PHOSPHOINOSITIDE 3-KINASE c MEDIATES MICROGLIAL
PHAGOCYTOSIS VIA LIPID KINASE-INDEPENDENT CONTROL OF cAMP
C. SCHMIDT,a?N. SCHNEBLE,a?J. P. MU¨LLER,a
R. BAUER,a,eA. PERINO,bR. MARONE,c
S. D. RYBALKIN,dM. P. WYMANN,eE. HIRSCHbAND
aInstitute of Molecular Cell Biology, Center for Molecular
Biomedicine, Jena University Hospital, 07745 Jena, Germany
bMolecular Biotechnology Center, University of Torino, 10126 Torino,
cDepartment of Biomedicine, Institute of Biochemistry and
Genetics, University Basel, 4058 Basel, Switzerland
dDepartment of Pharmacology, University of Washington,
Seattle, WA 98195, USA
eCenter for Sepsis Control and Care, Jena University Hospital,
07747 Jena, Germany
Abstract—Microglial phagocytosis plays a key role in neuro-
protective and neurodegenerative responses of the innate
immune system in the brain. Here we investigated the regu-
latory function of phosphoinositide 3-kinase c (PI3Kc) in
phagocytosis of bacteria and Zymosan particles by mouse
brain microglia in vitro and in vivo. Using genetic and phar-
macological approaches our data revealed PI3Kc as an
essential mediator of microglial phagocytosis. Unexpect-
edly, microglia expressing lipid kinase deficient mutant
PI3Kc exhibited similar phagocytosis as wild-type cells.
These data suggest kinase-independent stimulation of
cAMP phosphodiesterase activity by PI3Kc as a crucial
mediator of phagocytosis. In sum our findings indicate
PI3Kc-dependent suppression of cAMP signaling as a criti-
? 2012 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: PI3K, PDE, cAMP, microglia, phagocytosis.
Microglia represent main constituents of the immune
system in the central nervous system. They fulfill typical
functions of innate immune cells including migration to
sites of injury, phagocytosis of microbes and cell debris as
well as release of specific cytokines (Kettenmann et al.,
neurodegeneration (Prinz et al., 2011). In addition to the
generationof neurotoxic mediators,
neurons was considered as a main cause of the
neurodegenerative activities of microglia (Neher et al.,
2012). The expanding understanding of the physiological
and pathophysiological capacity of microglial phagocytosis
contrasts with the limited knowledge on intracellular
mediators of their phagocytic activities.
demonstrated inhibitory effects of cAMP on microglial
phagocytic activities (Makranz et al., 2006; Orr et al.,
2009; Steininger et al., 2011). Both cAMP-dependent
stimulation of protein kinase A (PKA) or Epac have
been shown to convey these inhibitory effects.
candidate mediators of microglial phagocytosis due to their
proposed regulatory function in phagocytic reactions of
neutrophils and macrophages (Lee et al., 2007; Tamura
et al., 2009). Four species of the class 1 PI3K family
produce the second messenger phosphatidylinositol-3,4,5-
trisphosphate (PIP3), which controls a multitude of cellular
functions (Hawkins et al., 2006). The PI3K isoform PI3Kc
was originally characterized as a major mediator of G-
eponymous lipid kinase activity PI3Kc has been shown to
repress cAMP production in cardiomyocytes via stimulation
of cAMP phosphodiesterase (Patrucco et al., 2004).
Expression of PI3Kc was documented in immune cells and
cells of the cardiovascular system (Patrucco et al., 2004;
Fruman and Bismuth, 2009). Recent investigations located
expression of PI3Kc in the nervous system, specifically in
dorsal root ganglia (Konig et al., 2010; Cunha et al., 2010)
and microglia (Jin et al., 2010).
Our present study uncovered a critical function of PI3Kc
in phagocytic reactions of microglia. PI3Kc-dependent
control of cAMP production has been disclosed as a
crucial regulator of microglial phagocytosis.
in the and
In addition to its
Monoclonal p110c antibodies and polyclonal antibodies against
p84, p101 and PDE3B have been produced in our facilities in
Jena, Basel and Seattle. Other antibodies were obtained from
Cell Signaling (Danvers, USA): p110a, p110b, pAKT Ser473,
0306-4522/12 $36.00 ? 2012 IBRO. Published by Elsevier Ltd. All rights reserved.
*Corresponding author. Tel: +49-3641-9395600; fax: +49-3641-
E-mail address: Reinhard.Wetzker@uni-jena.de (R. Wetzker).
?These authors contributed equally to this work.
Abbreviations: DMEM, Dulbecco’s modified Eagle’s medium; EDTA,
fluorescein-isothiocyanate; KD, kinase-dead; KO, knockout; PBS,
phosphate-buffered saline; PDE, cAMP phosphodiesterase; PI3K,
phosphoinositide 3-kinase; PI3Kc, phosphoinositide 3-kinase c; PIP3,
phosphatidylinositol-3,4,5-trisphosphate; PKA, protein kinase A.
