Protein Z Exerts Pro-Angiogenic Effects and Upregulates CXCR4

Abstract

Protein Z (PZ) is a vitamin K-dependent coagulation factor without catalytic activity. Evidence points towards PZ as an independent risk factor for the occurrence of human peripheral arterial disease. However, the role of PZ in ischemia-driven angiogenesis and vascular healing processes has not been elucidated so far.
Angiogenic potency of PZ was assessed in established in vitro assays using endothelial cells. PZ-deficient (PZ-/-) mice and their wild-type littermates (PZ+/+) were subjected to hindlimb ischemia. Furthermore, PZ-/- mice were exposed to PZ expressing adenovirus (AdV-PZ) or control adenovirus (AdV-GFP). In an additional set of animals, PZ-/- mice were exposed to AdV-PZ and AdV-GFP, each in combination with the CXCR4 antagonist AMD3100.
In vitro, PZ stimulated migratory activity and capillary-like tube formation of endothelial cells comparable to SDF-1. PZ-/- mice exhibited diminished hypoxia-driven neovascularization and reperfusion in post-ischemic hindlimbs, which was restored by adenoviral gene transfer up to levels seen in PZ+/+ mice. The stimulatory impact of PZ on endothelial cells in vitro was abolished by siRNA targeting against PZ and PZ was not able to restore reduced migration after knock-down of CXCR4. The increased surface expression of CXCR4 on PZ-stimulated endothelial cells and the abrogated restoration of PZ-/- mice via AdV-PZ after concomitant treatment with the CXCR4 antagonist AMD3100 supports the idea that PZ mediates angiogenesis via a G-protein coupled pathway and involves the SDF-1/CXCR4 axis. This is underlined by the fact that addition of the G-protein inhibitor PTX to PZ-stimulated endothelial cells abolished the effect of PZ on capillary-like tube formation.
The results of the current study reveal a role of PZ in ischemia-induced angiogenesis, which involves a G-protein coupled pathway and a raised surface expression of CXCR4. Our findings thereby extend the involvement of PZ from the coagulation cascade to a beneficial modulation of vascular homeostasis.

Full text

RESEARCH ARTICLE
Protein Z Exerts Pro-Angiogenic Effects
and Upregulates CXCR4
Antje Butschkau
1
*, Nana-Maria Wagner
2
, Berit Genz
1
, Brigitte Vollmar
1
1. Institute for Experimental Surgery, University Hospital Rostock, Rostock, Germany, 2. Clinic for
Anesthesiology and Critical Care Medicine, University Hospital Rostock, Rostock, Germany
*antje.butschkau@uni-rostock.de
Abstract
Objective: Protein Z (PZ) is a vitamin K-dependent coagulation factor without
catalytic activity. Evidence points towards PZ as an independent risk factor for the
occurrence of human peripheral arterial disease. However, the role of PZ in
ischemia-driven angiogenesis and vascular healing processes has not been
elucidated so far.
Approach: Angiogenic potency of PZ was assessed in established in vitro assays
using endothelial cells. PZ-deficient (PZ
2/2
) mice and their wild-type littermates
(PZ
+/+
) were subjected to hindlimb ischemia. Furthermore, PZ
2/2
mice were
exposed to PZ expressing adenovirus (AdV-PZ) or control adenovirus (AdV-GFP).
In an additional set of animals, PZ
2/2
mice were exposed to AdV-PZ and AdV-GFP,
each in combination with the CXCR4 antagonist AMD3100.
Results: In vitro, PZ stimulated migratory activity and capillary-like tube formation
of endothelial cells comparable to SDF-1. PZ
2/2
mice exhibited diminished
hypoxia-driven neovascularization and reperfusion in post-ischemic hindlimbs,
which was restored by adenoviral gene transfer up to levels seen in PZ
+/+
mice. The
stimulatory impact of PZ on endothelial cells in vitro was abolished by siRNA
targeting against PZ and PZ was not able to restore reduced migration after knock-
down of CXCR4. The increased surface expression of CXCR4 on PZ-stimulated
endothelial cells and the abrogated restoration of PZ
2/2
mice via AdV-PZ after
concomitant treatment with the CXCR4 antagonist AMD3100 supports the idea that
PZ mediates angiogenesis via a G-protein coupled pathway and involves the SDF-
1/CXCR4 axis. This is underlined by the fact that addition of the G-protein inhibitor
PTX to PZ-stimulated endothelial cells abolished the effect of PZ on capillary-like
tube formation.
Conclusions: The results of the current study reveal a role of PZ in ischemia-
induced angiogenesis, which involves a G-protein coupled pathway and a raised
OPEN ACCESS
Citation: Butschkau A, Wagner N-M, Genz B,
Vollmar B (2014) Protein Z Exerts Pro-Angiogenic
Effects and Upregulates CXCR4. PLoS ONE 9(12):
e113554. doi:10.1371/journal.pone.0113554
Editor: Ingo Ahrens, University Hospital Medical
Centre, Germany
Received: May 15, 2014
Accepted: October 28, 2014
Published: December 4, 2014
Copyright: ß2014 Butschkau et al. This is an
open-access article distributed under the terms of
the Creative Commons Attribution License, which
permits unrestricted use, distribution, and repro-
duction in any medium, provided the original author
and source are credited.
Data Availability: The authors confirm that all data
underlying the findings are fully available without
restriction. All relevant data are within the paper
and its Supporting Information files.
Funding: This work was supported by a grant of
the Deutsche Forschungsgemeinschaft DFG (VO
450/11-1). BV received the funding. The funders
had no role in study design, data collection and
analysis, decision to publish, or preparation of the
manuscript.
Competing Interests: The authors have declared
that no competing interests exist.
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 1/18

