European Journal of Cell Biology 94 (2015) 3245

Contents lists available at ScienceDirect

European Journal of Cell Biology

journal homepage: www.elsevier.com/locate/ejcb

CAP2 is a regulator of the actin cytoskeleton and its absence changes

inltration of inammatory cells and contraction of wounds
Kosmas Kosmas a , Ali Eskandarnaz a , Arya B. Khorsandi a , Atul Kumar a , Rajeev Ranjan d ,
Sabine A. Eming b,c,d , Angelika A. Noegel a,b,c , Vivek S. Peche a,b,c,
a
Institute of Biochemistry I, Medical Faculty, University of Cologne, Cologne, Germany
b
Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, Germany
c
Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Cologne, Germany
d
Department of Dermatology, University of Cologne, Cologne, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Cyclase associated protein (CAP) is a highly conserved protein with roles in actin dynamics and many cel-
Received 9 April 2014 lular processes. Two isoforms exist in higher eukaryotes, CAP1 and CAP2. CAP1 is ubiquitously expressed
Received in revised form 21 October 2014 whereas CAP2 shows restricted tissue distribution. In mice, ablation of CAP2 leads to development of
Accepted 21 October 2014
cardiomyopathy. CAP2 is expressed in skin. In human skin its expression is increased in wounds. To elu-
cidate the role of CAP2 in skin upon injury, we studied the wound healing in CAP2 decient mice and
Keywords:
found altered wound healing response presumably resulting from reduced levels of -SMA, decreased
Cyclase associated protein
macrophage inltration and slower neovascularization. In vitro cultured Cap2 decient keratinocytes
CAP1
Wound healing
showed reduced velocity and a delay in scratch closure. The analysis of primary mutant broblasts also
Cell migration showed reduced velocity and less contractibility. They had extended protrusions and more focal adhe-
Actin sions. In addition the F-actin content was increased keeping the total actin content unaltered. Mutant
Actin binding proteins broblasts furthermore exhibited an altered response during recovery from drug-induced disruption of
the actin cytoskeleton. Interestingly, CAP1 was upregulated in knockout unwounded skin and in wounds
which might partially compensate for the loss of CAP2. Taken together, our studies reveal a role for CAP2
in wound healing which may be based on its function as a regulator of the actin cytoskeleton.
2014 Elsevier GmbH. All rights reserved.

Introduction of signaling factors, and regeneration of normal skin architec-

Skin is the largest organ in the human body, which provides
a barrier between an organism and its surrounding environment.
Skin harbors compartments such as the dermis and epidermis,
which have a particular role in skin homeostasis. Cutaneous trauma
disrupts skin architecture and integrity, which elicits a highly reg-
ulated localized response that cleans, debrides, and heals the site
of injury. Trauma to skin can arise from abrasions, lacerations, and
thermal, electrical, or chemical burns (Almine et al., 2012; Trott,
1988). Cutaneous wound healing is a complex and dynamic pro-
cess involving soluble mediators, blood cells, extracellular matrix
(ECM), and parenchymal cells (Singer and Clark, 1999). The inam-
matory response to tissue injury includes differential expression
Corresponding author at: Institute of Biochemistry I, Medical Faculty, University
of Cologne, Joseph-Stelzmann-Str. 52, D-50931 Kln,
Germany. Tel.: +49 221 478 6988; fax: +49 221 478 6979.
E-mail address: pechev@uni-koeln.de (V.S. Peche).
ture (Almine et al., 2012). The process of wound healing normally
proceeds from coagulation and inammation through broplasia,
matrix deposition, angiogenesis, epithelialization, collagen matu-
ration and nally wound contraction (Schfer and Werner, 2008).
Although the actin cytoskeleton is an essential component of cells
for their physiological and pathophysiological processes, little focus
has been on the role of actin binding proteins which maintain the
G-/F-actin dynamics during wound healing.
Many proteins bind to either G- or F-actin and display a unique
function. One family which binds to G actin is the family of actin
sequestering proteins. Cyclase associated protein (CAP) falls under
this category and was rst described in yeast (Srv2) with roles
in Ras mediated responses of adenylyl cyclase and cell morphol-
ogy (Fedor-Chaiken et al., 1990). CAP is a protein composed of
474551 amino acid residues, which is highly conserved through-
out evolution. CAP/Srv2p is a multifunctional protein harboring
two functional domains, the N-domain consisting of an adenylyl
cyclase binding domain and the C-domain which is an actin binding
domain and is separated from the N-domain by a centrally located

