cell
Dambournet, D., Machicoane, M., Chesneau, L., Sachse, M., Rocancourt, M.,
El Marjou, A., Formstecher, E., Salomon, R., Goud, B., and Echard, A. (2011).
Rab35 GTPase and OCRL phosphatase remodel lipids and F-actin for
successful cytokinesis. Nat. Cell Biol. 13, 981–988.
Du, M.Q. (2007). MALT lymphoma : recent advances in aetiology and molec-
ular genetics. J. Clin. Exp. Hematop. 47, 31–42.
Erdmann, K.S., Mao, Y., McCrea, H.J., Zoncu, R., Lee, S., Paradise, S.,
Modregger, J., Biemesderfer, D., Toomre, D., and De Camilli, P. (2007). A
role of the Lowe syndrome protein OCRL in early steps of the endocytic
pathway. Dev. Cell 13, 377–390.
Faucherre, A., Desbois, P., Satre, V., Lunardi, J., Dorseuil, O., and Gacon, G.
(2003). Lowe syndrome protein OCRL1 interacts with Rac GTPase in the
trans-Golgi network. Hum. Mol. Genet. 12, 2449–2456.
Bcl10 Controls Endosome Traffic and Actin Dynamics
966 Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc.
Flannagan, R.S., Jaumouille ´, V., and Grinstein, S. (2012). The cell biology of Download full-text
phagocytosis. Annu. Rev. Pathol. 7, 61–98.
Goodridge, H.S., Shimada, T., Wolf, A.J., Hsu, Y.M., Becker, C.A., Lin, X., and
Underhill, D.M. (2009). Differential use of CARD9 by dectin-1 in macrophages
and dendritic cells. J. Immunol. 182, 1146–1154.
Gross, O., Gewies, A., Finger, K., Scha ¨fer, M., Sparwasser, T., Peschel, C.,
Fo ¨rster, I., and Ruland, J. (2006). Card9 controls a non-TLR signalling pathway
for innate anti-fungal immunity. Nature 442, 651–656.
Ha ¨cker, H., and Karin, M. (2006). Regulation and function of IKK and IKK-
related kinases. Sci. STKE 2006, re13.
Hara, H., and Saito, T. (2009). CARD9 versus CARMA1 in innate and adaptive
immunity. Trends Immunol. 30, 234–242.
Hara, H., Ishihara, C., Takeuchi, A., Imanishi, T., Xue, L., Morris, S.W., Inui, M.,
Takai, T., Shibuya, A., Saijo, S., et al. (2007). The adaptor protein CARD9 is
essential for the activation of myeloid cells through ITAM-associated and
Toll-like receptors. Nat. Immunol. 8, 619–629.
Hara, H., Ishihara, C., Takeuchi, A., Xue, L., Morris, S.W., Penninger, J.M.,
Yoshida, H., and Saito, T. (2008). Cell type-specific regulation of ITAM-
mediated NF-kappaB activation by the adaptors, CARMA1 and CARD9.
J. Immunol. 181, 918–930.
Hirst,J.,Motley,A.,Harasaki,K.,Peak Chew,S.Y., andRobinson,M.S.(2003).
EpsinR: an ENTH domain-containing protein that interacts with AP-1. Mol.
Biol. Cell 14, 625–641.
Klemm, S., Gutermuth, J., Hu ¨ltner, L., Sparwasser, T., Behrendt, H., Peschel,
C., Mak, T.W., Jakob, T., and Ruland, J. (2006). The Bcl10-Malt1 complex
segregates Fc epsilon RI-mediated nuclear factor kappa B activation and
cytokine production from mast cell degranulation. J. Exp. Med. 203, 337–347.
Loovers, H.M., Kortholt, A., de Groote, H., Whitty, L., Nussbaum, R.L., and van
Haastert, P.J. (2007). Regulation of phagocytosis in Dictyostelium by the
inositol 5-phosphatase OCRL homolog Dd5P4. Traffic 8, 618–628.
Malacombe, M., Bader, M.F., and Gasman, S. (2006). Exocytosis in neuroen-
docrine cells: new tasks for actin. Biochim. Biophys. Acta 1763, 1175–1183.
Massol, P., Montcourrier, P., Guillemot, J.-C., and Chavrier, P. (1998). Fc
receptor-mediated phagocytosis requires CDC42 and Rac1. EMBO J. 17,
Ngo,V.N., Davis, R.E., Lamy,L., Yu,X., Zhao, H., Lenz, G.,Lam, L.T., Dave, S.,
Yang, L., Powell, J., and Staudt, L.M. (2006). A loss-of-function RNA interfer-
ence screen for molecular targets in cancer. Nature 441, 106–110.
and AGAP2 distinguish between the adaptor protein complexes AP-1 and
AP-3. J. Cell Sci. 118, 3555–3566.
Niedergang, F., Colucci-Guyon, E., Dubois, T., Raposo, G., and Chavrier, P.
(2003). ADP ribosylation factor 6 is activated and controls membrane delivery
during phagocytosis in macrophages. J. Cell Biol. 161, 1143–1150.
Niedergang, F., and Chavrier, P. (2005). Regulation of phagocytosis by Rho
GTPases. Curr. Top. Microbiol. Immunol. 291, 43–60.
