Cardiovascular Research (2014) 104, 467–476
doi:10.1093/cvr/cvu213
PKCd is dispensible for oxLDL uptake and foam
cell formation by human and murine macrophages
Katka Szilagyi 1*, Alexander B. Meijer2, Annette E. Neele 1, Paul Verkuijlen 1,
Michael Leitges 3, Sandrine Dabernat 4, Elisabeth Förster-Waldl5, Kaan Boztug 5,6,
Alexandre Belot 7, Taco W. Kuijpers 1,8, Georg Kraal 9, Menno P. J. de Winther 10, and
Timo K. van den Berg 1
1
Department of Blood Cell Research, Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Plesmanlaan 125, Amsterdam 1066CX,
The Netherlands; 2Department of Plasma Proteins, Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands;
3
The Biotechnology Centre of Oslo, University of Oslo, Oslo, Norway; 4INSERM U1035, Université Bordeaux Segalen, Bordeaux, France; 5Divison of Neonatology, Paediatric Intensive Care
and Neuropaediatrics, Department of Paediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; 6CeMM Research Center for Molecular Medicine of the Austrian
Academy of Sciences, Vienna, Austria; 7Hôpital Femme Mère Enfant, Hospices Civils de Lyon and Université de Lyon, Lyon, France; 8Department of Pediatric Hematology, Immunology and
Infectious Disease, Emma Children’s Hospital, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; 9Department of Molecular Cell Biology and Immunology, VU
University Medical Center, Amsterdam, The Netherlands; and 10Department of Medical Biochemistry, Experimental Vascular Biology, Academic Medical Center, University of Amsterdam,
Amsterdam, The Netherlands
Received 9 May 2014; revised 10 September 2014; accepted 17 September 2014; online publish-ahead-of-print 24 September 2014
Time for primary review: 37 days
Aims
Uptake of oxidized lipoprotein particles (oxLDL) and foam cell formation by macrophages is one of the first steps in the
development of atherosclerosis. Recently, protein kinase C d (PKCd) has been implicated as a regulator of oxLDL uptake
and foam cell formation via down-regulation of PKCb and scavenger receptors CD36 and SR-A expression. Here, we
describe studies in which we have re-evaluated the role of PKCd in oxLDL uptake and foam cell formation.
.....................................................................................................................................................................................
Methods
PKCd expression was silenced in the human monocytic cell lines and also in primary human monocytes to analyse oxLDL
uptake and CD36 expression. Additionally, bone marrow-derived macrophages of PKCd knockout mice and macroand results
phages cultured from patients with rare null mutations in the PRKCD gene were tested for uptake of oxLDL and foam
cell formation. Expression of scavenger receptor CD36 was determined and levels of PKCb isoforms were quantified.
Neither a reduction in PKCd levels nor its complete absence resulted in a detectable effect on the uptake of oxLDL
and the formation of foam cells.
.....................................................................................................................................................................................
Conclusion
PKCd is dispensible for oxLDL uptake and foam cell formation by monocytes and macrophages.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Protein kinase Cd † OxLDL uptake † Foam cell † PKCd deficiency
1. Introduction
An early and central event leading to the development of atherosclerotic
lesions is the formation of foam cells in the intima of large blood vessels.
Foam cells develop from resident macrophages and newly recruited
monocytes by an intracellular accumulation of oxidized lipoprotein
particles (oxLDL) acquired from the plasma. Recognition and uptake
of oxLDL occurs predominantly through scavenger receptors such as
CD36 and SR-A.1
Although the (early) signalling events that regulate the uptake and
processing of oxLDL in developing foam cells remain largely unknown,
it is clear that the relevant signalling components comprise promising
targets for therapeutic strategies in atherosclerosis as this may prevent
further lipid accumulation and foam cell formation. For instance, the
endocytosis of lipoprotein particles involves the phosphorylation of
various proteins by kinases,2 the targeting of which could form an attractive approach to treat atherosclerosis. In particular, a role for
several members of the protein kinase C (PKC) family, among which
representatives of classical (PKCa, bI, bII, and g) and novel (PKCd, 1,
u, and h) members, have been already suggested in the uptake of
oxLDL. Using human macrophages, the role of PKCb in oxLDL
uptake has been investigated in vitro, showing that decreased PKCb expression or inhibition with a specific inhibitor reduces oxLDL uptake by
lowering the expression of SR-A.3 In vivo experiments using ApoEdeficient mice further supported the importance of PKCb in the development of atherosclerosis.4,5 Also other PKC family members, namely
* Corresponding author. Tel: +31 20 5123261; fax: +31 20 5123310. Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2014. For permissions please email: [email protected].
468
PKCd, have been implicated in the regulation of oxLDL uptake and foam
cell formation. This involved studies employing short hairpin RNA
(shRNA) and small interference RNA (siRNA) on human macrophages,
and based on the results a model was proposed in which PKCd regulates
PKCb levels via the PI3K/Akt-ERK pathway, and consequently expression of both major scavenger receptors CD36 and SR-A.6 However,
one potential pitfall in the reported study was that the knockdown of
PKCd could cause off-target effects reducing PKCb expression levels
instead of regulating it.
Here, we have re-evaluated the role of PKCd in oxLDL uptake and
foam cell formation. We demonstrate that the specific lack of PKCd,
in either human or mouse macrophages, does not have a measurable
impact on oxLDL internalization or foam cell formation, thereby demonstrating that PKCd is non-essential for these processes.