http://dx.doi.org/10.1016/j.ejcb.2014.10.004
0171-9335/ 2014 Elsevier GmbH. All rights reserved.
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245
proline-rich stretch. Mammals have two CAP genes encoding the
related CAP1 and CAP2. With specic antibodies we studied their
distribution and subcellular localization. CAP1 shows a broad tis-
sue distribution, CAP2 is signicantly expressed only in brain, heart
and skeletal muscle, and in skin (Peche et al., 2007). ASP56, the
rst identied mammalian homolog of CAP, was isolated from pig
platelets on the basis of its actin sequestering activity (Gieselmann
and Mann, 1992). The in vivo role of CAP in whole organisms has not
been studied in detail and is complicated by the presence of the two
isoforms. Earlier studies with CAP1 suggest that it could promote
rapid actin dynamics in conjunction with ADF/colin. Also, CAP1
is involved in cell motility and endocytosis in mammalian cells
(Bertling et al., 2004), and recently CAP1 was shown to translocate
rapidly to mitochondria and stimulate colin-induced apoptosis
upon treatment with a number of agents that induce apoptosis
(Wang et al., 2008).
Through the regulation of the actin cytoskeleton CAP plays a
major role in cell polarity and development. Mutation of CAP in the
Drosophila eye (acu) leads to a premature differentiation of pho-
toreceptors (Benlali et al., 2000). This is thought to be due to the
inability of the cells to undergo shape changes that are required for
proper hedgehog signaling. In D. discoideum as well as in Drosophila
oogenesis, CAP plays a key role in cell polarity (Baum et al., 2000;
Noegel et al., 2004). D. discoideum CAP is also involved in endocyto-
sis and vesicle transport (Sultana et al., 2005). Its proline rich region
appears to be essential for phototaxis and cell size maintenance
(Noegel et al., 2004). Loss of cas-1, the C. elegans cyclase-associated
protein, is homozygous lethal with defects in sarcomeric actin
organization (Nomura et al., 2012). Moreover, CAS-2 (C. elegans
CAP2) is important in the ATP-dependent regulation of the actin
monomer-lament equilibrium (Nomura and Ono, 2013). Further-
more, human CAP2 was shown to be overexpressed in multistage
hepatocarcinogenesis (Shibata et al., 2006).
We developed a mouse model lacking CAP2. The whole body
knock out was generated through a gene trap approach in which
exon 2 was targeted. CAP2 knock out mice showed a decrease in
body weight and had a decreased survival rate. Further, they devel-
oped a severe cardiac defect marked by dilated cardiomyopathy
(DCM) associated with drastic reduction in basal heart rate and
prolongations in atrial and ventricular conduction times. More-
over, CAP2-decient myobrils exhibited reduced cooperativity
of calcium regulated force development. At the microscopic level,
we observed disarrayed sarcomeres with development of brosis
(Peche et al., 2013).
Based on the restricted tissue distribution of CAP2 in mammals,
we sought to understand the role of CAP2 in skin repair upon injury.
For this purpose we performed in vivo wound healing. We noted
a reduced level of -SMA, decreased macrophage inltration and
slower neovascularization. At day 10 post wounding, we observed
a delayed wound contraction in Cap2 knockout mice compared to
their wild type (WT) littermates. In addition, primary broblasts
lacking CAP2 developed extended protrusions and increased focal
adhesions and exhibited slower migration velocities. We noticed
an upregulation of CAP1 in knockout skin and knockout wounds
which might partially compensate for the loss of CAP2. We conclude
that CAP2, a regulator of the actin cytoskeleton, has roles in cell
migration and wound healing.
Experimental procedures
CAP2 decient mice and skin wounding
Cap2 gene trap mice which have a whole body deletion of
the protein generated by targeting and trapping exon 2 were
described previously (Peche et al., 2013). For wound healing
33
experiments mice were rst anesthetized and four circular wounds
of 5 mm diameter were generated at the dorsal site by excising
skin and the subcutaneous fat and muscle panniculus carnosus
using a punch (pfm medical ag, Kln Germany). 4 mice per geno-
type per time point were used in these studies. For the day 10
time point, 8 mice per genotype were used. Wounds were left
uncovered, digitally photographed at the indicated time points
and harvested at days 3, 7 and 10 after wounding. Animals were
housed in specic pathogen-free facilities and all animal proto-
cols were approved by the local veterinary authorities. All animal
experiments were performed in accordance with the German ani-
mal protection law. Institutional Animal Care and Use Committee
approval was obtained for all animal studies. Tissue of normally
healing human wounds (n = 2 per time point) was taken by con-
sent from healthy volunteers. Wounds were created by performing
a punch biopsy (6 mm diameter, 0.5 cm depth, lower back), and at
indicated time points following wounding the wound was excised.
Wound tissues were processed for histological analysis as described
below. The study adhered to the Declaration of Helsinki Principles,
and skin biopsies were collected according to a protocol approved
by the Ethics committee at the University of Cologne.
Preparation of wound tissue
For histology, the complete wounds were excised with a small
margin of surrounding skin. Tissues were xed for 2 h in 4%
paraformaldehyde before parafn embedding or frozen unxed
in optimal cutting temperature compound (OCT, Sakura, Torrance,
CA). The parafn embedded and cryopreserved wounds were cut
in serial sections from the surrounding wound margin across the
center of the wound toward the opposite wound edge in the cau-
docranial direction.
Immunohistochemistry, antibodies and histology
For general histology, the samples (parafn sections of 7 m
thick) were stained with hematoxylin and eosin (H&E) according
to standard procedures. For immunouorescence, parafn sec-
tions were deparafnised in 2 changes of xylene and rehydrated
through a graded ethanol series, which was then followed by anti-
gen retrieval and antibody incubation. For skin samples incubation
was done with mAbs specic for -SMA (Sigma), Ki-67 (DAKO),
vinculin (Sigma). Appropriate secondary antibodies were conju-
gated with Alexa Fluor 488 and 568 (Molecular Probes). Nuclei
were visualized with either 4 ,6-diamidino-2-phenylindole (DAPI)
or propidium iodide (PI). Sections were mounted and imaged with
a Leica confocal microscope. Massons trichrome staining to detect
brosis was performed according to the manufacturers protocol
(Sigma).
Cultured cells were xed with paraformaldehyde and processed
for immunouorescence analysis for detecting CAP2 with mAb
K82381-1 which had been generated against bacterially expressed
N-terminal domain of CAP2 (aa 1-310). F-actin was visualized
with TRITC Phalloidin. Nuclei were stained with DAPI. Cells were
incubated with primary and secondary antibodies for 1 h at room
temperature each.
Isolation of primary broblasts from mouse skin
Primary skin broblasts were isolated from newborns of vari-
ous mouse strains. Mice were killed by decapitation and the skin
was carefully peeled off from the rest of the body. The skin was
surface-sterilized with Betaisodona and then transferred to a plate
with 0.01% trypsin in PBS followed by incubation at 4 C. On the
subsequent day, the dermis was separated from the epidermis, cut
into small pieces and then treated with collagenase type I (2 ml of
34 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245
400 U/ml diluted in medium) for 1 h at 37 C. The cells were then
plated on 10 cm  culture dishes with DMEM medium containing
FCS and grown to conuency, split once and frozen (5 106 cells per
vial) in FCS containing 10% DMSO. All experiments were performed
between passage 3 and passage 5.
Keratinocyte isolation and culture
Primary Keratinocytes were isolated using standard procedures.
Newborn mice were used for the isolation (not older than 3 days).
After decapitation, bodies were cooled on ice for 1 h. Bodies were
processed in 1 PBS/Betaisodona Iodine for 1 min followed by a
brief rinse in PBS. Further, bodies were processed in 70% ethanol for
1 min and rinsed once with PBS. Tail cuts were taken for genotyp-
ing and skin was carefully peeled off from the rest of the body. The
skin was transferred to a dish with antibiotic/antimycotic solution
(Gibco) 1:100 in PBS without calcium and magnesium and incu-
bated for a few minutes. Subsequently the skin was transferred to
a new 35 mm dish and spread on the plastic surface with the epi-
dermal side facing upward. 2 ml of trypsin for cell culture (0.025%
trypsin in PBS/EDTA) was added from the side ensuring that only
the dermal part came into contact with trypsin. It was kept at
4 C overnight. The next day, dermis and epidermis were sepa-
rated and the epidermis was mined into very small pieces using
two scalpels. They were then resuspended in 1.5 ml of keratinocyte
growth media (low calcium) (Promocell) and kept shaking in a ster-
ile Eppendorf tube for 30 min, room temperature at 1000 rpm. After
30 min cells were deposited on collagen coated 6 well plates. Mean-
while mitomycin C (Sigma) treated feeder cells were trypsinized,
resuspended in keratinocyte growth media and added to the 6 well
plates containing epidermal cells to make up the volume to 2 ml.
Cells were grown to conuency and split 1:2 on collagen coated
plates. Four wild type and four knock out preparations were used
in the experiments. All experiments were performed between pas-
sage 3 and passage 5.
In vitro keratinocyte proliferation assay
WT and knock out primary keratinocytes were seeded at a
density of 0.05 106 cells in 24 well plates. Cells were grown to
7080% conuency. Later they were trypsinized and living cells
were counted with a Bio-Rad automated cell counting chamber
which uses trypan blue dye to distinguish living and dead cells.
The generation time was calculated according to the formula
G = t log 2/[log(Nt log(N0 ); t = growth time,
N0 = seeding density at time0; Nt = cell count at timet.
Four different preparations of WT and knock out keratinocytes
were used for the proliferation assay (generation time). The exper-
iment was performed twice.
Ki67 staining protocol
Cell proliferation assays were performed with an antibody
directed against proliferation-specic protein Ki-67 using rat
anti-mouse Ki-67, clone TEC-3 (DAKO). Proliferating cells stain
redbrown by this technique whereas non-proliferating cells stain
blue. After deparafnizing and rehydration of the sections, the
endogenous peroxidase activity was blocked by incubating the sec-
tions in PBS containing 0.4% hydrogen peroxide. The sections were
then blocked with avidin and biotin (Dako). The cells were perme-
abilized with 0.01% (w/v) saponin followed by blocking for 30 min