surface expression of CXCR4. Our findings thereby extend the involvement of PZ
from the coagulation cascade to a beneficial modulation of vascular homeostasis.
Introduction
Protein Z (PZ) is a 62 kDa vitamin K-dependent coagulation glycoprotein with a
molecular structure similar to those of factor VII, IX, X and protein C [1] and a
biological half-life of 2.5 days [2]. PZ was initially identified by Prowse and Esnouf
in bovine plasma in 1977 [3] and the human homologue was subsequently
isolated in 1984 [2]. In contrast to the serine protease zymogens, PZ lacks catalytic
activity [4] and serves as a cofactor for the protein Z-dependent protease inhibitor
(ZPI), a 72 kDa member of the serpin superfamily of protease inhibitors [5,6]. PZ
and ZPI deficiency have been shown to enhance thrombosis in mouse models
[7,8]. Whether PZ and ZPI are involved in clinical thrombotic disease is
controversially discussed, with some, but not all, studies suggesting a relation. In
humans, low PZ-levels are associated with the occurrence of deep vein thrombosis
[9] and increased risk of ischemic stroke [10]. In 2007, Sofi et al. observed an
association between low PZ-levels and both the occurrence and the severity of
peripheral arterial disease (PAD) in a case control study, postulating evidence for
a role of PZ in the pathogenesis of atherosclerotic disease [11]. In 2009 they could
confirm these results by another case-control study, demonstrating again a
significant association of low PZ-levels with the occurrence and severity of PAD
[12].
Several studies point towards a role of coagulation factors in angiogenic
processes. After first evidence for a relevance of heparin cofactor II in the
development of atherosclerosis [13], Ikeda and coworkers found that heparin
cofactor II potentiates the activation of vascular endothelial cells and promotes
angiogenesis in response to hindlimb ischemia via an AMP-activated protein
kinase-endothelial nitric-oxide synthase signaling pathway [14]. Protein C, a
structural homologue to PZ, was shown to be associated with lower leg ulcers in
patients with diabetes, when protein C plasma levels were diminished [15].
Already in the early 1990’s, endothelial cells were identified as a source for protein
C[
16]. Uchiba et al. found that activated Protein C (aPC) induces endothelial cell
proliferation in vitro and angiogenesis in mouse cornea in vivo [17]. Moreover, it
was shown that aPC promotes the barrier function of HUVECs by utilization of
the angiopoietin/Tie2 axis [18,19].
Like protein C, PZ is expressed by endothelial cells and immunolocalized in
human vessel sections, where the labeling was found stronger on arterial than
venular endothelium [20]. Though reasonable, the role of PZ for angiogenesis and
neovascularization has not been elucidated so far. In the present study, we
investigated the modulatory capacity of PZ on ischemia-induced neovascular-
ization employing PZ-deficient mice and explored the response of endothelial
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 2/18