2. Methods
2.1 Primary cells and cell lines
Peripheral mononuclear blood cells were isolated from peripheral blood or
buffy coats of healthy individuals collected by Sanquin Blood Supply (Amsterdam, The Netherlands) or patients who signed informed consent based on
principles outlined in the Declaration of Helsinki. The studies were approved
by the local ethics committee of Sanquin Blood Supply, Medical University
of Vienna, Vienna, Austria and Comite’ de Protection des Personnes
Sud-EST IV, France. Patients’ characteristics including details about PRKCD
gene mutations have been described before.7,8 Monocyte isolation was
performed by gradient centrifugation on Percoll (Pharmacia, Uppsala,
Sweden) following by either magnetic-activated cell separation using
human anti-CD14 antibody (Miltenyi Biotec B.V., Utrecht, The Netherlands)
or if higher amounts of monocytes were required, by counterflow centrifugal
elutriation as described before.9 For the purpose of phosphoproteomic analysis, freshly isolated cells were plated for 30 min at 378C and 5% CO2 in
IMDM medium (Invitrogen, Eugene, OR, USA) supplemented with 10%
foetal calf serum (FCS; Bodinco, Alkmaar, The Netherlands), 100 U/mL
of penicillin, 100 mg/mL of streptomycin, and 2 mM L-glutamine to adhere
(all Gibco Invitrogen, Breda, The Netherlands). Non-adherent cells were
discarded and the purity of the attached monocytes was .90% as determined after detachment with citric saline solution (135 mM potassium chloride and 15 mM sodium citrate) on Casy Cell Counter (Schärfe System
GmbH, Reutlingen, Germany).
For a foam cell formation assay and uptake of oxLDL, freshly isolated
monocytes were cultured for 7 – 8 days to differentiate into macrophages
in IMDM medium supplemented with 10% FCS, 100 U/mL of penicillin,
100 mg/mL of streptomycin, 2 mM L-glutamine, and 20 ng/mL of human
M-CSF (eBioscience, San Diego, CA, USA).
PLB-985 cell line was a kind gift of Dr M. Dinauer (Indiana University
Medical Center, Indianapolis, IN, USA) and THP-1 cell line was obtained
from ATCC (Teddington, UK). Cell lines were cultured in IMDM medium
supplemented with 10% FCS, 100 U/mL of penicillin, 100 mg/mL of streptomycin, and 2 mM L-glutamine. PLB-985 cells were differentiated into monocytes using 100 nM 1,25-dihydroxyvitamin D3 (vitamin D3; Sigma-Aldrich,
St Louis, MO, USA) for 5 –6 days. Expression of monocyte/macrophage
differentiation markers was determined by flow cytometry using antiCD14-FITC (Santa Cruz, Santa Cruz, CA, USA), mouse anti-human
CD11b-APC-Cy7 (BD Biosciences, Franklin Lakes, NJ, USA), and mouse
anti-human CD36-APC (BD Bioscience, San Jose, CA, USA).
2.2 Phosphoproteomic analysis
CD14+ monocytes were starved for 30 min in serum-free IMDM medium
before being exposed to 25 mg/mL of oxLDL (Biomedical Technologies,
Stoughton, MA, USA) for 5, 15, or 30 min. Preparation of lysates, enrichment
for phosphoproteins, cleavage into peptides and labelling with isobaric
K. Szilagyi et al.
labels for phosphoproteomic analysis is in more detail described Supplementary material online. Equal amounts of a control and oxLDL-treated sample
containing labelled peptides were pooled and further fractionated using the
HPLC strong cation exchange system with a polySULPHOETHYL A Column
(PolyLC, Columbia, MD, USA).10 Solvents were evaporated and fractionated
peptides were resuspended in 1% formic acid. Peptides were separated
by reversed phase chromatography and analysed on a LTQ OrbitrapXL
mass spectrometer (Thermo Fisher Scientific, Inc., Bremen, Germany) as
described.11 Collision induced dissociation (CID) spectra and higher
energy induced dissociation (HCD) spectra were acquired as described.12
The three most intense precursor ions in the full scan (300– 2000 m/z, resolving power 30 000) with a charge state of 2+ or higher were selected
for CID using an isolation width of 2 Da, a 35% normalized collision
energy, and an activation time of 30 ms. The same precursor ions were subjected to HCD with a normalized collision energy of 60%, which allows for
the identification of the reporter group from the TMT label. The identity of
the peptides including TMT labelled lysine residues and the TMT-127/
TMT-126 ratio thereof were assessed employing the Proteome discoverer
1.1 software (Thermo Fisher Scientific, Bremen, Germany). The SEQUEST
search algorithm and protein database 25.H_sapiens.fasta were used to identify the peptides. The presence of the TMT label on peptides generates additional false-negative ion fragments in the MS/MS spectra that negatively
influence the SEQUEST Xcorr score.13 We therefore employed the following peptide selection criteria: (i) three and more peptides of a particular
protein were identified in at least two time points; (ii) peptides were identified from both CID and HCD spectrum; (iii) lysine residues within the
peptide were all modified by a TMT label; and (iv) phosphorylated peptides
were identified in at least one time point. TMT ratios were normalized to the
average ratio obtained for all peptides within one experiment. A 1.8-fold
cut-off value was set to identify proteins whose expression was significantly
increased or decreased.
2.3 shRNA lentiviral transduction of PLB-985
and THP-1 cell line
To silence expression of human PKCd in cell lines, pTRIPZ-inducible lentiviral vector was used (TET Systems, Heidelberg, Germany). 293T cell line
(ATCC) was transiently transfected with shRNA constructs and packaging
constructs to generate lentiviral particles, and supernatant was collected
to transduce the cells. Briefly, 1 × 106 cells were resuspended in virus containing medium (5 MOI) to which DOTAP transfection reagent (1 : 100;
Roche Diagnostics Nederland, Almere, The Netherlands) was added.