Neuroscience 233 (2013) 44–53
panAKT, pCREB Ser133, panCREB and PKA Ca, Upstate (Lake
Placid, USA): p110d and Sigma Aldrich (Steinheim, Germany): b-
For immunoprecipitation BV-2 cell lysates were cleared by
centrifugation at 13.000 rpm, 15 min at 4 ?C. Supernatants were
incubated with p110c monoclonal antibody and Protein-G-
Sepharose beads (GEHealthcare)
immunoprecipitate was analyzed by Western blotting.
for 24 h.The
The PI3Kc Inhibitor AS605240 was obtained from Alexis
(Lausen, Switzerland). The inhibitors TGX221 (PI3Kb) and
IC87114 (PI3Kd) were purchased from Baker Heart Research
Institute (Melbourne, Australia). Inhibitor A66 (PI3Ka) was
obtained from Symansis (Auckland, New Zealand). Wortmannin
as a general PI3K inhibitor was purchased from Sigma (St.
Louis, USA). Other inhibitors obtained from Sigma include
IBMX, rolipram and cilostamide. Forskolin, PTX and H89 were
obtained from Enzo Life Science (New York, USA).
BV-2 an immortalized murine microglia cell line was cultured at
37 ?C and 5% CO2 in Dulbecco’s modified Eagle’s medium
(DMEM) high glucose from PAA Laboratories (Dartmouth,
USA) containing 10% fetal calf serum (FCS), 1% penicillin/
streptomycin and 1% amphotericin B.
shRNA-mediated down regulation of gene expression
The generation of specific shRNA cell lines has been described
previously (Bekhite et al., 2011). Plasmid pLKO.1 vectors
encoding shRNA constructs targeting PI3K p110a, b, c or d
catalytic subunits and regulatory p84 and p101 subunits, or
plasmid pLKO.1 encoding a non-targeting control shRNA were
obtained from the Sigma–Aldrich (Taufkirchen, Germany). For
generationof lentiviral particles,
transfected with pLKO.1 derivative plasmids in combination
with pRev, pEnv-VSV-G, and pMDLg. BV-2 cells were infected
three times with the pseudotyped particles in the presence of
8 lg/ml polybrene. The transduced cell pools were selected
with 2 lg/ml puromycin 48 h post transduction for 10 days.
Down regulation of the catalytic and regulatory subunits of the
PI3K class I was tested by Western blot analysis.
PI3Kc knockout (KO) and PI3Kc kinase-dead (KD) mice
(Patrucco et al., 2004) were on the C57BL/6J background.
C57BL/6J wild-type mice were used as controls.
Primary microglia cells
The microglia cells were isolated from neonatal mouse cerebral
cortex as described (Giulian and Baker, 1986). The cells were
co-cultivated with astrocytes for 14 days at 37 ?C and 5% CO2
in DMEM high glucose containing 10% FCS, 1% penicillin/
streptomycin and 1% amphotericin B. After 14 days adherent
microglia were separated from astrocytes by adding PBS/EDTA
and careful shaking. After harvesting microglial cells were
seeded in well plates.
SDS PAGE and Western blotting
For quantification of protein expression and phosphorylation cells
were seeded into 6-well plates and incubated at 37 ?C (5% CO2).
After becoming adherent, cells were incubated over night in
DMEM medium w/o FCS and treated for 24 h with agonists or
inhibitors. Thereafter cells were suspended in RIPA lysis buffer
composed of 50 mM Tris/HCl pH 8, 150 mM NaCl, 1% (v/v)
NP-40, 0.5% (v/v) deoxycholate, 0.1% (w/v) SDS, 100 lg/ml
Pefa-Block, 1 lg/ml pepstatin, 10 lM sodium vanadate and
1 lg/ml leupeptin. Referring to a 20-min centrifugation at
13500g, 4 ?C, supernatants were mixed with 5? protein sample
buffer (5% SDS, 33% glycerol, 25% b-mercaptoethanol) and
heated to 95 ?C for 5 min. Protein samples were separated on
10% polyacrylamide-Gel, transferred to a polyvinylidenfluorid
followed by enhanced chemiluminescence reaction.
Microglial cells were seeded in 96-well clear-bottom plates and
treated with different inhibitors. cAMP was measured following
the manufacturer’s protocol (Promega, cAMP GloAssayKit).