with 10% (v/v) FCS in PBS. The primary antibody was diluted in 2%
(v/v) FCS and the sections were incubated overnight at 4 C. On
the following day slides were rinsed with PBS containing 0.05%
(v/v) Tween 20. Biotinylated rat anti-mouse secondary antibody
was then added to the sections and incubated for further 45 min
at room temperature. The slides were again washed and incubated
in HRP-conjugated streptavidin (Dako) for 45 min. After washing,
the sections were developed in 10% AEC stock solution (Sigma) in
acetate buffer. AEC is the substrate for the peroxidase reaction. The
sections were then counter stained with hematoxylin and mounted
in aqueous mounting medium Paramount (Dako).
Two step collagen contraction assay
The collagen gel contraction assay was performed on 24 well
plates according to the manufactures instructions (Cell Biolabs,
Inc.). Briey, a collagen solution is added to a cold sterile tube. Sub-
sequently 5 DMEM solution is added and the solutions mixed well.
Immediately neutralization solution was added and mixed and the
sample kept on ice until further processing. Primary broblasts
were harvested and resuspended in broblast growth media at a
concentration of 5 106 cells/ml. A collagen lattice was prepared
by mixing 2 parts of the cell suspension and 8 parts of collagen gel
solution. 0.5 ml of cellcollagen mixture was added per well in a 24-
well plate and incubated at 37 C for 1 h. After the polymerization
of collagen, 1 ml of culture media was added atop each collagen lat-
tice. For stress development, cultures were incubated for two days.
To initiate the contraction, collagen gels in each well were gently
release from the side using a sterile spatula. The collagen gel size
was measured using a ruler. 4 different primary broblast prepa-
rations from 4 different mice each (WT and mutant) were used in
this experiment. The experiment was performed twice.
Cell culture and cell scratch assay
Keratinocytes and broblasts were isolated from both WT and
Cap2gt/gt mice and were cultured in primary keratinocyte growth
media (Promocell) and DMEM medium with 10% fetal bovine serum
(FBS), l-Glutamine and antibiotics, respectively, in 5% CO2 in a 37 C
incubator. For observing cell spreading and morphology, WT and
Cap2gt/gt broblasts were trypsinized for 5 min and centrifuged. The
cells were resuspended in medium as mentioned above and placed
in a 10 cm petri-dish overnight.
For the cell scratch assays, mouse primary keratinocytes and
broblasts from Cap2gt/gt and WT were cultured in medium as
described above and placed in a -slide 8 well (Ibidi) attached
to a culture insert (Ibidi). Keratinocytes and broblasts were
trypsinized for 5 min, centrifuged, and resuspended in medium.
35 103 cells were seeded and cultured overnight at 37 C with 5%
CO2 . The next day, the culture inserts were removed to create the
scratch and cells were rinsed with fresh medium once and fed with
culture medium. Migration of wild type and mutant keratinocytes
and broblasts into the open area was analyzed by time lapse video
microscopy (37 C, 5% CO2 ) using a Leica CTR 7000 HS microscope
(LAS AF-AF6000 software) equipped with a Hamamatsu A3472-07
camera and a Plan-Neouar 10/0.30 Ph1 objective. For the cell-
tracking analysis movies were made for 24 h with frames taken
every 15 min and quantication of cell migration speed was done
using ImageJ tool. 4 different primary broblast and keratinocytes
preparations from 4 different mice per genotype (WT and mutant)
were used in this experiment. The experiment was performed
thrice.
Western blot analyses
Tissues and cells were homogenized and lysed (1% Triton X-100,
0.1 M NaCl, 0.05 M Tris-HCl, pH 7.5, 0.01 M EDTA, freshly added 1x
protease inhibitor cocktail (PIC), 0.5 mM PMSF, 0.01 M DTT) and
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245
proteins were resolved on polyacrylamide SDS gels, transferred
to nitrocellulose membranes, and then subjected to immunola-
beling. Primary antibodies used were polyclonal antibodies (pAb)
against CAP2 (Peche et al., 2007) and mAb K82-381-1 specic
for CAP2. Horseradish peroxidase conjugated secondary antibod-
ies were used for detection. mAb against GAPDH conjugated with
horseradish peroxidase (Sigma, St. Louis, MO, USA) was used for
analyzing equal loading. For determining the G/F actin ratio, cells
were washed once in ice-cold PBS and lysed with actin stabilization
buffer (0.1 M PIPES, pH 6.9, 30% glycerol, 5% DMSO, 1 mM MgSO4 ,
1 mM EGTA, 1% TX-100, 1 mM ATP, and protease inhibitor) on ice
for 10 min. Cells were dislodged by scraping and centrifuged at 4 C
for 75 min at 16,000 g. The supernatant (G-actin) and the pellet
(F-actin) fraction were resolved on 12% SDS-PAGE gels, the proteins
transferred to nitrocellulose membranes and probed with mono-
clonal anti--actin antibody (Sigma). Densitometric analysis was
done using Image J for estimation of the cellular G/F-actin ratio.
Focal adhesion assay
Focal adhesion assay was carried out as described by Taranum
et al. (2012). Briey, trypsinised cells were seeded on coverslips
in culture dishes with an initial cell number of 1 103 and sub-
jected to immunouorescence as described above by staining for
vinculin. Analysis was carried out with a confocal laser scanning
microscope TCS-SP5 (Leica) equipped with a very sensitive hybrid
detector (HyD). The individual immunouorescences shown have
the same magnications and were taken in the same z-plane so that
the spreading of focal adhesions on the surface of the coverslip is
comparable. LAS-AF Lite Application Suite software from Leica was
used to quantify the spreading area in m2 .
Disruption of the actin cytoskeleton and recovery
WT and Cap2gt/gt broblasts were plated on coverslips overnight
in 24 well plates in normal growth medium. Next day cells were
washed three times with PBS and 500 l of DMEM medium con-
taining latrunculin B at a concentration of 2.5 M (without FBS and
antibiotics) were added. For solvent control, on a separate cover-
slip medium containing 2.5 l DMSO was added. After a 30 min
incubation in a humidied chamber (5% CO2 , 37 C), the medium
containing latrunculin B was removed and cells were washed three
times with PBS to remove any traces of the drug. Normal growth
medium was added for cell recovery. Cells were xed at various
time points of recovery (10, 20, 30 and 60 min). After permeabiliza-
tion cells were stained with TRITC-phalloidin to visualize F-actin.
Nuclei were visualized using DAPI. Coverslips were mounted and
processed for confocal microscopy.
Quantication of staining
Wound healing was done with 4 WT and 4 mutant mice for
each time point (day 3, day 7) and for day 10, 8 WT and 8 mutant
mice were used. Each mouse had four different wounds. At least 12
sections (three sections per slide) from each mouse wound were
analyzed for quantication. Parafn sections from 3, 7 and/or 10 d
after wounding were stained with specic antibodies as mentioned
(-SMA, F4/80, CD31). To quantify the stainings, uorescence sig-
nal intensity of -SMA, F4/80 or CD31 per unit area was measured
using the quantify function in the TCS SP II confocal software.
CAP2 stainings on human wounds were quantied as uorescence
signal intensity per unit area using TCS SP II confocal software.
Results are presented as mean values standard deviation (S.D.).
For comparison of means between different groups the Students t
test was used.
35
Results
Loss of CAP2 results in delayed wound repair
To test the response of the mutant skin to an injury we
used Cap2gt/gt mice and their WT littermates at the age of 3
months. Full thickness wounds of 5 mm were generated by cir-
cular excision on the shaved back of WT and Cap2gt/gt mice
and the closure was followed over a period of ten days. Macro-
scopic observation of Cap2gt/gt wounds revealed a delayed wound
closure (Fig. 1A). To conrm the macroscopic observations, we
performed H&E stainings on wounds at different stages and
observed that the wound healing was slowed down in Cap2gt/gt
mice (Fig. 1B). At day 3, 7 and 10 the distance between the
wound ends was greater in sections from Cap2gt/gt mice when com-
pared with WT sections (Fig. 1B, C) (Day 0, WT 5000 m, Cap2gt/gt
5000 m; Day 3, WT 2728 588 m, Cap2gt/gt 2958 295 m;
Day 7, WT 1475 449 m, Cap2gt/gt 1606 372 m; Day 10, WT
820 176 m, Cap2gt/gt 1406 188 m Day 10, p = 0.0003). Statis-
tical analysis showed a signicant difference in WT and mutant
wounds at day 10. To assess the dermal architecture we performed
masson trichrome staining that detects the collagen in mutant and
control mice. We did not nd a signicant difference in collagen
content (Fig. 1D).
Cap2gt/gt mice show delayed wound contraction
Restoration of the dermal matrix requires the migration and
proliferation of broblasts and overlaps with reepithelializa-
tion. Fibroblasts in the dermis synthesize extracellular matrix to
strengthen the damaged tissue and subsequently to contract the
granulation tissue. For wound contraction, the broblasts dif-
ferentiate into specialized cells called myobroblasts which are
characterized by stress bers containing smooth muscle actin
(-SMA) (Blumbach et al., 2010; Hinz, 2007; Hinz et al., 2001).
When we investigated the expression of -SMA in the wounds of
WT and Cap2gt/gt mice by immunouorescence analysis at day 7,
a signicantly lower intensity of -SMA staining per unit wound
area in Cap2gt/gt samples was observed indicating that fewer bro-
blasts had differentiated into myobroblasts than in WT which
resulted in delayed wound contraction (Fig. 2A) (-SMA inten-
sity per unit area Day 7, WT 59.67 17.25, Cap2gt/gt 33.54 14.54,
p = 0.0018).
Keratinocyte proliferation is a crucial event for reepithelializa-
tion of the wound and alterations in this process might cause a
delay in wound closure. When we assessed keratinocyte prolifera-
tion with the cell proliferation marker Ki-67 we did not observe
a signicant alteration in the number of Ki-67 positive cells in
Cap2gt/gt wounds compared to the WT wounds (Fig. 2B) (percent-
age Ki-67 positive cells in hyperproliferative epidermis, Day 3 WT
11.39 1.03, Cap2gt/gt 10.87 2.1, Day 7, WT 9.45 0.79, Cap2gt/gt
9.81 1.03).
Cap2gt/gt mice show decreased macrophage inltration and slower
neovascularization
Inammation is an important phase of the wound healing
process, which is followed by reepithelialization. To investigate
whether the inltration of wounds by macrophages is affected by
CAP2 deciency, we analyzed macrophage inltration by F4/80
staining. We found less macrophage inltration at day 3 rela-
tive to the WT wounds (Fig. 3A) (F4/80 intensity per unit area,
WT 21.6 7.899, Cap2gt/gt 13.1 5.462, Day 3, p = 0.044839061).
Although at day 7 we observed a slight increase in macrophage
inltration in Cap2gt/gt wounds, it was not signicant.
36 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245