Cells with virus were centrifuged for 2 h at 1467 g, after which medium
was refreshed. Doxycycline (Sigma-Aldrich) was used to induce expression
of shRNA sequence as well as RFP expression which allowed sorting of successfully transduced cells based on their fluorescence.
2.4 siRNA transfection of human monocytes
Human monocytes were isolated from healthy donors as described above by
magnetic-activated cell separation using human anti-CD14 antibody (Miltenyi Biotec B.V.). The cells were transfected using DharmaFect 1 transfection
reagent with either mock siRNA or RNA targeting PKCd (all purchased from
Dharmacon, Lafayette, CO, USA). Cells were used in experiments 72 h after
transfection.
2.5 Isolation of bone marrow and bone
marrow-derived macrophage culture
Bone marrow cells isolated from femur and tibia of PKCd2/2 ,14
NDK-A2/2 ,15 and wild-type (wt) mice (PKCd2/2 and matching wt mice
were on a 129/SV background; NDK-A2/2 mice and matching wt controls
were on a C57Bl/6 background) were cultured for 7 – 8 days in RPMI-1640
(Gibco Invitrogen) supplemented with 10% of heat-inactivated FCS,
100 U/mL of penicillin, 100 mg/mL of streptomycin, 2 mM L-glutamine, and
15% L929-cell conditioned medium (as source of M-CSF) to generate
469
PKCd is dispensible for oxLDL uptake and foam cell formation
bone marrow-derived macrophages (BMDMs). Expression of CD36 was
determined by flow cytometry using anti-mouse CD36 Alexa Fluor 488
(BioLegend, San Diego, CA, USA). Animal tissue was obtained from
animals treated according to the European Commission guidelines; cervical
dislocation was applied to euthanize the animals prior to tissue isolation. The
study was approved by the ethics committee of the University of Oslo, Oslo,
Norway (PKCd2/2 mice) and Université Bordeaux, Bordeaux, France
(NDK-A2/2 mice).
(phosphoPKCd Thr505) followed by incubation with goat anti-mouse-IgG
IRDye 700CW or goat anti-rabbit-IgG IRDye 800CW (LI-COR Biosciences,
Lincoln, NE, USA). Quantification of bound antibodies was performed on an
Odyssey Infrared Imaging system (LI-COR Biosciences). For the whole cell
lysates, immunostaining for alpha-tubulin (mouse anti-alpha-tubulin,
Sigma-Aldrich, Missouri, USA) followed by secondary goat anti-mouse-IgG
IRDye 700CW (LI-COR Biosciences) served as a control for equal loading
and electrophoretic transfer of each lane in the gels.
2.6 Preparation of fluorescently labelled oxLDL
2.11 Respiratory burst assay
OxLDL was purchased from Biomedical Technologies. To achieve fluoresceinated oxLDL, it was labelled using DiD dye (Molecular Probes, Eugene,
OR, USA) for 18 h at 378C. Non-bound label was separated from the labelled fraction using gel filtration through PD-10 desalting columns following
manufacturer’s instructions (GE Healthcare, Backinghamshire, UK). Concentration of labelled oxLDL was determined by NanoPhotometerw
(Implen, Munich, Germany).
Activation of respiratory burst after phorbol 12-myristate 13-acetate
(PMA, 100 ng/mL) in scrambled or shRNA targeting PKCd transduced
PLB-985 cells, BMDM, and HMDM was measured with the Amplex
Red (10-acetyl-3,7-dihydroxyphenoxazine) Hydrogen Peroxidase Assay
kit (Molecular Probes, Eugene, OR, USA). Fluorescence was measured at
60 s intervals for 60 min with the HTS7000 + plate reader.
2.7 OxLDL uptake measurement by flow
cytometry
Human monocyte-derived macrophages (HMDMs), BMDM, THP-1, and
vitamin D3 differentiated PLB-985 cell lines were incubated for 3 h at
378C with 25 mg/mL of DiD-labelled oxLDL. Adherent cells were detached
using citric saline solution and all cells were washed in PBS containing
2.5% human albumin (200 g/L, Sanquin Blood Supply). The amount of internalized particles was analysed using the FACS Canto II HTS, FACS Diva software. The inhibitor rottlerin (final concentration 10 mM, Sigma-Aldrich) was
added 30 min prior to the uptake assay and washed with PBS before oxLDL
was added, inhibitor Gö6983 (final concentration 10 mM, Sigma-Aldrich)
was added 30 min prior to the oxLDL uptake assay.
2.8 Confocal fluorescent microscopy
Cells were washed in PBS with 2.5% human albumin, fixed with 4% paraformaldehyde, and permeabilized with 0.1% Triton. Lysosomes were stained
with primary antibody (rabbit anti-LAMP-1, Thermo Fisher Scientific, Rockford, IL, USA) followed by secondary antibody (goat anti-rabbit Alexa 546,
Invitrogen, Eugene, OR, USA) or Lysotrackerw (Lysotracker, Molecular
Probes, Carlsbad, CA, USA). Cells were mounted in Vectashieldw with
DAPI (Vector Laboratories, Peterborough, UK) and analysed with a Zeiss
LSM 510 confocal laser scanning microscope (Carl Zeiss, Jena, Germany).
2.9 Foam cell formation assay
For foam cell formation assays, cells were plated on plastic slides (Thermo
Fisher Scientific, Rochester, NY, USA) and incubated for 24 h with 25 mg/
mL of oxLDL (Biomedical Technologies). Afterwards, cells were washed,
fixed with 4% paraformaldehyde, pretreated with 60% isopropanol, and
stained for lipid droplets with Oil Red O in 60% isopropanol (Sigma-Aldrich).
Cells were analysed using a light microscope at ×50 magnification (Leica,
Germany). The percentage of positively stained cells was calculated and normalized to control values using wt macrophages in the mouse studies, or
macrophages obtained from healthy donors.