cells to stimulation with physiological concentrations of PZ in vitro. Moreover, we
identified the receptor of the stromal cell-derived factor-1 (SDF-1) CXCR4 as
being part of the molecular mechanism underlying the effects of PZ on endothelial
cells.
Materials and Methods
Cell Culture of Endothelial Cells
HUVECs were purchased from Lonza (Basel, Switzerland) and cultivated in
endothelial basal medium with supplements and growth factors (EBM-2 and
EGM-2, Lonza, Basel, Switzerland) supplemented with 10% fetal calf serum on
gelatin-coated dishes (Attachment Factor, Gibco, Germany). Cells were harvested
by trypsinization (0.05%, Gibco, Germany) and used from passage 2 to 5.
Matrigel Angiogenesis Assay
As described previously [21], 1610
4
HUVECs were incubated with PZ (3 mg/ml;
Sigma Aldrich, St. Louis, MO, USA/Enzyme Research Laboratories, South Bend,
IN, USA), SDF-1 (50 ng/ml; R&D Systems, Minneapolis, MN, USA; #350-NS-
010) and Pertussis Toxin (PTX, 100 ng/ml; Tocris Bioscience, Bristol, UK) in
duplicate in 100 mL endothelial growth medium for 8 hours in 96-well plates
precoated with 70 mL Matrigel Basement Membrane Matrix (BD Bioscience,
USA). Tubular HUVEC structures were photographed using a fluorescence
microscope (Leica, Germany) employing 1006magnification at 8 random high
power fields (HPF) per variant. Tubular length was assessed per high-power field
employing ImageProPlus Software, (CA, USA). Per independent experiment,
mean values of all variants were expressed as relative to control (ctrl51.0).
Migration Assay
HUVECs were grown on 12-well plates until confluence. A 10 mL pipette tip was
employed to scratch over well plates twice vertically and horizontally for obtaining
four 90˚crosses on each well. Wells were gently washed and 2 mL of medium
containing PZ (3 mg/ml), SDF-1 (50 ng/ml) or vehicle (aqueous glycerol solution)
were added in triplicate per variant. All scratched-crosses were photographed
every other hour until a total of 24 hours of incubation and scratch wounds were
analyzed employing ImageProPlus Software. Per independent experiment, mean
values of all variants were expressed as relative to control (ctrl51.0).
Flow cytometry analysis
Following incubation with PZ (3 mg/ml) and vehicle, HUVECs were detached,
washed and resuspended in 0.5% BSA in PBS in a concentration of 1610
6
cells
per ml. A volume of 100 ml cell suspension was incubated with 10 ml of PE-
conjugated monoclonal antibody against human CXCR4 or control IgG (both
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 3/18