Fig. 1. Wound healing is altered in Cap2gt/gt mice. A, Macroscopic photos of wounds from WT and Cap2gt/gt mice at day 0, 3, 7 and 10. The dotted black line indicates the
wound area left open at each day. B, Skin sections of WT and Cap2gt/gt were stained with HE (day 3, 7 and 10). The position of the wound margins are indicated by arrows. In
Cap2gt/gt mice, the wound closure was affected in contrast to WT. Scale bar, 250 m. C, The distance between the wound ends during wound healing was measured. At day
10 the distance differs signicantly in WT and Cap2gt/gt (for each time point, 4 mice per genotype, at least 2 wounds per mice and 12 sections per wound; for day 10 time
point, 8 mice per genotype, 2 wounds per mice and 12 sections per wound **p < 0.0005). D, Massons trichrome staining of the wounds. Scale bars, 500 m (day 3), 250 m
(day 7, 10).

Neovascularization is a key event during wound healing in
which the functional microvascular networks develop. Sections
were stained with anti CD31 antibody, a specic marker for
endothelial cells, to evaluate neovascularization by intensity mea-
surement. We found that at day 3 there were less CD31 positive
cells which could contribute to neovascularization (Fig. 3B) (CD31
intensity per unit area, Day 3, WT 32.134 11.565, Cap2gt/gt
17.576 2.607; Day 7, WT 36.73 14.562, Cap2gt/gt 38.35 17.171
Day 3, p = 9.10011 E05) whereas at day 7 we did not nd a
signicant difference between mutant and WT wounds. From
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245 37

Fig. 2. Differentiation of broblasts into myobroblasts and proliferation are reduced in Cap2gt/gt mice. A, In WT and Cap2gt/gt mice, differentiated myobroblasts were
identied by staining for -SMA at day 7. Differentiation of broblasts into myobroblasts appeared reduced during the wound healing process in Cap2gt/gt mice. 4 mice per
genotype, 2 wounds per mice and 12 sections per wound. Scale bar, 50 m; **p < 0.01. B, Sections from WT and Cap2gt/gt wounds were stained with a Ki-67 specic antibody
as marker for keratinocyte proliferation at day 3 and day 7 post wounding. 4 mice per genotype, at least 2 wounds per mice and 12 sections per wound Scale bar, 100 m.

these observations it is clear that lack of CAP2 has an effect
on macrophage inltration and neovascularization during early
phases of wound healing.
Expression of CAP2 in human wounds
Interestingly, the protein was mainly located in the cytosol and less
at the cell periphery whereas in the stratum spinosum where kera-
tinization begins it was also observed at the cell periphery (Fig. 4A).
When we quantied the expression of CAP2 in human wounds by
intensity per unit area, we found a signicant upregulation of CAP2
at day 5, 20 and 30 (Fig. 4B).

Earlier we have reported that CAP2 is present in murine and

human skin where it localizes to the nucleus, the cytosol and also to
the cell periphery (Peche et al., 2007). It remained however unclear
whether CAP2 has any role in human wound healing. To study this
we performed immunohistochemistry on unwounded human skin
and normally healing human wounds obtained from healthy vol-
unteers at different stages after injury. In accordance with earlier
studies (Peche et al., 2007), in unwounded human skin, we found
expression of CAP2 in all living layers of the epidermis where it
was found at the cell periphery and in the nucleus. Interestingly,
in human wounds, CAP2 was also expressed in hyperproliferative
epidermis and at the migrating tongue (Fig. 4A). In hyperprolif-
erative epidermis of wounds at day 5, day 20 and day 30 we
found expression of CAP2 in basal cells which undergo proliferation.
Cultured Cap2gt/gt primary keratinocytes show reduced speed and
unaltered proliferation
During wound healing, keratinocytes play an important role
in the formation of the migrating tongue, its migration and the
wound closure. Therefore we analyzed the migratory behavior
of primary keratinocytes from WT and Cap2gt/gt by scratching
monolayers and monitoring the closure of the scratch over 12 h
using time-lapse video microscopy. Control keratinocytes invaded
the denuded zone and displayed directional migration leading
to the closure of the scratch within 12 h (Fig. 5A). In con-
trast, Cap2gt/gt keratinocytes invaded the denuded area with a
reduced speed and at 12 h the scratch remained partially open
38 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245

Fig. 3. Decreased macrophage inltration and slower neovascularization. A, Analysis of macrophage inltration using F4/80 antibody, a macrophage specic marker. Sig-
nicant reduction in F4/80 positive macrophages in mutant animals inltrating during wound healing process at day 3. B, Immunouorescence analysis with CD31, an
endothelial cell specic marker at day 3 and day 7 post wounding in WT and Cap2gt/gt . 4 mice per genotype, at least 2 wounds per mice and 12 sections per wound. Scale bar,
50 m; *p < 0.05, ***p < 0.001.