Phagocytosis assay in vitro
Efficiency of phagocytosis was investigated as described (Sun
et al., 2008). Primary microglia cells and BV-2 cells were
seeded into 6-well plates and incubated at 37 ?C (5% CO2) for
24 h. After attachment cells were starved for 24 h in DMEM
without FCS. Phagocytosis assay was performed using GFP-
producing Escherichia coli. Forty microliters of the suspended
bacteria was added to the microglial cells and incubated 1 h at
37 ?C (5% CO2). After incubation the cells were harvested,
washed and re-suspended in PBS. The phagocytic activity of
the cells was measured by flow cytometry using FACS Canto
(BD, Heidelberg, Germany).
Phagocytosis assay in vivo
Experiments were performed on adult PI3Kc KO and wild-type
mice (5 mice per group, weighing 19–24 g). Animal procedures
were approved by the committee of the Thuringian State
Government on Animal Research. Mice were anesthetized by
intraperitoneal injection of a cocktail comprising midazolam
(5 mg/kg), fentanyl (0.05 mg/kg), and medetomidin (0.5 mg/kg).
Kainate (0.21 lg in 50 nl saline) was stereotaxically injected
into the dorsal hippocampus on each side with a 0.5-ll
microsyringe over 60 s. Stereotaxic coordinates were AP,
?1.6 mm; L, ±1.6 mm; and V, ?1.5 mm, respectively (Paxinos
and Franklin, 2001). Subsequently, anesthesia was terminated
by intraperitoneal injection of atipamezol (2.5 mg/kg), naloxon
(1.2 mg/kg), and flumazenil (0.5 mg/kg). After antagonization of
anesthesia animals woke up within 1–2 min and developed
signs of nonconvulsive seizures as shown by clonic forepaw
movements and chewing lasting several hours. Thereby, an
intensified microglial cell activation is induced by increased
expression of inflammatory mediators in the hippocampus,
followed within 24–48 h by microglial proliferation, and a
modification of microglia morphology characterized by larger
somata and thicker primary processes (Avignone et al., 2008).
Furthermore, microglial cells express the metabotropic P2Y6
receptor whose activation by endogenous agonist UDP triggers
microglial phagocytosis (Koizumi et al., 2007). Accordingly,
24 h later mice were re-anesthetized and 50 nl of fluorescein-
isothiocyanate (FITC)-labeled Zymosan particles (9800 per ll)
were stereotaxically (identical stereotaxic coordinates) injected
into the dorsal hippocampus. Twenty-four hours later mice were
deeply anesthetized and perfused with 4% paraformaldehyde in
phosphate buffer by cardiac puncture via the left ventricle.
Brains were removed immediately after fixation, post-fixated in
4% paraformaldehyde at 4 ?C for 1 day and embedded in
paraffin. Referring to Koizumi et al. (2007), for quantitative
immunohistochemical analysis of in vivo phagocytosis, coronal
sections (6 lm) of the hippocampus containing the FITC-
C. Schmidt et al./Neuroscience 233 (2013) 44–53
labeled Zymosan particles-injected sites were thoroughly washed
with 0.3% Triton X-100 containing PBS (shaking for 10 min, 10
times) to remove non-specific binding of the microspheres.
Chosen thicknessof coronal
phagocytized fluorescent Zymosan particles. Sections were
then stained with anti-Iba1 antibody using the antigen retrieval
(microwave, 750 W, 10 min, 0.01 mol/l citrate buffer) and the
avidin–biotin–peroxidase complex methods (Vectastain ABC
staining kit, Vector Labs) to visualize microglia. Evaluation of
phagocytosis of the fluorescent particles by microglial cells was
performed undera light microscope
Localization of labeled Zymosan was performed by merged
pictures obtained by bright-field imaging and fluorescence
microscopy. The total number of microglial cells incorporating
fluorescent particles in relation to the total number of Iba1-
positive cells was counted and used as an index of in vivo
phagocytosis. Three sections containing the microspheres-
injected sites per each animal were analyzed, and at least four
animals per each group were used. In each case the slice with
the highest number of FITC-labeled Zymosan particles was
chosen together with a rostral as well as a caudal slice in a
distance of 48 lm in order to warrant that every phagocytizing
microglial cell has been counted just for one time. In each slice
between 186 and 223 microglial cells were scored.
(6 lm) warrants
Data are reported as means ± standard error of the mean
(SEM). Comparisons between groups were made with one-way
analysis of variance (ANOVA). Post hoc comparisons were
made with the Holm–Sidak test. Comparisons between groups
were made with unpaired t tests using Bonferroni correction for
multiple uses, if appropriate. Differences were considered
significant when P < 0.05.
PI3Kc mediates microglial phagocytosis
To explore the role of PI3K in the control of phagocytosis
microglia were isolated from mouse brain and treated with
inhibitors specifically inhibiting the lipid kinase activities of
the PI3K species PI3Ka (A66), PI3Kb (TGX221), PI3Kc
(AS605240) and PI3Kd (IC87114) and incubated with
GFP producing E. coli. Only AS605240 suppressed
uptake of bacteria indicating PI3Kc as a mediator of
phagocytic activities of microglia (Fig. 1A). The uptake
of GFP-labeled E. coli by primary microglia has been
further characterized by kinetic analysis (Fig. 1B, C).