2.10 Western blotting
Cell lysates were prepared using NP-40 lysis buffer as described in Supplementary material online, Methods. Prior to immunoblotting, lysates were
boiled 5 min in Laemmli sample buffer containing 1% b-mercaptoethanol.
Equal sample volumes were subjected to 7.5% sodium dodecyl sulfate–
PAGE and transferred to nitrocellulose membranes (Schleicher & Schuell,
Dassel, Germany). Membranes were washed and blocked in 5% non-fat
milk (Elk Campina, Zaltbommel, The Netherlands). The following primary
antibodies were used: total PKCd (Invitrogen, Camarillo, CA, USA) and
phosphoPKCd Thr505 (Cell Signaling, #9374, Danvers, MA, USA); membranes were incubated in 1% non-fat milk (total PKCd) or 5% BSA
2.12 Statistical analysis
Statistical analysis was performed using Prism 5.01 (GraphPad Software, San
Diego, CA, USA). Data were evaluated by two-tailed Student’s t-test if two
columns were compared. The results are presented as mean + SEM and a
significant difference was assessed as P –value ,0.05.
3. Results
3.1 Phosphorylation of PKCd during
oxLDL uptake
To identify kinases involved in the regulation of oxLDL uptake and foam
cell formation, method of semi-quantitative phosphoproteomics was
applied.16 This allowed for a targeted search for phosphorylation
events on proteins identified by mass spectrometry with an accurate relative quantification using isobaric tag labelling. Rapid phosphorylation of a
number of proteins over time was observed after exposure of human
monocytes to oxLDL when compared with cells without oxLDL treatment (see Supplementary material online, Figure S1 and Table S1). Two
closely related kinases, nucleoside diphosphate kinase (NDK) A and B,
showed the highest relative increase in phosphorylation at 30′ , with
an 4-fold increase within 5 min and 10-fold increases within 15–
30 min after oxLDL addition. NDK A and B are widely expressed and
can form hexamer complexes mediating transfer of phosphates from
ATP to nucleoside diphosphates.17 To test whether the massive and
rapid NDK-A phosphorylation was actually required for oxLDL uptake
and localization, BMDMs of NDK-A-deficient mice were used.15
However, from the normal uptake and localization of oxLDL by macrophages of NDK-A2/2 mice when compared with control mice (see Supplementary material online, Figure S2), we could rule out an important
role for at least NDK-A, and likewise for the NDK A/B complex too.
The next identified kinase with increased phosphorylation over
time was protein kinase C d (PKCd). An increase in the protein phosphorylation was less rapid and pronounced when compared with
NDK-A and B with a 2-fold increase observed after 30 min incubation
with oxLDL (Figure 1A and see Supplementary material online, Table S2).
Interestingly, using phosphoproteomics, we could detect increased
phosphorylation of the peptide sequence that contains T507 residue
of PKCd located within the activation loop of the kinase (see Supplementary material online, Table S2). We confirmed increased phosphorylation of the T507 residue by western blot (Figure 1B). We also
detected a number of peptides unique for another PKC expressed in
monocytes, namely PKCb. However, there were no changes in PKCb
470
K. Szilagyi et al.
material online, Figure S3A). However, we observed a decrease on
oxLDL uptake also in PKCd-deficient patient (see Supplementary material online, Figure S3B). Notably, the concentration of Gö6983
needed to inhibit PKCd is affecting also PKCb, which is known to be
involved in oxLDL uptake.3
3.3 Specific silencing of PKCd does not
affect oxLDL uptake
Figure 1 Treatment of human monocytes leads to rapid phosphorylation of PKCd. Freshly isolated human monocytes were exposed to
25 mg/mL of oxLDL for 5, 15, and 30 min, respectively. (A) Protein
lysates of monocytes were enriched for phosphoproteins, digested
with trypsin, labelled using TMT 126 label (control) or TMT 127
label (oxLDL-treated cells) and fractionated as in detail described in
the ‘Methods’ section. Samples were analysed by mass spectrometry
allowing relative quantification of protein phosphorylation. Monocytes
of two different healthy donors were pooled and used for analysis of
protein phosphorylation. (B) Cell lysates of control monocytes or
monocytes treated with 25 mg/mL of oxLDL for 5, 15, and 30 min,
respectively were enriched for phosphoproteins. Protein concentration was measured and equal amount (20 mg per well) of enriched proteins was loaded for analysis. Phosphorylation of T507 was analysed by
western blot using a specific a-T507 PKCd antibody. The band intensity
of oxLDL-treated cells is expressed relative to the intensity of control
cell band (set up as 1). Data represent the analysis of pooled monocytes
obtained from two healthy individuals.
phosphorylation status with increasing time of oxLDL exposure (data
not shown).
3.2 Inhibition of oxLDL uptake by the PKCd
inhibitor rottlerin
To explore, and re-evaluate,6 a potential role for PKCd in the uptake
of oxLDL and foam cell formation by macrophages, the effect of the
widely used pharmacological PKCd inhibitor rottlerin was tested using
different populations of freshly cultured human macrophages and also
monocytic/macrophage cell lines. As can be seen in Figure 2, we were
able to dramatically decrease oxLDL uptake by human macrophages
(Figure 2A) and also by the human monocytic cell lines PLB-985
(Figure 2B) and THP-1 (Figure 2C and D), and this was consistent with a
potential role of PKCd in oxLDL uptake. We additionally tested
another reported PKC inhibitor Gö6983, which shows even a stronger
decrease in oxLDL uptake as achieved by rottlerin (see Supplementary
However, although rottlerin was originally claimed to be a selective
PKCd inhibitor, this has also been questioned in more recent years
(e.g. Soltoff18), and there is lack of a specific PKCd inhibitor that would
not affect other PKC isoforms. We therefore applied RNA interference
to more specifically address the involvement of PKCd in oxLDL uptake.