(Fig. 5A). We also monitored cell speed and directionality by track-
ing single cells at the migration front. We observed a speed of
0.8933 0.19 m/min with directionality of 0.50 by control kera-
tinocytes versus 0.77 0.22 m/min with directionality of 0.53 by
Cap2gt/gt keratinocytes (Fig. 5B) (p < 0.0001; 130 cells per genotype
using 4 different preparations; the experiments were performed
three times). Thus, isolated Cap2gt/gt keratinocytes show lowered
speed with unaltered directionality.
Additionally, we also calculated the generation time for iso-
lated primary keratinocytes to support our ndings on keratinocyte
proliferation during wound closure. Equal numbers of cells were
seeded and upon reaching conuency, the generation time was cal-
culated. We did not observe any difference for Cap2gt/gt keratinocyte
generation time (Fig. 5C) (52.09 3.69 h for WT and it 51.56 3.21 h
for Cap2gt/gt ).
Cap2gt/gt broblasts show altered cellular properties
To ask whether altered broblast migration contributed to the
wound healing defect we analyzed the migratory behavior of
primary broblasts. WT and Cap2gt/gt broblasts were isolated and
cultured. Immunocytochemistry and western blot analysis were
performed to ensure the loss of CAP2 (Fig. 6A).
To monitor the speed and migratory properties of primary bro-
blasts from mutant animals we performed scratch assays using
monolayers of WT and Cap2gt/gt broblasts. Pictures were taken at
different time points using time lapse video microscopy. We found
that the gap in WT cells had closed completely 20 h after it was
introduced, whereas those in the Cap2gt/gt had closed only partly
after 20 h (Fig. 6B). Also, Cap2gt/gt cells had a signicantly decreased
cell speed compared to control cells as revealed by quantication of
the speed of migration (Fig. 6B) (WT 0.35 0.05 m/min, Cap2gt/gt
0.29 0.03 m/min; p = 0.004438).
Furthermore, we performed collagen gel contraction assay to
analyze the ability of cultured broblasts for collagen lattice con-
traction upon stress development. Fibroblasts were seeded in the
collagen lattice and allowed to grow for 24 h in culture media.
After 24 h the collagen lattices were analyzed and the width was
quantied. We found that Cap2gt/gt cells showed a reduced con-
traction ability as compared to WT cells (Fig. 6C) (percentage area
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245 39

Fig. 4. CAP2 is expressed in human wounds. A, Unwounded skin and human wound (day 5, day 20, day 30 post injury) parafn sections were stained with CAP2 specic
polyclonal antibodies. In unwounded skin CAP2 is found in all living layers of the epidermis (upper panel). Note that CAP2 is present in the cells of the stratum basale, stratum
spinosum and also in the migrating tongue of the wound. Cells marked with asterisk indicate cytosolic CAP2 staining (stratum basale) and also in keratinizing cells (stratus
spinosum) at the cell periphery. Nuclei were stained with propidium iodide (PI). E, Epidermis; D, Dermis; MT, Migrating tongue. Scale bars, 20 m. B, Quantication of CAP2
intensity per unit area in normal skin versus skin at day 5, day 20 and day 30 post injury (**p < 0.001); normally healing human wounds from 2 different healthy volunteers
per time point were analyzed.

Fig. 5. CAP2 regulates keratinocyte migration in vitro. A, In vitro scratch assay was performed on a conuent keratinocyte monolayer and cells were allowed to invade
denuded area. Time-laps video microscopy was performed. Asterisk indicates the uncovered area by mutant broblasts after 12 h of migration. B, Migration speed was
calculated and quantied, p < 0.0001 (n = 130 cells per genotype, experiment was done thrice with four different keratinocyte preparations). C, Generation time of WT and
mutant primary keratinocytes was calculated and quantied.

for WT 60.33 6.97%, percentage area for Cap2gt/gt 74.30 8.47%;
p = 0.0055, N = 2 from 4 different broblast isolations).
To probe whether CAP1 could compensate the loss of CAP2 in
skin, we checked for the expression of CAP1 in unwounded WT and
knockout skin, WT and knockout wounds and in WT and knockout
broblasts. In unwounded WT murine skin CAP1 was expressed
at very low levels and hardly detectable, CAP2 ablation led to an
increase of CAP1. Upon wounding CAP1 amounts increased in the
WT, and in the knockout wounds the already signicant levels
further increased. In WT primary broblasts CAP1 is expressed in
signicant amounts and there is no difference seen between knock-
out and WT (Fig. 6D).
Actin rearrangement plays an important role in cell polar-
ity, migration, and differentiation. Although other factors like
nuclear positioning or transcriptional switches also affect these
processes, the ability of a cell to develop protrusions using the actin
cytoskeleton strongly governs the above-mentioned phenomenon.
To analyze this, we performed phalloidin staining on primary
40 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245

Fig. 6. Analysis of primary Cap2gt/gt and WT broblasts. A, Fibroblasts were xed and stained with mAb K82381-1 for detection of CAP2. FITC Phalloidin was used to visualize
F-actin, DAPI for nuclei. Scale bar, 10 m. Complete deletion of CAP2 was also conrmed by western blot analysis using polyclonal anti-CAP2 antibodies. GAPDH was used as
a loading control. B, Scratch assay revealed a migration defect in Cap2gt/gt primary broblasts. Wound closure was followed by live cell microscopy (020 h after scratching).
The speed of migration for WT and Cap2gt/gt broblasts was determined (m/min) (**p < 0.01 n = 100 cells, per cell type). C, Fibroblasts were seeded in the collagen lattice
and allowed to grow for 24 h in culture media. After 24 h collagen lattice were analyzed using a ruler and quantied. Percentage area for WT 60.33 6.97%, percentage
area for Cap2gt/gt 74.30 8.47%; p = 0.0055, N = 2 from 4 different broblasts preparation. D, CAP1 is upregulated in Cap2gt/gt . WT unwounded skin lysates and lysates from
wounds at day 3 were separated on 12% SDS-PAGE, blotted and probed with CAP1 antibody. Lysates from WT and primary knockout broblasts were also probed with CAP1
specic antibody. GAPDH was used as a loading control. E, Fibroblasts were xed, permeabilized and incubated with TRITC-Phalloidin to stain lamentous actin. Extended
protrusions (indicated by arrows) were observed in Cap2gt/gt cells. The graph depicts the percentage of normal and extended protrusions as visualized by F-actin staining in
WT and Cap2gt/gt broblasts (n = 185 cells per cell type, *p < 0.05). Scale bars, WT, 20 m; Cap2gt/gt , 50 m.
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245 41

Fig. 7. Altered cellular morphology in Cap2gt/gt broblasts. A, Formation of focal adhesions in control and Cap2gt/gt primary broblasts. Fibroblasts were trypsinized and
plated and focal adhesion formation was assessed after 12 h by staining with mAb specic for vinculin. DAPI was used to visualize the nuclei. Scale bar, 20 m Quantication
of focal adhesions per cell. Graph shows increased number of focal adhesions in Cap2gt/gt broblasts compared to WT (Scale bar ***p < 0.001; n = 40 cells per cell type). B,
Vinculin staining of WT and Cap2gt/gt broblasts showed that mutant cells were faster in adherence upon seeding compared to WT. Scale bar, 20 m. C, F-actin organization
is altered in mutant cells as visualized by phalloidin at different time points after seeding. Scale bar, 20 m. D, Total actin content was determined using anti-actin antibody.
Total actin remains unaltered in Cap gt/gt broblasts compared to WT (n = 4). E, Quantication of G/F actin level in cultured primary broblasts showing signicant increase
in F-actin levels (4 experiments from 4 different preparations per genotype).

broblasts and observed a higher number of cells that had extended
protrusions in cells lacking CAP2 as compared to wild type bro-
blasts. We observed a signicantly lower percentage of mutant
broblasts extending normal protrusions (Fig. 6E).
Focal adhesions are altered in Cap2gt/gt broblasts
Since altered cell migration is often associated with altered
cell adhesion properties, we investigated the effects of CAP2
deletion on the formation and size of focal adhesions. Cell adhesion
contributes substantially to the maintenance of tissue structure,
the promotion of cell migration, and the transduction of informa-
tion about the microenvironment of the cell. Focal adhesions are
large macromolecular assemblies through which both mechanical
force and regulatory signals are transmitted (Chen et al., 2003). To
visualize focal adhesions we used mAbs specic for vinculin and
stained cells 12 h post trypsinization and seeding. Immunouores-
cence analysis revealed that Cap2gt/gt cells had formed more focal
42 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245

Fig. 8. Recovery of the actin cytoskeleton upon disruption is altered in Cap2gt/gt broblasts. A, WT and mutant broblasts were treated with latrunculin B (lat B), a drug that
disrupts the actin cytoskeleton. Cap2 gt/gt and WT cells were incubated with latrunculin B (2.5 M) for 30 min and then washed out and followed at different time points.
F-actin was visualized by TRITC phalloidin, nuclei were stained with DAPI. Scale bars, 20 m. B, Quantication of the cells that developed extended protrusions with increased
cell surface contacts upon drug treatment and subsequent washout in mutant and WT broblasts. *p < 0.05, **p < 0.01.