Fig. 1. PI3Kc mediates phagocytosis of E. coli by primary microglia. (A) Microglia isolated from mouse brain were pretreated for 60 min with
inhibitors of the PI3K species PI3Ka [30 nM A66], PI3Kb [20 nM TGX221], PI3Kc [1 lM AS605240] and PI3Kd [200 nM IC87114] and phagocytosis
was assayed by FACS after 60-min incubation. n = 4,⁄⁄⁄P < 0.001, compared to control. (B, C) Time-dependent uptake of GFP-producing E. coli
by primary microglia. Microglia isolated from mouse brain were incubated with GFP-producing E. coli particles for the indicated periods and
phagocytic active cells were quantified using by flow cytometry. (D) Pulse chase experiment. Phagocytosis of GFP-producing E. coli was quenched
after 60 min by addition of a 80-fold excess of unlabeled bacteria and GFP fluorescence was monitored for indicated time points using flow
46C. Schmidt et al./Neuroscience 233 (2013) 44–53
The pulse chase experiment shown in Fig. 1D reveals
absence of significant degradation of the GFP-labeled
E. coli or unspecific attachment of bacteria to the
To verify pharmacological indications for an important
role of PI3Kc in microglia phagocytosis siRNA technology
has been used. The BV-2 mouse microglia cell line was
genes specifically targeting the PI3K species a, b, c,
and d. As shown in Fig. 2A solely the decline of PI3Kc
phagocytosis of E. coli by BV-2 cells. These data
phagocytosis. Down regulation of PI3Ka, PI3Kb and
PI3Kd did not affect the phagocytic activity. As has been
shown in Fig. 2B expression of the catalytic subunits of
all PI3K species assayed by Western blotting was
We next compared phagocytosis of microglia from
wild-type and PI3Kc-deficient mice in vitro and in vivo.
As shown in Fig. 3A, B and the histograms in Fig. 3C,
isolated wild-type microglia expressed significantly a
higher uptake of E. coli. Similar results were obtained
using FITC Zymosan particles for investigating PI3Kc
dependence of microglial phagocytosis (Kettenmann
et al., 2011) (data not shown).
To investigate the phagocytosis in vivo FITC-labeled
hippocampus of adult PI3Kc KO and wild-type mice
after microglial cell activation by focal seizure induction
bya reduction of
injectedinto the dorsal
(Groticke et al., 2008; Koizumi et al., 2007) (Fig. 4A). As
shown in Fig. 4B microglial phagocytosis in the brain of
KOmice is significantly
comparison to wild-type brain. Taken together these
data disclose PI3Kc as a crucial mediator of microglial
Signaling reactions of PI3Kc mediating microglial
Next we asked for specific signaling functions of PI3Kc
involved in microglial phagocytosis. PI3Kc exists in
different cell types as a dimer including a p84 or p101
regulatory subunits associated with the catalytic subunit
p110c and both regulatory subunits were demonstrated
to exert stimulatory effects on PI3Kc lipid kinase activity
(Bohnacker et al., 2009). In the microglia cell line BV-2
decreased expression of p84 or p101 induced by
specific shRNA did not affect microglial phagocytosis in
contrast to suppression of the catalytic subunit p110c
(Fig. 5A). In these experiments shRNA knock down led
to about 80% decline of the regulatory subunits as
assayed by Western blotting (Fig. 5B). These data
suggest minor importance of p84 and p101 in the
control of phagocytic reactions of microglia.
In addition to its lipid kinase activity PI3Kc was shown
to induce stimulation of cAMP phosphodiesterase 3B
(PDE3B) via a lipid kinase-independent direct interaction
in cardiomyocytes (Patrucco et al., 2004). Regulatory
functions of this scaffold protein activity of PI3Kc have
Fig. 2. PI3Kc mediates phagocytosis of E. coli by microglial BV-2 cells. (A) Bacteria were added to BV-2 cell lines stably expressing shRNA genes
targeting p110a p110b p110c p110d or non-targeting control. Phagocytosis was measured by flow cytometry. n = 3,⁄⁄⁄P < 0.001, compared to
control. (B) shRNA effects on expression levels of PI3K species or regulatory subunits.