R&D Systems) and analyzed employing a BD Biosciences Becton Dickinson flow
cytometer FACSCalibur (Heidelberg, Germany) after 8 and 24 hours of
stimulation.
Knock-down of CXCR4 and PZ in HUVECs
For knock-down of CXCR4- and PZ, 70–80% confluent HUVECs in a 6-well or
12-well plate were transfected with CXCR4 siRNA (sc-35421, Santa Cruz
Biotechnology, USA) or PZ siRNA (sc-106450, Santa Cruz Biotechnology, USA)
employing lipofectamine (Invitrogen). Cells were transfected under antibiotic-free
and serum-reduced conditions for 6 hours. Then, medium was changed and cells
were incubated for another 24 hours according to the manufactur’s instructions.
In the following, cells were used for employing the migration assay and matrigel
angiogenesis assay. HUVECs transfected with ctrl siRNA (sc-37007, Santa Cruz
Biotechnology, USA) and un-transfected cells served as controls.
Immunofluorescence staining and confocal laser scanning
microscopy
HUVECs were seeded in gelatin-coated 8 well m-slides (ibidi GmbH, Martinsried,
Germany) at a density of 5610
3
cells per well. Cells were serum-starved for
24 hours before stimulation with PZ (3 mg/ml) or SDF-1 (50 ng/ml) in serum-
and growth-factor reduced medium (containing 5% FCS). After 8 and 24 hours of
stimulation cells were fixed with 4% paraformaldehyde, permeabilized with 0.1%
Triton-X-100 and blocked with 1% bovine serum albumin in PBS for 1 hour.
Immunofluorescence staining was performed with primary antibody anti-CXCR4
(1:50, MAB21651, R&D Systems) over night at 4˚
C followed by incubation with
secondary antibody goat-anti-rat-Alexa
555
(1:400, Life Technologies GmbH,
Darmstadt, Germany) for 1 hour at room temperature. Samples without primary
antibodies served as negative control. Additionally, nuclei were stained with DAPI
(1:1000; AppliChem, Darmstadt, Germany) for 10 min at room temperature. The
fluorescence signals were visualized by using a confocal laser scanning microscope
(LSM 780 ELYRA PS.1 microscope, Carl Zeiss Microscopy GmbH, Jena,
Germany).
Mice
The experiments were conducted in accordance with the guidelines for the Care
and Use of Laboratory Animals and the Institutional Animal Care and Use
Committee (University of Rostock, Medical Faculty, Rostock, Germany; 7221.3-1-
055/13). PZ-deficient mice (PZ
2/2
) in a C57Bl/66129 genetic background, as
described by Yin et al. and Zhang et al., were compared to their respective wild-
type littermates (PZ
+/+
). Representative image and method of PCR for genotyping
of PZ mice can be found in (Figure S1). Male mice were used at an age of 2–4
months and a body weight of 25–30 g. Animals were kept on water and standard
laboratory chow ad libitum.
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 4/18

Ethic statement
All experiments were approved by the local government (Landesamt fu
¨r
Landwirtschaft, Lebensmittelsicherheit und Fischerei Mecklenburg-Vorpommern
LALLF M-V/TSD/7221.3-1-055/13-1) and performed in accordance with the
German legislation on protection of animals and the National Institutes of
‘‘Health Guide for the ‘‘Health Guide for the Care and Use of Laboratory Animals’’
(Institute of Laboratory Animal Resources, National Research Council; NIH
publication 86-23 revised 1985).
Experimental groups
To verify the contribution of PZ in murine hindlimb ischemia, PZ
+/+
and PZ
2/2
mice were used. Subsequently, PZ expressing adenoviral vector (AdV-PZ) or the
only GFP expressing control adenoviral vector (AdV-GFP; which was provided by
B. M. Pu
¨tzer; Institue for Experimental Gene Therapy and Cancer Research,
University Hospital Rostock) were applicated to PZ
2/2
mice by intravenous
injection (1*10
8
particles in saline) one day before and every seventh day after
preparation of hindlimb ischemia. Constant expression and secretion of PZ over
96 hours into blood was ensured by an ELISA for PZ (for further details of
adenovirus vector production, please see Material and Methods S1 and for in vitro
and in vivo kinetics of adenovirus see Figure S2 and S3). To address the possible
interaction of PZ with CXCR4, PZ
2/2
mice were exposed to AdV-PZ or AdV-
GFP in combination with three single injections of AMD3100 (5 mg/kg bw;
subcuteanous; AMD3100 octahydrocloride hydrate, Sigma Aldrich, St. Louis, MO,
USA) on post-operative day (POD) one, three and five.
Murine hindlimb ischemia model
Unilateral hindlimb ischemia was induced as previously described [21,22]. Mice
were anesthetized by an intraperitoneal injection of ketamine (75 mg/kg bw) and
xylazine (5 mg/kg bw) and placed on a warming pad to maintain the body
temperature at 37˚
C. Then, the right femoral artery (immediately distal to the
branch of the deep femoral artery) as well as the distal portion of the saphenous
artery were permanently ligated employing a 7-0 polypropylene suture (Prolene,
Ethicon, Norderstedt, Germany) and the ligated femoral artery was removed.
Wounds were carefully sutured using 6-0 sutures (Prolene).
Infrared thermal imaging (thermography)
Thermal imaging was performed as previously described [23]. In brief, before and
immediately after surgical ligation of the femoral artery and during follow-up on
post-operative day (POD) 21, mice were anesthetized as described above and
placed on a 37˚C heating pad for 6 min following 3 min on a table surface at
room temperature and infrared imaging was performed employing a Therma
CAM B20HS camera (FLIR Systems, Wilsonville, OR, USA). Images were
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 5/18