Highly specific silencing of PKCd expression in monocytic cell lines
PLB-985 (Figure 3A) and THP-1 (Figure 3B) using lentiviral transduction
allowed us to study oxLDL uptake in two cell types at different stages
of monocytes differentiation. The PLB-985 cell line represents an early
stage of monocyte differentiation and the cells do not express detectable surface levels of scavenger receptor CD36 and SR-A (data not
shown). When vitamin D3-differentiated PLB-985 cells with decreased
PKCd expression were incubated in the presence of oxLDL, we did not
measure any decrease in oxLDL uptake, but instead detected even
slightly higher uptake compared with scrambled shRNA-transduced
cells (Figure 3C), which was therefore apparently due to scavenger
receptor-independent mechanism(s) of uptake. Similar results were
obtained in THP-1 cells, where we observed almost identical levels of
oxLDL uptake in both scrambled and PKCd shRNA-transduced cells
(Figure 3D). Although decreased expression of PKCd has been reported
to cause decreased CD36 scavenger receptor expression,6 we could not
detect differences in CD36 expression on normal and PKCd-silenced
THP-1 monocytes (see Supplementary material online, Figure S4A). In
our experiments, we have established an approximately 60– 75%
decrease in PKCd expression in both cell lines and the functional
consequence of that was tested by measuring extracellular hydrogen
peroxide, which is known to be produced in human monocytes as
well as in PLB985 cells through the phagocyte NADPH oxidase
(NOX2),19 regulated by PKCd.20 We could indeed confirm that decreasing expression of PKCd in the PLB-985 cell line led to a significantly
lower production of H2O2 in response to PMA stimulation (Figure 3E).
We could not perform similar tests in the THP-1 cell line, because
these cells are not capable to produce measurable levels of extracellular
reactive oxygen species (ROS) when treated with either PMA or other
stimuli (data not shown).
Despite the functionally relevant decrease in PKCd protein levels, as
measured by extracellular H2O2, we could still see a detectable decrease
in oxLDL uptake when the cells were treated with the PKCd inhibitor
rottlerin (Figure 3F and G, respectively), suggesting a PKCd-independent
mechanism.
We also performed silencing of PKCd expression in human monocytes freshly isolated from healthy donors, which probably constitutes
a more relevant population of cells in the context of atherosclerosis.
We used siRNA sequences selectively targeting PKCd as shown by analysing also expression of closely related PKCbII, (Figure 4A). Although we
were able to reduce PKCd expression to levels below 50% of mock
siRNA-treated cells, we did not observe any difference in oxLDL
uptake (Figure 4B) or CD36 levels (see Supplementary material online,
Figure S4B). On the other hand, we could observe near normal H2O2
production of silenced monocytes (Figure 4C), which could be explained
PKCd is dispensible for oxLDL uptake and foam cell formation
471
Figure 2 Treatment of human monocytes/macrophages with PKCd inhibitor rottlerin decreases uptake of oxLDL. (A) HMDMs of healthy donors or
(B) human monocytic cell line PLB-985, and (C) THP-1 were either incubated with 25 mg of fluorescently labelled oxLDL alone or pretreated with
10 mM of rottlerin for 30 min. Uptake of oxLDL was analysed by flow cytometry and presented as the percentage of uptake by non-treated cells. (D)
Human monocytic THP-1 cells were either treated with 25 mg/mL of fluorescently labelled oxLDL (green) for 3 h or additionally pretreated for
30 min with 10 mM of rottlerin. Cells were stained with lysotracker (red) and lysosomal localization of oxLDL was visualized using confocal microscopy.
Data are presented as mean + SEM and are representative of three (PLB-985) or six independent experiments (THP-1 and HMDMs) (T-test, ***P , 0.001,
**P , 0.01, *P , 0.05).
by less efficient knockdown of PKCd compared with knockdown
achieved in the PLB-985 cell line (50 vs. 75%). Furthermore, as
seen with the cell lines, the reducing effect of rottlerin on oxLDL
uptake was not altered by PKCd silencing (Figure 4D).
3.4 Absence of PKCd expression has
no influence on oxLDL uptake and foam
cell formation
Although we were able to achieve a prominent decrease in PKCd expression with RNAi in multiple cell systems, the remaining expression
levels of the protein might still have been sufficient to mediate normal
uptake of oxLDL by monocytes and macrophages. To circumvent this
problem, we used BMDMs of mice completely lacking PKCd protein,
but showing normal expression of PKCbI and II, for studying oxLDL
uptake and foam cell formation (Figure 5A). Even in the complete
absence of PKCd the ability of macrophages to take up oxLDL
showed no detectable reduction (Figure 5B), whereas at the same
time the cells had considerably decreased hydrogen peroxide production (Figure 5C), confirming the functional consequence of PKCd
absence. If PKCd is indeed regulating CD36 expression,6 we expected
PKCd2/2 cells to show changes in the expression of this receptor.
472
K. Szilagyi et al.
Figure 3 PKCd knockdown does not reduce uptake of oxLDL in human monocytic cell lines PLB-985 and THP-1. Lentiviral transduction was used to
silence expression of PKCd in (A) PLB-985 and (B) THP-1 cell line and knockdown efficiency was determined by western blot analysis of PKCd expression. In
PLB-985 cell line, specificity of used shRNA sequence was tested by staining for protein expression of other two PKC isoforms, PKCbI and II, respectively
(A). PLB-985 (C) or THP-1 cells (D) transduced with scrambled or PKCd targeting shRNA were incubated for 3 h with 25 mg of fluorescently labelled
oxLDL and uptake was determined using flow cytometry. (E) Functional relevance of PKCd knockdown was tested by measuring H2O2 release in
PLB-985 cells, when control cells transduced with scramble shRNA and shRNA targeting PKCd were treated with PMA to induce production of ROS.