Popoff, V., Mardones, G.A., Tenza, D., Rojas, R., Lamaze, C., Bonifacino, J.S.,
Raposo, G., and Johannes, L. (2007). The retromer complex and clathrin
define an early endosomal retrograde exit site. J. Cell Sci. 120, 2022–2031.
Rawlings, D.J., Sommer, K., and Moreno-Garcı ´a, M.E. (2006). The CARMA1
signalosome links the signalling machinery of adaptive and innate immunity
in lymphocytes. Nat. Rev. Immunol. 6, 799–812.
Rosebeck, S., Rehman, A.O., Lucas, P.C., and McAllister-Lucas, L.M. (2011).
From MALT lymphoma to the CBM signalosome: three decades of discovery.
Cell Cycle 10, 2485–2496.
Rueda, D., Gaide, O., Ho, L., Lewkowicz, E., Niedergang, F., Hailfinger, S.,
Rebeaud, F., Guzzardi, M., Conne, B., Thelen, M., et al. (2007). Bcl10 controls
TCR- and FcgammaR-induced actin polymerization. J. Immunol. 178, 4373–
Ruland,J., Duncan, G.S., Elia, A., del Barco Barrantes, I., Nguyen,L., Plyte, S.,
Millar, D.G., Bouchard, D., Wakeham, A., Ohashi, P.S., and Mak, T.W. (2001).
Bcl10 is a positive regulator of antigen receptor-induced activation of
NF-kappaB and neural tube closure. Cell 104, 33–42.
Sarantis, H., Balkin, D.M., De Camilli, P., Isberg, R.R., Brumell, J.H., and
Grinstein, S. (2012). Yersinia entry into host cells requires Rab5-dependent
dephosphorylation of PI(4,5)P₂ and membrane scission. Cell Host Microbe
Scott, C.C., Dobson, W., Botelho, R.J., Coady-Osberg, N., Chavrier, P., Knecht,
D.A., Heath, C., Stahl, P., and Grinstein, S. (2005). Phosphatidylinositol-4,5-
bisphosphate hydrolysis directs actin remodeling during phagocytosis. J. Cell
Biol. 169, 139–149.
Shiba, Y., Ro ¨mer, W., Mardones, G.A., Burgos, P.V., Lamaze, C., and
Johannes, L. (2010). AGAP2 regulates retrograde transport between early
endosomes and the TGN. J. Cell Sci. 123, 2381–2390.
Strasser, D., Neumann, K., Bergmann, H., Marakalala, M.J., Guler, R.,
Rojowska, A., Hopfner, K.P., Brombacher, F., Urlaub, H., Baier, G., et al.
(2012). Syk kinase-coupled C-type lectin receptors engage protein kinase
C-s to elicit Card9 adaptor-mediated innate immunity. Immunity 36, 32–42.
Swanson, J.A. (2008). Shaping cups into phagosomes and macropinosomes.
Nat. Rev. Mol. Cell Biol. 9, 639–649.
Thome, M., and Weil, R. (2007). Post-translational modifications regulate
distinct functions of CARMA1 and BCL10. Trends Immunol. 28, 281–288.
Thome, M., Charton, J.E., Pelzer, C., and Hailfinger, S. (2010). Antigen
receptor signaling to NF-kappaB via CARMA1, BCL10, and MALT1. Cold
Spring Harb. Perspect. Biol. 2, a003004.
Touret,N., Paroutis, P.,and Grinstein, S. (2005). The nature of the phagosomal
membrane: endoplasmic reticulum versus plasmalemma. J. Leukoc. Biol. 77,
Uren, A.G., O’Rourke, K., Aravind, L.A., Pisabarro, M.T., Seshagiri, S., Koonin,
E.V., and Dixit, V.M. (2000). Identification of paracaspases and metacaspases:
two ancient families of caspase-like proteins, one of which plays a key role in
MALT lymphoma. Mol. Cell 6, 961–967.
Welch, H.C., Coadwell, W.J., Stephens, L.R., and Hawkins, P.T. (2003).
Phosphoinositide 3-kinase-dependent activation of Rac. FEBS Lett. 546,
Wu, W., Hsu, Y.M., Bi, L., Songyang, Z., and Lin, X. (2009). CARD9 facilitates
microbe-elicited production of reactive oxygen species by regulating the
LyGDI-Rac1 complex. Nat. Immunol. 10, 1208–1214.
Xue, L., Morris, S.W., Orihuela, C., Tuomanen, E., Cui, X., Wen, R., and Wang,
D. (2003). Defective development and function of Bcl10-deficient follicular,
marginal zone and B1 B cells. Nat. Immunol. 4, 857–865.
Zhang, X., Jefferson, A.B., Auethavekiat, V., and Majerus, P.W. (1995). The
protein deficient in Lowe syndrome is a phosphatidylinositol-4,5-bisphos-
phate 5-phosphatase. Proc. Natl. Acad. Sci. USA 92, 4853–4856.
Bcl10 Controls Endosome Traffic and Actin Dynamics
Developmental Cell 23, 954–967, November 13, 2012 ª2012 Elsevier Inc. 967