C. Schmidt et al./Neuroscience 233 (2013) 44–53
Fig. 3. PI3Kc mediates phagocytosis of E. coli by primary microglia. (A) Representative confocal images of E. coli phagocytosis by primary
microglia isolated from wild-type and PI3Kc-deficient mice. Microglia were stained with WGA-wheat germ agglutinin (red). The visualized bacteria
(green) are intracellular (merged image). (B) Phagocytosis of E. coli by PI3Kc KO microglia in comparison to wild-type cells. The total number of
microglial cells incorporating fluorescent bacteria in relation to the absolute number of microglial cells was counted and used as an index of in vitro
phagocytosis. n = 3, each group;
demonstrate GFP fluorescence before (red) and after phagocytosis for 60 min (green) for wild-type and PI3Kc KO primary microglia.
⁄P < 0.05, compared to wild-type microglia. (C) Flow cytometric analysis of phagocytosis. Histograms
48 C. Schmidt et al./Neuroscience 233 (2013) 44–53
been explored by investigating of the phenotype of
cardiomyocytes isolated from transgenic mice, which
express a lipid kinase inactive version of the catalytic
comparative analysis of microglia isolated from PI3Kc
KO mice with microglia produced from mice expressing
potential role of the scaffold function of PI3Kc in
expressing the lipid kinase deficient mutant of PI3Kc
revealed similar phagocytic activity as wild-type cells
(Fig. 6). In contrast, PI3Kc KO microglia exhibited low
phagocytic activity. Since stimulation of PDE activity
represents the only known signaling activity shared by
wild-type PI3Kc and catalytically inactive KD mutant
these data suggest a major role of the cAMP repressing
function of PI3Kc in the control of phagocytosis.
inmutant). We used
cAMP-dependent inhibition of microglial
Tracing these data to PI3Kc controlled regulatory function
ofcAMP wefurther investigated
cAMP-dependent PKA or Epac activities on microglial
adenylate cyclase and treatment with the inhibitors of
cAMP phosphodiesterases IBMX, rolipram (PDE4) and
cilostamide (PDE3B) or addition of the agonists of PKA
(6-BenzcAMP) or Epac
produced uniformly inhibition of microglial phagocytosis
down to the level of PI3Kc KO microglia (Fig. 7A). In
line with these data treatment of microglia with agonists
of the Gs-coupled A2A adenosine receptor as well as
CGS-21680 suppressed the phagocytic reaction of
Fig. 4. Phagocytosis of FITC-labeled Zymosan particles by microglia in mouse brain. (A) Representative images of phagocytosis of FITC Zymosan
particles by microglia in mouse brain from wild-type (upper panel) and PI3Kc-deficient mice (lower panel). The left-sided image section shows the
distribution of FITC Zymosan particles (green), the intermediate image section shows Iba1-positive (brownish-stained) cells and the right-sided
image section presents the merged picture. Note the intracellular localization of FITC-labeled Zymosan within Iba1-positive (brownish-stained) cells,
indicated by white arrows. (B) Phagocytosis of FITC Zymosan particles in the brain of PI3Kc KO mice in comparison to wild-type brain. The total
number of microglial cells incorporating fluorescent particles in relation to the total number of microglial cells was counted and used as an index of
in vivo phagocytosis. n = 5, each group;⁄P < 0.05, compared to wild-type mice.
C. Schmidt et al./Neuroscience 233 (2013) 44–53
microglia again down to the level of phagocytosis by
PI3Kc KO microglia (Fig. 7B). Conversely, the PKA
inhibitor H89stimulated phagocytosis
microglia and rescued the inhibitory effects of A2A
receptor agonists. The incomplete rescue effect of H89
in the presence of CGS-21680 or adenosine may
indicate significant contribution of Epac to the cAMP-
dependent inhibitory effect on phagocytosis. Taken
together, these data specify cAMP as a mediator
blocking microglial phagocytosis. Both cAMP-dependent
signaling proteins PKA and Epac may be involved in
suppression of the phagocytic reaction. Collectively our
data are in accordance with recent reports on cAMP
effects on phagocytosis in microglia and macrophages
(Steininger et al., 2011; Orr et al., 2009; Gosain et al.,
2007), which reveal inhibition of microglial phagocytosis
by specific stimulation of cAMP signaling.
Next we sought to directly assess specific regulatory
effects of PI3Kc on cAMP signaling in microglia. cAMP
levels were measured in microglia isolated from brain of
wild-type, PI3Kc KO or PI3Kc KD mice. The data shown
in Fig. 8A consistently revealed strongly increased cAMP
levels in PI3Kc KO cells. In contrast, microglia expressing
the catalytically inactive PI3Kc KD mutant exhibited low
cAMP level, which was comparable to wild-type cells. A
similar relation was observed if cAMP levels in these
microglia were measured by assaying phosphorylation of
the cAMP response element-binding transcription factor
CREB (Fig. 8B). These data suggest PI3Kc-dependent
stimulation of cAMP phosphodiesterase as a critical
mediator of cAMP signaling in microglia.