PLB-985 (F) or THP-1 (G) cells were either treated for 3 h with only fluorescently labelled oxLDL (25 mg/mL) or pretreated for 30 min with the inhibitor
rottlerin, and uptake of oxLDL was determined by flow cytometry. Western blot images show representative results of three (PKCb) or four (PKCd)
individual experiments. Graphs show results from three individual experiments for oxLDL uptake by THP-1 cell line, four individual experiments for measurement of hydrogen peroxide levels by PLB-985 cell line, and seven individual experiments for oxLDL uptake by PLB-985 cell line, all normalized for values
of control cells transduced with non-silencing-scrambled shRNA and are presented as mean + SEM (T-test, ***P , 0.0001, **P , 0.01, *P , 0.05).
473
PKCd is dispensible for oxLDL uptake and foam cell formation
Figure 4 A decreased level of PKCd in primary monocyte-derived macrophages has no effect on oxLDL uptake. (A) HMDMs were treated with mock
siRNA and siRNA targeting PKCd, and the levels of PKCd and bII were determined by western blot analysis. (B) Uptake of fluorescently labelled oxLDL
(25 mg/mL) by macrophages was measured after 3 h of incubation using flow cytometry. (C) Hydrogen peroxide production by macrophages was tested by
stimulating the cells with PMA. (D) Uptake of oxLDL was analysed by flow cytometry in the presence of rottlerin, when cells were first pretreated for 30 min
with the inhibitor followed by 3 h incubation with fluorescently labelled oxLDL (25 mg/mL) and compared those incubated with oxLDL only. Western blot
images show representative results of seven (PKCd) and three (PKCbII) individual experiments. Graphs show results from four individual measurements of
hydrogen peroxide release (D) and seven individual experiments for oxLDL uptake (B and D), all normalized for values of control cells transduced with
non-silencing mock siRNA and are presented as mean + SEM (T-test, ***P , 0.0001, **P , 0.01).
However, macrophages missing PKCd expressed normal levels of CD36
(Figure 5D) and were able to form foam cells in an extent comparable
with that of wt cells (Figure 5E).
Finally, we analysed macrophages from three patients with reported
mutations in the PRKCD gene,7,8 leading either to the complete
absence of PKCd (Figure 6A) or a considerably reduced expression
(Figure 6B and C), while the expression of PKCbI/II remained unaffected.
Monocytes-derived macrophages of all patients were analysed in comparison with macrophages cultured from monocytes of two or three
healthy donors in each experiment. The results clearly showed that all
individuals have completely normal uptake of oxLDL (Figure 6D) as
well as expression of CD36 (Figure 6E). Furthermore, macrophages
from the patient with a complete PKCd deficiency displayed a level of
foam cell formation that clearly falls within a range of that of healthy
controls (Figure 6G), whereas hydrogen peroxide release has been
decreased upon PMA stimulation in neutrophils of the patient (K.S.
et al., manuscript in preparation). Patients with partial PKCd deficiency
have shown normal capacity to produce ROS by neutrophils (data not
shown) probably due to remaining expression of the protein. Of note,
rottlerin could still reduce the uptake of oxLDL in the cells from all
studied controls and patients (Figure 6F).
4. Discussion
In the present study, we provide definitive evidence that PKCd does not
play a non-redundant role in the regulation of oxLDL uptake and foam
cell formation by macrophages. Perhaps, the most compelling evidence
for this is based on studies with macrophages from patients with distinct
mutations in the PRKCD gene (Figure 6), leading either to a complete
absence8 or a dramatically decreased stability and expression of
PKCd.7 We have additionally tested the function of PKCd in oxLDL
uptake and foam cell formation by using macrophages derived from
PKCd2/2 mice (Figure 5) and by applying methods of RNA interference
to decrease PKCd expression in human cell lines and primary monocytes
(Figure 3 and 4). Clearly, these findings are in sharp contrast to data recently reported by Lin et al. 6 A major difference between the results
from Lin et al. and our own study is that the RNAi strategy used by Lin
et al. not only silenced PKCd expression, but also resulted in a significant
reduction in expression of the PKCbI/II isoforms. Of relevance, the
PKCbI/II isoforms have independently been shown to be important
for oxLDL uptake and SR-A expression.3 In contrast, our experiments
were performed with macrophages in which a highly selective reduction
or deletion in PKCd was achieved. Therefore, we conclude that PKCd is
not involved in the regulation of oxLDL uptake and foam cell formation.
474
K. Szilagyi et al.
Figure 5 The complete absence of PKCd in mouse BMDMs has no effect on oxLDL uptake and foam cell formation. (A) Mouse BMDMs from wt and
PKCd2/2 mice were cultured as described in ‘Methods’ and PKCd, bI, and II expression levels in the cell lysates were determined using western blot. (B)
Uptake by the macrophages was measured after 3 h of incubation with fluorescently labelled oxLDL (25 mg/mL) using flow cytometry. (C) Hydrogen peroxide release was tested by stimulating the macrophages with PMA. (D) Surface expression of scavenger receptor CD36 on the macrophages was analysed
by flow cytometry. (E) Bone marrow-derived macrophages of wt and PKCd2/2 mice were incubated for 24 h in the presence of 25 mg of oxLDL and foam
cell formation was determined by Oil Red O staining of lipid droplets as described in the ‘Methods’ section. Western blot images show representative
results of three (PKCb) and four (PKCd) individual experiments. Graphs show results from three individual experiments for foam cell formation and
six individual experiments for oxLDL uptake, CD36 expression, and extracellular hydrogen peroxide measurement, normalized for values of wt cells,
and are presented as mean + SEM (T-test or ANOVA, ***P , 0.001).