Fig. 9. Model illustrating cellular functions of CAP2. Ablation of CAP2 leads to cytoskeleton defects. CAP2 has a potential to sequester G-actin and also sever F-actin laments.
Increased F-actin content in cells lacking CAP2 and altered reorganization of cortical actin cytoskeleton in mutant cells could be seen. Depletion of CAP2 leads to an increase
in focal adhesion in resting (A) and in migrating cells (B, C). In CAP2 knockout, cell motility is affected either by stabilization of focal adhesions and/or disruption of cell
polarity (B, C). Dense meshwork of peripheral actin laments with less stress bers may lead to reduced cell motility (B) accounting for altered wound healing response and
contraction in vivo.
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245
adhesions than WT (Fig. 7A) (focal adhesions per cell: WT, 72 28,
Cap2gt/gt , 102 30, p = 0.000000009). As the number of focal adhe-
sions was enhanced in Cap2gt/gt broblasts, we further analyzed
the development of focal adhesions. To evaluate the development
of focal adhesions upon seeding, we followed the WT and mutant
broblasts at several time points for focal adhesions through vin-
culin staining. We observed that at 40 min post seeding mutant
broblasts were atter and started to develop focal adhesions,
whereas this process for WT occurred at 90 min. Mutant broblasts
exhibited at every time point a larger area of spreading on the sub-
stratum than WT (Fig. 7B). This prompted us to visualize F-actin
during progressive attachment. Similarly, we found that mutant
broblasts were faster in spreading and in rearrangement of the
F-actin cytoskeleton visualized through FITC-phalloidin (Fig. 7C).
We excluded that this was due to alterations of the total actin con-
tent in mutant broblasts as in western blot analysis we did not
observe a change in the total actin content in mutant broblasts
when compared with WT broblasts (Fig. 7D). Since CAP2 is a pro-
tein regulating actin polymerization we determined the G/F-actin
ratio. We found an increased F-actin content in mutant broblasts
when compared to WT (relative reduction in G/F actin ratio 20%)
(Fig. 7E). This points at a differential actin regulation upon loss of
CAP2. Thus, ablation of CAP2 results in increased focal adhesion,
rapid development of focal adhesions and increased F-actin content
whereas the total actin content was unaltered.
Recovery of the actin cytoskeleton is faster in mutant broblasts
Many cellular processes require a balanced reorganization of the
actin cytoskeleton. To follow actin reorganization upon disruption
we treated cells with latrunculin B which disassembles the actin
cytoskeleton and followed the recovery after removal of the drug.
Treatment was with 2.5 M latrunculin B for 30 min which led to
complete disruption of the F-actin network (Fig. 8A). After washout,
the cortical cytoskeleton was seen already at the 10 min time point
of post recovery in mutant cells indicating a faster rearrangement.
For WT broblast this stage was observed at 30 min post recovery.
However, we noted a disturbed development of actin stress bers in
mutant broblasts when compared to the WT control. Additionally
we noticed more abnormal protrusions with increased cell-surface
contacts (Fig. 8B, marked with asterisk) (% of cells with abnormal
protrusions, 10 min, WT 23 1.414, Cap2gt/gt 41 1.414, 20 min,
WT 30 1.414, Cap2gt/gt 43 1.767, 10 min, p = 0.006116; 20 min,
p = 0.044144). This was reminiscent of the observation that mutant
broblasts had formed increased focal adhesions as visualized with
vinculin staining (Fig. 7A, B).
Discussion
CAPs are highly conserved proteins which play a role in vari-
ous cellular activities. Knockout of CAP in Dictyostelium discoideum
and in yeast has revealed functions of CAP in cell polarity and cell
migration (Noegel et al., 2004; Vojtek et al., 1991). We reported pre-
viously that CAP2 did not only sequester actin but also sever actin
laments. The rst mouse knockout of a CAP developed a cardiomy-
opathy and the deletion of CAP2 was lethal in postnatal stages
(Peche et al., 2007, 2013). Here we describe the expression of CAP2
in human wounds in the hyperproliferative epidermis suggesting a
role in the normal wound healing process. Furthermore we charac-
terize functions of CAP2 showing that it is important in the wound
healing process as its ablation impairs macrophage inltration and
neovascularization and wound contraction. Notably there is only
a delay and not the absence of an event and the defects in in vivo
wound healing which we observed were comparatively mild. One
possible explanation could be compensation by CAP1, which was
43
upregulated in the wounded skin of CAP2 decient mice and also
during wound healing.
At the amino acid level, the N-terminal domain of CAP1 and
CAP2 show less similarity whereas the C terminal domain is highly
conserved (Peche et al., 2007). Interestingly, the C-terminal domain
of CAPs has been associated with actin binding and remodeling.
This might be relevant in wound healing where CAP1 may have the
potential to compensate CAP2 with regard to its role in the actin
cytoskeleton. CAP1 was reported as differentially regulated gene
in sciatic nerve crush (SNC). In normal sciatic nerve the expression
level of CAP1 was low but upon injury it was highly expressed dur-
ing the different stages of recovery. It was highest at day 5 post
SNC and slowly decreased through a 4 weeks span. Schwann cells,
which undergo differentiation, are important and support success-
ful regeneration of axons. Through knockdown and differentiation
studies, CAP1 was shown to be important in differentiation of
Schwann cells (Zhu et al., 2014). The regulation of CAP2 in Schwann
cells upon CAP1 knock down and sciatic nerve injury has not been
reported so far. Interestingly when we quantied the CAP2 inten-
sity per unit area in human wounds we observed a signicant
increase in CAP2 expression at day 5, 20 and 30 post injury which
emphasizes the role of CAP2 in the healing process. In general, our
data and reports for CAP1 indicate that the CAP family of proteins
plays an essential role in the healing processes post injury.
At the single cell level, primary keratinocytes and broblasts
showed signicantly altered cellular properties. At the organis-
mal level we observed that macrophage inltration was delayed
in wounds of CAP2 decient mice. This might result in delays in the
following steps of reepithelialization and wound closure. Interest-
ingly, depletion of macrophages using antisera resulted in delayed
wound healing (DiPietro and Polverini, 1993; Eming et al., 2007;
Leibovich and Ross, 1975). Macrophage inltration at the wound
site is a complex and highly orchestrated event. It is regulated by
gradients of different chemotactic factors, including growth fac-
tors, proinammatory cytokines and chemokines (Eming et al.,
2007; Frank et al., 2000; Wetzler et al., 2000). The major source
of these chemoattractants includes platelets trapped in the brin