Stimulatory effects of PI3Kc on cAMP phosphodiesterase
activity and direct binding to PDE3B have been identified
in cardiomyocytes (Patrucco et al., 2004). To assess the
possible role of this mechanism for cAMP-mediated
PI3Kc from microglia and examined associated proteins
by Western blotting (Fig. 8C). Indeed PDE3B could be
detected in the immunoprecipitate favoring this PDE
isoform as the relevant interaction partner for the
inhibitory effects of PI3Kc on cAMP level in microglia.
The dominant regulatory effects of PI3Kc on cAMP
controlled phagocytosis seem contradictory to the strong
suppression of microglial phagocytosis by inhibitors of
Fig. 5. Role of PI3Kc adaptor proteins p84 and p101 in microglial phagocytosis. (A) Phagocytosis of E. coli by BV-2 cells. Suspended bacteria were
added to BV-2 cell lines stably expressing shRNA genes targeting p84, p101 or catalytic subunit p110c and phagocytic activity was measured by
flow cytometry. n = 6,⁄⁄⁄P < 0.001, compared to control. (B) shRNA-induced suppression of regulatory subunits p84 and p101 of PI3Kc in BV-2
Fig. 6. Role of PI3Kc lipid kinase activity in microglial phagocytosis.
Phagocytosis of E. coli by primary microglia. Microglia isolated from
mouse brain of wild type, PI3Kc KO and lipid kinase-deficient PI3Kc
KD mutant were incubated with E. coli and phagocytosis was
assayed by FACS. n = 4,⁄⁄⁄P < 0.001, compared to control.
50C. Schmidt et al./Neuroscience 233 (2013) 44–53
PI3Kc lipid kinase activity shown in Fig. 1A. Thus we
AS605240 and wortmannin on cAMP level. Intriguingly,
our investigations revealed strong stimulatory effects of
both PI3K inhibitors on cAMP level in wild-type microglia
and cells expressing the lipid kinase inactive KD mutant
Taken together, these data indicate control of cAMP
level by direct interaction of PI3Kc with PDE3B as a
main mediator of microglial phagocytosis. Functional
relevance of unknown other protein interactions of
PI3Kc cannot be excluded.
microglial phagocytic activities for their physiological and
pathological functions (Kettenmann et al., 2011; Neher
et al., 2012). Whereas a whole series of extracellular
phagocytosis-promoting agonists and receptors have
intracellular mediators of microglial phagocytosis has
remained very limited.
Here we describe a key function of the signaling
protein PI3Kc in the control of microglial phagocytosis.
Employing pharmacological and genetic approaches our
results unveiled PI3Kc as an essential mediator of the
phagocytic uptake of E. coli bacteria and FITC Zymosan
particles in the in vitro and in vivo models used. These
results seem in line with recent reports on an essential
studiesrevealed particularimportance of
role of the PI3K signaling in phagocytosis of the
monocytic cell line THP-1 (Lee et al., 2007) or the
macrophage cell line RAW.264.7 (Tamura et al., 2009).
Using siRNA approaches it was shown that knock down
suppression of the PI3K antagonists PTEN or SHIP-1
(Tamura et al., 2009) provoked stimulatory effects on
the phagocytic activities. Both authors concluded that
phosphatidylinositol (3,4,5)-trisphosphate the product of
PI3K lipid kinase activity functioned as an essential
mediator of phagocytic activities of monocytes and
In contrast to these literature data, our results indicate
stimulatory effects of PI3Kc on PDE3B activity followed
by a decline of cAMP level as crucial mediator of
microglial phagocytosis. The data disclose PI3Kc as a
novel player in cAMP signaling in the innate immune
system and corroborate recent reports on inhibitory
functions of cAMP and its downstream mediators PKA
and Epac on microglial phagocytosis (Lee et al., 2007;
Tamura et al., 2009). Analogous functions of the cAMP
repressing function of PI3Kc in other immune cells can
Intriguingly the repressing effects of PI3K inhibitors
wortmannin and AS605240 on microglial phagocytosis
were accompanied by an increase of cAMP level. These
data suggest indirect control of cAMP by PI3K lipid
kinase in microglia. Indeed, stimulatory effects of PI3K
on PDE3B activity have been proposed in adipocytes
Fig. 7. Role of cAMP, PKA or Epac in phagocytosis by wild-type and PI3Kc KO microglia. (A) Effects of pharmacological modulation of cAMP, PKA
or Epac activity on E. coli phagocytosis by wt, or PI3Kc KO primary microglia. Cells were treated with 100 lM forskolin (activator of adenylate
cyclase), 500 lM IBMX (pan-PDE inhibitor), 10 lM rolipram (PDE4 inhibitor), 10 lM cilostamide (PDE3B inhibitor), 2 mM 6-Benz cAMP (agonist of
PKA) or 2 mM 8-CPT-20-O-Me-cAMP (agonist of Epac). Phagocytosis of E. coli was assayed by flow cytometry.