Our interest in PKCd was initially triggered by the rapid T507 phosphorylation within the kinase activation loop observed during phosphoproteomic screen of human monocytes exposed to oxLDL (Figure 1).
Phosphorylation of highly conserved Thr located close to catalytic
sites by phosphoinositide-dependent kinase-1 generally appears to be
required to obtain catalytically competent PKC enzymes.21 An increase
in T507 phosphorylation leads to catalytically more potent PKCd once
activated and associated with a membrane.22
In favour of a role for PKCd is the observation of the pronounced
effect of rottlerin on oxLDL uptake (Figure 2). However, currently
there is no specific PKCd inhibitor available and a number of studies
have discredited rottlerin as a specific PKCd inhibitor and seriously questioned its inhibitory effects towards any of the PKC isoforms.23 Rottlerin
has been reported to display a wide range of effects, including inhibition
of PRAK (p38-regulated/activated protein kinase) and MAPKAP-K2 or
such as CHK2, PLK1 (Polo-like kinase 1), PIM3, and SRPK1 [SRSF
PKCd is dispensible for oxLDL uptake and foam cell formation
475
Figure 6 PKCd-deficient patients show no difference in oxLDL uptake and foam cell formation by monocyte-derived macrophages. (A – C) Western blot
analysis of monocyte-derived macrophages of three patients (V, L1, and L2) with reported mutations in the PRKCD gene was performed to determine levels
of PKCd and bI/II. Macrophages of all patients (D) were incubated for 3 h with 25 mg/mL of oxLDL and uptake was measured by flow cytometry. Surface
expression of scavenger receptor CD36 was analysed by flow cytometry on macrophages of 14 controls and all patients (E). Macrophages of the controls
and patients were pretreated with 10 mM of rottlerin, washed, and incubated for 3 h with oxLDL to measure uptake (F). Macrophages of the patient V (G)
incubated for 24 h with 25 mg/mL of oxLDL were stained with Oil Red O as described in ‘Methods’ to determine foam cell formation. Every oxLDL uptake
measurement consists of two technical replicates that were averaged prior to statistical analysis. Monocytes-derived macrophages of 2 – 3 healthy donors
were used as day controls for each patient. Each symbol represents a separate patient sample (filled squares for patient V; filled diamonds for patient L1;
filled inverted triangles for patient L2). Macrophages of patient V and L1 were analysed at two different occasions. Fluorescence of internalized oxLDL
measured by flow cytometry in healthy donors macrophages was averaged for each individual experiment (day control oxLDL uptake ¼ 100%) and
oxLDL uptake by patient macrophages is displayed as a percentage of a day control.
(serine/arginine-rich splicing factor) protein kinase 1].23 Our own results
add to the scepticism with respect to the PKCd specificity of rottlerin, as
we have observed a pronounced rottlerin-mediated decrease in oxLDL
uptake by macrophages with strongly reduced expression of PKCd. It
seems feasible, among other options, that the effect of rottlerin on
oxLDL uptake is explained by its potent capacity to inhibit macropinocytosis,24 one of the scavenger receptor-independent ways of oxLDL internalization.1 The biological function of rapid PKCd phosphorylation
after oxLDL exposure in macrophages is probably in induction of apoptosis via p53 destabilization as has been previously reported.25
476
In summary, our results argue against a direct intrinsic role for PKCd
involvement in oxLDL uptake and foam cell formation, thereby questioning the previously suggested role of PKCd in these processes that
are pivotal to the development of atherosclerosis.
Supplementary material
Supplementary material is available at Cardiovascular Research online.
Acknowledgement
We thank Carmen van der Zwaan and Martijn Veldthuis for their technical assistance with phosphoproteomic analysis.
Conflict of interest: none declared.
Funding
This work was supported by the Netherlands Organisation for Scientific
Research (NWO TOP 91208001 to G.K., M.P.J.W., and T.K.B).
References
1. Moore KJ, Sheedy FJ, Fisher EA. Macrophages in atherosclerosis: a dynamic balance.
Nat Rev Immunol 2013;13:709–721.
2. Pelkmans L, Fava E, Grabner H, Hannus M, Habermann B, Krausz E, Zerial M. Genomewide analysis of human kinases in clathrin- and caveolae/raft-mediated endocytosis.
Nature 2005;436:78–86.
3. Osto E, Kouroedov A, Mocharla P, Akhmedov A, Besler C, Rohrer L, Iliceto S, Volpe M,
Luscher TF, Cosentino F. Inhibition of protein kinase C beta prevents foam cell formation
by reducing scavenger receptor A expression in human macrophages. Circulation 2008;
118:2174 –2182.
4. Kong L, Shen X, Lin L, Leitges M, Rosario R, Zou YS, Yan SF. PKCbeta promotes vascular
inflammation and acceleration of atherosclerosis in diabetic ApoE null mice. Arterioscler
Thromb Vasc Biol 2013;33:1779 –1787.
5. Harja E, Chang JS, Lu Y, Leitges M, Zou YS, Schmidt AM, Yan SF. Mice deficient in PKCbeta
and apolipoprotein E display decreased atherosclerosis. FASEB J 2009;23:1081 – 1091.
6. Lin CS, Lin FY, Ho LJ, Tsai CS, Cheng SM, Wu WL, Huang CY, Lian CH, Yang SP, Lai JH.
PKCdelta signalling regulates SR-A and CD36 expression and foam cell formation. Cardiovasc Res 2012;95:346–355.