clot at the wound surface and hyperproliferative keratinocytes
at the wound edge. Macrophages play an important role in the
healing process synthesizing potent growth factors like TGF-, TGF-
, basic broblast growth factor, platelet derived growth factor
and vascular endothelial growth factor which promote cell pro-
liferation and synthesis of extracellular matrix by resident skin
cells. Actin binding proteins Filamin A and drebrin were shown
to be involved as modulator of the chemokines signaling path-
way through an interaction with CXCR4. Filamin A is proposed as
an adaptor protein whereas drebrin was involved in the recruit-
ment of CXCR4 (Jimnez-Baranda et al., 2007; Prez-Martnez et al.,
2010). Further, CAP mutants of D. discoideum had altered chemo-
tactic signaling (Noegel et al., 2004). The observed delayed wound
closure in mutant mice coupled with earlier data from various stud-
ies in D. discoideum CAP mutants points at a possible role of CAP2 in
chemokine release and/or chemokine sensing by numerous recep-
tors and its absence results in delayed macrophage inltration and
thus altered wound healing.
CAP2 loss also led to altered migratory properties of pri-
mary keratinocytes. In migration experiments we observed a
decrease in cellular velocity. Proliferation remained unaffected.
Taking these results into account one would expect a delay in
the closure of the epithelial tips in in vivo wound healing. Sur-
prisingly we did not observe such a defect. Similar ndings were
made in RhoA null mice, which showed normal wound closure
upon wounding whereas cultured primary keratinocytes showed
a reduced speed of migration (Jackson et al., 2011). This indi-
cates that alterations in physiological interactions of keratinocytes
with numerous factors like extracellular matrix, other cells and
44 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245
soluble factors could inuence the role of CAP2 during keratinocyte
migration.
Fibroblasts regulate the turnover of ECM under normal
conditions. In injured tissues, broblasts differentiate into myo-
broblasts which contract and participate in the healing process
by reducing the size of the wound and secreting ECM proteins. The
differentiation of broblasts into myobroblasts is a key event in
connective tissue wound healing (Grotendorst et al., 2004; Li and
Wang, 2011). At day 10 which is the phase when myobroblasts
contribute to the wound healing process Cap2gt/gt mice showed a
signicant delay. Loss of CAP2 delayed or altered myobroblast dif-
ferentiation as indicated by the reduced levels of -SMA from day
7.
We propose that the altered properties of the mutant pri-
mary keratinocytes and broblasts cells contribute to observed
wound healing defects. Cap2gt/gt broblasts develop abnormal
protrusions and more focal adhesions and show reduced veloc-
ity. The abnormal protrusion formation may lead to the altered
velocity observed in mutant broblasts. Focal adhesions play
an important role in cell spreading and cell attachment. Focal
adhesions are important for stability and subsequent movement
of a cell in a 3D environment. We observed a higher number
of focal adhesions in mutant broblasts, which may contribute
to the delay in wound closure. During reepithelialization cellu-
lar movement is a critical parameter and reduced cell velocity
may contribute to the delay in wound healing. Moreover, latrun-
culin B washout experiment revealed an important function of
CAP2 in organization of the actin cytoskeleton, which is highly
relevant in cell migration. Cellular morphology and protrusive
outgrowth is controlled by actin cytoskeleton rearrangement.
Rho-GTPases are major regulators of the cytoskeleton which
coordinate the growth of lamellipodia (Rac1), lopodia (Cdc42)
and stress bers (RhoA). The role of CAP in the focal adhesion
complex was proposed to be through FAK and Talin. Activated
adhesion complexes have a potential to regulate Rac/Cdc42
signaling which are well documented drivers of cell migra-
tion (Zhang et al., 2013). Extended and abnormal protrusions in
CAP2 null cells might be a result of altered Rho-GTPases activ-
ities. This could ultimately affect the cell migration which we
observed with primary cells. This is an interesting topic for further
research.
From our in vivo and in vitro data we propose a model
for CAP2 functions in wounded skin. Depletion of CAP2 leads
to an increase in focal adhesion in resting (Fig. 9A) and in
migrating cells (Fig. 9B, C). In CAP2 knockout cells motility
is affected either by stabilization of focal adhesions and/or
disruption of cell polarity. A dense meshwork of peripheral
actin laments with fewer stress bers may also lead to
reduced cell motility leading to the altered wound healing pro-
cess.
Acknowledgements
KK was supported by the International Graduate School in
Development Health and Disease (IGSDHD), Cologne. VSP acknowl-
edges support by Kln Fortune and the Maria Pesch Foundation.
We thank Dr. Raphael Rastetter for his help in scratch assays and
Berthold Gaen and Rolf Mller for help in antibody generation.
There are no nancial disclosures or conicts of interest existing
for any of the authors.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ejcb.2014.10.004.
References
Almine, J.F., Wise, S.G., Weiss, A.S., 2012. Elastin signaling in wound repair. Birth
Defects Res. C: Embryo Today Rev. 96, 248257, http://dx.doi.org/10.1002/
bdrc.21016.
Baum, B., Li, W., Perrimon, N., 2000. A cyclase-associated protein regulates
actin and cell polarity during Drosophila oogenesis and in yeast. Curr. Biol.,
http://dx.doi.org/10.1016/S0960-9822(00)00640-0.
Benlali, A., Draskovic, I., Hazelett, D.J., Treisman, J.E., 2000. act up controls actin poly-
merization to alter cell shape and restrict Hedgehog signaling in the Drosophila
eye disc. Cell 101, 271281, http://dx.doi.org/10.1016/S0092-8674(00)80837-5.
Bertling, E., Hotulainen, P., Mattila, P.K., Matilainen, T., Salminen, M., Lappalainen,
P., 2004. Cyclase-associated protein 1 (CAP1) promotes colin-induced actin
dynamics in mammalian nonmuscle cells. Mol. Biol. Cell 15, 23242334,
http://dx.doi.org/10.1091/mbc.E04-01-0048.
Blumbach, K., Zweers, M.C., Brunner, G., Peters, A.S., Schmitz, M., Schulz, J.-N., Schild,
A., Denton, C.P., Sakai, T., Fssler, R., Krieg, T., Eckes, B., 2010. Defective granula-
tion tissue formation in mice with specic ablation of integrin-linked kinase in
broblasts role of TGF1 levels and RhoA activity. J. Cell Sci. 123, 38723883,
http://dx.doi.org/10.1242/jcs.063024.
Chen, C.S., Alonso, J.L., Ostuni, E., Whitesides, G.M., Ingber, D.E., 2003. Cell shape
provides global control of focal adhesion assembly. Biochem. Biophys. Res. Com-
mun. 307, 355361, http://dx.doi.org/10.1016/S0006-291X(03)01165-3.
DiPietro, L.A., Polverini, P.J., 1993. Angiogenic macrophages produce the angiogenic
inhibitor thrombospondin 1. Am. J. Pathol. 143, 678684.
Eming, S.A., Krieg, T., Davidson, J.M., 2007. Inammation in wound repair:
molecular and cellular mechanisms. J. Invest. Dermatol. 127, 514525,
http://dx.doi.org/10.1038/sj.jid.5700701.
Fedor-Chaiken, M., Deschenes, R.J., Broach, J.R., 1990. SRV2, a gene required
for RAS activation of adenylate cyclase in yeast. Cell 61, 329340,
http://dx.doi.org/10.1016/0092-8674(90)90813-T.
Frank, S., Kmpfer, H., Wetzler, C., Stallmeyer, B., Pfeilschifter, J., 2000. Large induc-
tion of the chemotactic cytokine RANTES during cutaneous wound repair: a
regulatory role for nitric oxide in keratinocyte-derived RANTES expression.
Biochem. J. 347 (Pt 1), 265273, http://dx.doi.org/10.1042/0264-6021:3470265.
Gieselmann, R., Mann, K., 1992. ASP-56, a new actin sequestering protein from
pig platelets with homology to CAP, an adenylate cyclase-associated pro-
tein from yeast. FEBS Lett. 298, 149153, http://dx.doi.org/10.1016/0014-
5793(92)80043-G.
Grotendorst, G.R., Rahmanie, H., Duncan, M.R., 2004. Combinatorial signaling
pathways determine broblast proliferation and myobroblast differentiation.
FASEB J. 18, 469479, http://dx.doi.org/10.1096/fj.03-0699com.
Hinz, B., 2007. Formation and function of the myobroblast during tissue repair. J.
Invest. Dermatol. 127, 526537, http://dx.doi.org/10.1038/sj.jid.5700613.
Hinz, B., Mastrangelo, D., Iselin, C.E., Chaponnier, C., Gabbiani, G., 2001. Mechanical
tension controls granulation tissue contractile activity and myobroblast dif-
ferentiation. Am. J. Pathol. 159, 10091020, http://dx.doi.org/10.1016/S0002-
9440(10)61776-2.
Jackson, B., Peyrollier, K., Pedersen, E., Basse, A., Karlsson, R., Wang, Z., Lefever,
T., Ochsenbein, A.M., Schmidt, G., Aktories, K., Stanley, A., Quondamatteo, F.,
Ladwein, M., Rottner, K., van Hengel, J., Brakebusch, C., 2011. RhoA is dispensable
for skin development, but crucial for contraction and directed migra-
tion of keratinocytes. Mol. Biol. Cell 22, 593605, http://dx.doi.org/10.1091/
mbc.E09-10-0859.
Jimnez-Baranda, S., Gmez-Moutn, C., Rojas, A., Martnez-Prats, L., Mira, E., Ana
Lacalle, R., Valencia, A., Dimitrov, D.S., Viola, A., Delgado, R., Martnez-A, C.,
Manes, S., 2007. Filamin-A regulates actin-dependent clustering of HIV recep-
tors. Nat. Cell Biol. 9, 838846, http://dx.doi.org/10.1038/ncb1610.
Leibovich, S.J., Ross, R., 1975. The role of the macrophage in wound repair. Am. J.
Pathol. 78, 71100.
Li, B., Wang, J.H.C., 2011. Fibroblasts and myobroblasts in wound heal-
ing: force generation and measurement. J. Tissue Viability 20, 108120,
http://dx.doi.org/10.1016/j.jtv.2009.11.004.
Noegel, A.A., Blau-Wasser, R., Sultana, H., Mller, R., Israel, L., Schleicher, M., Patel,
H., Weijer, C.J., 2004. The cyclase-associated protein CAP as regulator of cell
polarity and cAMP signaling in Dictyostelium. Mol. Biol. Cell 15, 934945,
http://dx.doi.org/10.1091/mbc.E03-05-0269.
Nomura, K., Ono, K., Ono, S., 2012. CAS-1, a C. elegans cyclase-associated pro-
tein, is required for sarcomeric actin assembly in striated muscle. J. Cell Sci.,
http://dx.doi.org/10.1242/jcs.104950.
Nomura, K., Ono, S., 2013. ATP-dependent regulation of actin monomer-lament
equilibrium by cyclase-associated protein and ADF/colin. Biochem. J. 453,
249259, http://dx.doi.org/10.1042/BJ20130491.
Peche, V., Shekar, S., Leichter, M., Korte, H., Schrder, R., Schleicher, M., Holak, T.A.,
Clemen, C.S., Ramanath-Y, B., Ptzer, G., Karakesisoglou, I., Noegel, A.A., 2007.
CAP2, cyclase-associated protein 2, is a dual compartment protein. Cell. Mol. Life
Sci. 64, 27022715, http://dx.doi.org/10.1007/s00018-007-7316-3.
Peche, V.S., Holak, T.A., Burgute, B.D., Kosmas, K., Kale, S.P., Wunderlich, F.T.,
Elhamine, F., Stehle, R., Ptzer, G., Nohroudi, K., Addicks, K., Stckigt, F.,
Schrickel, J.W., Gallinger, J., Schleicher, M., Noegel, A.A., 2013. Ablation of
cyclase-associated protein 2 (CAP2) leads to cardiomyopathy. Cell. Mol. Life Sci.
70, 527543, http://dx.doi.org/10.1007/s00018-012-1142-y.
Prez-Martnez, M., Gordn-Alonso, M., Cabrero, J.R., Barrero-Villar, M., Rey, M.,
Mittelbrunn, M., Lamana, A., Morlino, G., Calabia, C., Yamazaki, H., Shi-
rao, T., Vzquez, J., Gonzlez-Amaro, R., Veiga, E., Snchez-Madrid, F., 2010.
 K. Kosmas et al. / European Journal of Cell Biology 94 (2015) 3245
F-actin-binding protein drebrin regulates CXCR4 recruitment to the immune
synapse. J. Cell Sci. 123, 11601170, http://dx.doi.org/10.1242/jcs.064238.
Schfer, M., Werner, S., 2008. Cancer as an overhealing wound: an
old hypothesis revisited. Nat. Rev. Mol. Cell Biol. 9, 628638,
http://dx.doi.org/10.1038/nrm2455.
Shibata, R., Mori, T., Du, W., Chuma, M., Gotoh, M., Shimazu, M., Ueda, M.,
Hirohashi, S., Sakamoto, M., 2006. Overexpression of cyclase-associated pro-
tein 2 in multistage hepatocarcinogenesis. Clin. Cancer Res. 12, 53635368,
http://dx.doi.org/10.1158/1078-0432.CCR-05-2245.
Singer, A.J., Clark, R.A., 1999. Cutaneous wound healing. N. Engl. J. Med. 341, 738746,
http://dx.doi.org/10.1056/NEJM199909023411006.
Sultana, H., Rivero, F., Blau-Wasser, R., Schwager, S., Balbo, A., Bozzaro, S., Schleicher,
M., Noegel, A.A., 2005. Cyclase-associated protein is essential for the functioning
of the endo-lysosomal system and provides a link to the actin cytoskeleton.
Trafc 6, 930946, http://dx.doi.org/10.1111/j. 1600-0854.2005.00330.x.
Taranum, S., Vaylann, E., Meinke, P., Abraham, S., Yang, L., Neumann, S.,
Karakesisoglou, I., Wehnert, M., Noegel, A.A., 2012. LINC complex alter-
ations in DMD and EDMD/CMT broblasts. Eur. J. Cell Biol. 91, 614628,
http://dx.doi.org/10.1016/j.ejcb.2012.03.003.
Trott, A., 1988. Mechanisms of surface soft tissue trauma. Ann. Emerg. Med. 17,
12791283, http://dx.doi.org/10.1016/S0196-0644(88)80353-6.
45
Vojtek, A., Haarer, B., Field, J., Gerst, J., Pollard, T.D., Brown, S., Wigler, M., 1991.
Evidence for a functional link between prolin and CAP in the yeast S. cerevisiae.
Cell 66, 497505, http://dx.doi.org/10.1016/0092-8674(81)90013-1.
Wang, C., Zhou, G.-L., Vedantam, S., Li, P., Field, J., 2008. Mitochondrial shutt-
ling of CAP1 promotes actin- and colin-dependent apoptosis. J. Cell Sci. 121,
29132920, http://dx.doi.org/10.1242/jcs.023911.
Wetzler, C., Kampfer, H., Stallmeyer, B., Pfeilschifter, J., Frank, S., 2000. Large
and sustained induction of chemokines during impaired wound healing in
the genetically diabetic mouse: prolonged persistence of neutrophils and
macrophages during the late phase of repair. J. Invest. Dermatol. 115, 245253,
http://dx.doi.org/10.1046/j. 1523-1747.2000.00029.x.
Zhang, H., Ghai, P., Wu, H., Wang, C., Field, J., Zhou, G.L., 2013. Mammalian
adenylyl cyclase-associated protein 1 (CAP1) regulates colin function, the
actin cytoskeleton, and cell adhesion. J. Biol. Chem. 288, 2096620977,
http://dx.doi.org/10.1074/jbc.M113.484535.
Zhu, X., Yao, L., Guo, A., Li, A., Sun, H., Wang, N., Liu, H., Duan, Z., Cao, J.,
2014. CAP1 was associated with actin and involved in Schwann cell differ-
entiation and motility after sciatic nerve injury. J. Mol. Histol. 45, 337348,
http://dx.doi.org/10.1007/s10735-013-9554-z.