⁄⁄⁄P < 0.001, compared to wild-type control. (B) Effects of adenosine A2Areceptor agonists adenosine (50 lM) and CGS-21680 (50 lM) on
phagocytosis by wild-type or KO primary microglia and rescue with PKA inhibitor H89 (10 lM). After treatment with the agonists and/or PKA inhibitor
cells were incubated with E. coli and phagocytosis was assayed by flow cytometry.⁄P < 0.05,⁄⁄P < 0.01,⁄⁄⁄P < 0.001, compared to wild type
without (w/o) H89 administration.
⁄P < 0.05,
⁄⁄P < 0.01,
C. Schmidt et al./Neuroscience 233 (2013) 44–53
(Baragli et al., 2011; Degerman et al., 2011). In line with
these results PI3K inhibition in microglia may lead to
suppression of PDE3B activity by preventing PKB/Akt-
increasing cAMP will possibly repress phagocytosis.
Following these results PI3K inhibitors seem to be
phagocytosisof viable neurons,
observed during inflammatory processes in the central
nervous system (Neher et al., 2012). If this could be
established, inhibitors of PI3K and also PDE3B might be
used as potential setscrews for the pharmacological
treatment of neurodegenerative diseases.
Our data introduce the signaling protein PI3K c as a key
mediator of microglial phagocytosis. In contrast to
recent reports on regulatory functions of PI3K in
phagocytosis of macrophages our data suggest lipid
kinase-independent stimulation of PDE activity by PI3Kc
microglia. These results implement the suppressive
effect of PI3Kc on cAMP as a mediator of immune cell
phagocytic activities of
Acknowledgement—This work was supported by DFG Research
Grant RTG 1715.
Avignone E, Ulmann L, Levavasseur F, Rassendren F, Audinat E
(2008) Status epilepticus induces a particular microglial activation
state characterized by enhanced purinergic signaling. J Neurosci
Baragli A, Ghe C, Arnoletti E, Granata R, Ghigo E, Muccioli G (2011)
Acylated and unacylated ghrelin attenuate isoproterenol-induced
lipolysis in isolated rat visceral adipocytes through activation of
phosphoinositide 3-kinase gamma and phosphodiesterase 3B.
Biochim Biophys Acta 1811:386–396.
Bekhite MM, Finkensieper A, Binas S, Mu ¨ ller J, Wetzker R, Figulla
HR, Sauer H, Wartenberg M (2011) VEGF-mediated PI3K class
IA and PKC signaling in cardiomyogenesis and vasculogenesis of
mouse embryonic stem cells. J Cell Sci 124:1819–1830.
Bohnacker T, Marone R, Collmann E, Calvez R, Hirsch E, Wymann
MP (2009) PI3Kgamma adaptor subunits define coupling to
degranulation and cell motility by distinct PtdIns(3,4,5)P3 pools
in mast cells. Sci Signal 2:ra27.
Cunha TM, Roman-Campos D, Lotufo CM, Duarte HL, Souza GR,
Verri Jr WA, Funez MI, Dias QM, Schivo IR, Domingues AC,
Sachs D, Chiavegatto S, Teixeira MM, Hothersall JS, Cruz JS,
Cunha FQ, Ferreira SH (2010) Morphine peripheral analgesia
depends on activation of the PI3Kgamma/AKT/nNOS/NO/KATP
signaling pathway. Proc Natl Acad Sci U S A 107:4442–4447.
Fig. 8. Effects of PI3Kc on cAMP level in microglia. (A) cAMP level in primary microglia expressing wild-type, KO and lipid kinase-deficient KD
mutant of PI3Kc. Microglial cells were seeded in 96-well plates and cAMP was measured using cAMP GloAssayKit. n = 3.⁄⁄P < 0.01, compared to
wild-type microglial cells. Mean cAMP level in wild-type microglia cells (100%) corresponds to 127 pmol per 1 lg protein. (B) cAMP-dependent
CREB phosphorylation in wt, KO and KD primary microglia. CREB phosphorylation has been assayed by Western blotting using specific pCREB
antibodies. A representative blot is shown. (C) Interaction of PI3Kc with PDE3B in BV-2 cells. Co-immunoprecipitation of p110c with PDE3B. (D)
cAMP level in wild-type, KO and KD primary microglia. Effects of AS605240 (1 lM) or wortmannin (100 nM). Microglial cells were seeded in 96-well
plates and cAMP was measured. n = 3.⁄P < 0.05,⁄⁄P < 0.01,⁄⁄⁄P < 0.001, compared to wild-type control (mean 127 pmol cAMP per 1 lg
52 C. Schmidt et al./Neuroscience 233 (2013) 44–53
Degerman E, Ahmad F, Chung YW, Guirguis E, Omar B, Stenson L, Download full-text
Manganiello V (2011) From PDE3B to the regulation of energy
homeostasis. Curr Opin Pharmacol 11:676–682.