7. Belot A, Kasher PR, Trotter EW, Foray AP, Debaud AL, Rice GI, Szynkiewicz M,
Zabot MT, Rouvet I, Bhaskar SS, Daly SB, Dickerson JE, Mayer J, O’Sullivan J, Juillard L,
Urquhart JE, Fawdar S, Marusiak AA, Stephenson N, Waszkowycz B, Beresford W,
Biesecker LG, Black CM, Rene C, Eliaou JF, Fabien N, Ranchin B, Cochat P,
Gaffney PM, Rozenberg F, Lebon P, Malcus C, Crow YJ, Brognard J, Bonnefoy N.
Protein kinase c delta deficiency causes Mendelian systemic lupus erythematosus with
B cell-defective apoptosis and hyperproliferation. Arthritis Rheum 2013;65:2161 –2171.
8. Salzer E, Santos-Valente E, Klaver S, Ban SA, Emminger W, Prengemann NK,
Garncarz W, Mullauer L, Kain R, Boztug H, Heitger A, Arbeiter K, Eitelberger F,
Seidel MG, Holter W, Pollak A, Pickl WF, Forster-Waldl E, Boztug K. B-cell deficiency
and severe autoimmunity caused by deficiency of protein kinase C delta. Blood 2013;
121:3112 –3116.
K. Szilagyi et al.
9. Sanderson RJ, Shepperdson RT, Vatter AE, Talmage DW. Isolation and enumeration of
peripheral blood monocytes. J Immunol 1977;118:1409 –1414.
10. Mischerikow N, Altelaar AF, Navarro JD, Mohammed S, Heck AJ. Comparative assessment of site assignments in CID and electron transfer dissociation spectra of phosphopeptides discloses limited relocation of phosphate groups. Mol Cell Proteomics 2010;9:
2140– 2148.
11. van Haren SD, Herczenik E, ten BA, Mertens K, Voorberg J, Meijer AB.
HLA-DR-presented peptide repertoires derived from human monocyte-derived dendritic cells pulsed with blood coagulation factor VIII. Mol Cell Proteomics 2011;10:M110.
12. Dayon L, Pasquarello C, Hoogland C, Sanchez JC, Scherl A. Combining low- and highenergy tandem mass spectra for optimized peptide quantification with isobaric tags.
J Proteomics 2010;73:769 –777.
13. Thingholm TE, Palmisano G, Kjeldsen F, Larsen MR. Undesirable charge-enhancement of
isobaric tagged phosphopeptides leads to reduced identification efficiency. J Proteome
Res 2010;9:4045 –4052.
14. Leitges M, Mayr M, Braun U, Mayr U, Li C, Pfister G, Ghaffari-Tabrizi N, Baier G, Hu Y,
Xu Q. Exacerbated vein graft arteriosclerosis in protein kinase C delta-null mice. J Clin
Invest 2001;108:1505 –1512.
15. Arnaud-Dabernat S, Bourbon PM, Dierich A, Le MM, Daniel JY. Knockout mice as model
systems for studying nm23/NDP kinase gene functions. Application to the nm23-M1
gene. J Bioenerg Biomembr 2003;35:19–30.
16. Nilsson CL, Dillon R, Devakumar A, Shi SD, Greig M, Rogers JC, Krastins B, Rosenblatt M,
Kilmer G, Major M, Kaboord BJ, Sarracino D, Rezai T, Prakash A, Lopez M, Ji Y, Priebe W,
Lang FF, Colman H, Conrad CA. Quantitative phosphoproteomic analysis of the STAT3/
IL-6/HIF1alpha signaling network: an initial study in GSC11 glioblastoma stem cells.
J Proteome Res 2010;9:430 –443.
17. Gilles AM, Presecan E, Vonica A, Lascu I. Nucleoside diphosphate kinase from human
erythrocytes. Structural characterization of the two polypeptide chains responsible
for heterogeneity of the hexameric enzyme. J Biol Chem 1991;266:8784 –8789.
18. Soltoff SP. Rottlerin: an inappropriate and ineffective inhibitor of PKCdelta. Trends Pharmacol Sci 2007;28:453–458.
19. van Beek EM, Zarate JA, van BR, Schornagel K, Tool AT, Matozaki T, Langenbach GE,
Kraal G, Everts V, van den Berg TK. SIRPalpha controls the activity of the phagocyte
NADPH oxidase by restricting the expression of gp91(phox). Cell Rep 2012;2:
748 – 755.
20. Bey EA, Xu B, Bhattacharjee A, Oldfield CM, Zhao X, Li Q, Subbulakshmi V,
Feldman GM, Wientjes FB, Cathcart MK. Protein kinase C delta is required for
p47phox phosphorylation and translocation in activated human monocytes.
J Immunol 2004;173:5730 – 5738.
21. Newton AC. Protein kinase C: poised to signal. Am J Physiol Endocrinol Metab 2010;298:
E395– E402.
22. Le Good JA, Ziegler WH, Parekh DB, Alessi DR, Cohen P, Parker PJ. Protein kinase C
isotypes controlled by phosphoinositide 3-kinase through the protein kinase PDK1.
Science 1998;281:2042 –2045.
23. Wu-Zhang AX, Newton AC. Protein kinase C pharmacology: refining the toolbox.
Biochem J 2013;452:195 –209.
24. Sarkar K, Kruhlak MJ, Erlandsen SL, Shaw S. Selective inhibition by rottlerin of
macropinocytosis in monocyte-derived dendritic cells. Immunology 2005;116:
513 – 524.
25. Giovannini C, Vari R, Scazzocchio B, Sanchez M, Santangelo C, Filesi C, D’Archivio M,
Masella R. OxLDL induced p53-dependent apoptosis by activating p38MAPK and
PKCdelta signaling pathways in J774A.1 macrophage cells. J Mol Cell Biol 2011;3:
316 – 318.