Franklin KBJ, Paxinos G (2001) The mouse brain in sterotaxic
coordinates. 2nd ed. Orlando, Fl, USA: Academic Press.
Fruman DA, Bismuth G (2009) Fine tuning the immune response with
PI3K. Immunol Rev 228:253–272.
Giulian D, Baker TJ (1986) Characterization of ameboid microglia
isolated from developing mammalian brain. J Neurosci 6:2163–2178.
Gosain A, Muthu K, Gamelli RL, DiPietro LA (2007) Norepinephrine
suppresses wound macrophage phagocytic efficiency through
alpha- and beta-adrenoreceptor dependent pathways. Surgery
Groticke I, Hoffmann K, Loscher W (2008) Behavioral alterations in a
mouse modelof temporal
intrahippocampal injection of kainate. Exp Neurol 213:71–83.
Hawkins PT, Anderson KE, Davidson K, Stephens LR (2006)
Signalling through class I PI3Ks in mammalian cells. Biochem
Soc Trans 34:647–662.
Jin R, Yu S, Song Z, Quillin JW, Deasis DP, Penninger JM, Nanda A,
Granger DN, Li G (2010) Phosphoinositide 3-kinase-gamma
expression is upregulated in brain microglia and contributes to
ischemia-induced microglial activation in acute experimental
stroke. Biochem Biophys Res Commun 399:458–464.
Kettenmann H, Hanisch UK, Noda M, Verkhratsky A (2011)
Physiology of microglia. Physiol Rev 91:461–553.
Koizumi S, Shigemoto-Mogami Y, Nasu-Tada K, Shinozaki Y,
Ohsawa K, Tsuda M, Joshi BV, Jacobson KA, Kohsaka S,
Inoue K (2007) UDP acting at P2Y6 receptors is a mediator of
microglial phagocytosis. Nature 446:1091–1095.
Ko ¨ nig C, Gavrilova-Ruch O, von Banchet GS, Bauer R, Grun M,
Hirsch E, Rubio I, Schulz S, Heinemann SH, Schaible HG,
phosphoinositide 3-kinase gamma. Neuroscience 169:449–454.
Lee JS, Nauseef WM, Moeenrezakhanlou A, Sly LM, Noubir S, Leidal
KG, Schlomann JM, Krystal G, Reiner NE (2007) Monocyte
p110alpha phosphatidylinositol 3-kinase regulates phagocytosis,
the phagocyte oxidase, and cytokine production. J Leukoc Biol
Makranz C, Cohen G, Reichert F, Kodama T, Rotshenker S (2006)
phosphodiesterases) regulates myelin phagocytosis mediated
by complement receptor-3 and scavenger receptor-AI/II in
microglia and macrophages. Glia 53:441–448.
Neher JJ, Neniskyte U, Brown GC (2012) Primary phagocytosis of
neuronsby inflamed microglia:
neurodegeneration. Front Pharmacol 3:27.
Orr AG, Orr AL, Li XJ, Gross RE, Traynelis SF (2009) Adenosine
A(2A) receptor mediates microglial process retraction. Nat
Patrucco E, Notte A, Barberis L, Selvetella G, Maffei A, Brancaccio
M, Marengo S, Russo G, Azzolino O, Rybalkin SD, Silengo L,
Altruda F, Wetzker R, Wymann MP, Lembo G, Hirsch E (2004)
PI3Kgamma modulates the cardiac response to chronic pressure
overload by distinct kinase-dependent and -independent effects.
Prinz M, Priller J, Sisodia SS, Ransohoff RM (2011) Heterogeneity of
CNS myeloid cells and their roles in neurodegeneration. Nat
Steininger TS, Stutz H, Kerschbaum HH (2011) Beta-adrenergic
stimulation suppresses phagocytosis via Epac activation in
murine microglial cells. Brain Res 1407:1–12.
Sun HN, Kim SU, Lee MS, Kim SK, Kim JM, Yim M, Yu DY, Lee DS
(2008) Nicotinamide adenine dinucleotide phosphate (NADPH)
oxidase-dependent activation of phosphoinositide 3-kinase and
p38 mitogen-activated protein kinase signal pathways is required
Pharm Bull 31:1711–1715.
Tamura N, Hazeki K, Okazaki N, Kametani Y, Murakami H, Takaba
Y, Ishikawa Y, Nigorikawa K, Hazeki O (2009) Specific role of
phagocytosis and pinocytosis in macrophages. Biochem J
adenylyl cyclase,Gi, and
(Accepted 19 December 2012)
(Available online 29 December 2012)
C. Schmidt et al./Neuroscience 233 (2013) 44–53