analyzed using FLIR QuickReport 1.2 software by determination of the
temperature at the middle of the pad of both the operated and non-operated
hindlimb.
Immunohistochemistry
Prior to harvest of M. gastrocnemius muscle tissue on POD 21, 100 mlof
fluorescein griffonia (bandeiraea) simplicifolia lectin I (Vector Laboratories,
Burlingame, CA, USA) were applied by left ventricular injection in anesthetized
mice for visualization of perfused tissues. Ten min later, mice were euthanized
and capillary density in the gastrocnemius muscle was assessed on 6 mm thick,
acetone-fixed frozen sections after staining with antibody against CD31 (1:50
dilution; Santa Cruz, USA) followed by Alexa555-labeled secondary antibody
(Molecular Probes, USA). Cell nuclei were counterstained with DAPI. The
number of CD31-immunopositive cells per muscle fiber was manually counted on
7 random microscope fields per section (2006magnification).
Statistical analysis
All data are given as median and interquartile range (IQR; the 25% and 75%
percentiles). Differences between more than two groups were calculated using
ANOVA on ranks, followed by the appropriate post-hoc comparison test and
between two groups using Mann-Whitney rank-sum test. Overall statistical
significance was set at p,0.05 and Bonferroni corrected for Mann-Whitney rank-
sum test. The statistical power was calculated for each significance at a50.05 with
a level of 80%. Statistics, power calculation and graphics were performed using the
software packages SigmaStat software version 3.5 and SigmaPlot software version
12.5 (Jandel Corporation, San Rafael, CA, USA).
Results
PZ promotes the angiogenic potency of endothelial cells in vitro
In the process of new blood vessel formation (angiogenesis), endothelial cell
migration and the formation of early capillaries are prerequisites [24]. We
investigated the effect of PZ on the migratory capacity of endothelial cells
employing a scratch-wound assay of confluent monolayers. In the presence of PZ
in a physiological concentration (3 mg/ml) endothelial cells exhibited an
significantly enhanced potency to close scratch wounds by endothelial migration
within 8 and 24 hours of incubation, compared with unstimulated cells (ctrl;
Figure 1 A–C). This promotion of endothelial cell migration by PZ was
comparable with that induced by 50 ng/ml SDF-1 (serving as positive control).
Furthermore, we investigated the capacity of PZ to modulate endothelial tube
formation on matrigel compared with SDF-1. After 8 hours of incubation,
endothelial cells had formed significantly increased lengths of tubular networks in
response to incubation with SDF-1 and PZ compared with untreated control (
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 6/18

Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 7/18

Figure 1 D and E). The effect on tube formation of PZ was comparable in
magnitude with that exerted by SDF-1.
PZ deficiency causes decreased ischemia-induced
neovascularization
Based on the pro-angiogenic properties of PZ in vitro, we evaluated the relevance
of these findings in vivo by using the murine model of hindlimb ischemia.
Thermography served for indirect assessment of hindlimb perfusion by calculating
temperature differences between the non-operated and operated extremity (
Figure 2 A). Quantification of temperature differences between the hindlimbs
revealed a significant increase in both PZ
+/+
and PZ
2/2
mice after preparation of
hindlimb ischemia compared to the pre-operation conditions (Figure 2 B). In
PZ
+/+
mice temperature difference between mouse pads decreased significantly on
POD 21, indicating enhanced reperfusion in the ischemic hindlimb. Of note,
temperature difference between hindlimbs of PZ
2/2
mice remained at a level
comparable to that found directly after the induction of ischemia (p,0.05 vs.
PZ
+/+
mice on POD 21). At POD 21 neovascularization was evaluated by
enumeration of CD31/DAPI double positive capillaries in the M. gastrocnemius.
As shown in Figure 2 D and E, induction of ischemia in PZ
+/+
mice resulted in an
increased capillary density, while PZ
2/2
mice did not exhibit increased angiogenic
activity in response to ischemia (p,0.05 vs. PZ
+/+
mice).
PZ expressing adenovirus reverses the diminished angiogenic
phenotype of PZ
2/2
mice
In order to analyse whether reduced angiogenesis in PZ
2/2
mice could indeed be
ascribed to the deficiency for PZ and is not related to other features of this genetic
phenotype, PZ
2/2
mice were administered a PZ expressing adenovirus (AdV-PZ)
resulting in physiological PZ-plasma concentrations with a median of 0.8 mg/ml at
a range of 0.5–1.0 mg/ml (25
th
and 75
th
percentiles) on POD 21. Quantification of
thermal images revealed a significant reduction in temperature difference on POD
21 in PZ
2/2
mice exposed to AdV-PZ, indicating improved blood perfusion in
Figure 1. PZ promotes the migration and capillary-like tube formation of endothelial cells in vitro.A,
Representative images of scratch-wound closures after 8 and 24 hours of incubation with SDF-1 or PZ, ctrl
corresponds to untreated cells. 100-fold magnification. Exposure of endothelial cells to PZ (3 mg/ml) led to a
significantly higher wound closure after 8 (B) and 24 hours (C) comparable to stimulation with SDF-1 (50 ng/
ml). Data are given as box plots indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks; $
p,0.05 vs. ctrl; n54–7 independent experiments. D, Representative images of tubular networks after 8 hours
of incubation in matrigel angiogenesis assay in the presence of SDF-1 (50 ng/ml), PZ (3 mg/ml), PZ+PTX and
PTX alone (100 ng/ml), ctrl are untreated cells. 100-fold magnification. E, HUVECs showed significantly
enhanced formation of capillary-like tubular structures on Matrigel when incubated with SDF-1 and PZ
compared with ctrl. Coincubation of endothelial cells with PZ and PTX abolished the PZ-mediated increase in
tube formation, while incubation with PTX alone had no effect on tube formation. Data are given as box plots
indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks; $ p,0.05 vs. ctrl; ß p,0.05 vs. PZ;
n56 independent experiments.
doi:10.1371/journal.pone.0113554.g001
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 8/18

Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 9/18

the ischemic hindlimb by reversing the phenotype of PZ
2/2
mice (Figure 2 C).
Analysis of the number of CD31/DAPI double positive capillaries in M.
gastrocnemius tissue revealed higher numbers in PZ
2/2
mice following
application of AdV-PZ compared to their non-ischemic hindlimb and also
compared to mice receiving AdV-GFP (serving as control virus) only, displaying
improved neovascularization by substitution of PZ
2/2
mice with PZ (Figure 2 D
and F).
Targeting PZ and CXCR4 with siRNA results in decreased
migratory capacity and endothelial tube formation of HUVECs in
vitro
To strengthen the findings of pro-angiogenic action of PZ in vitro knock-down of
PZ and CXCR4 was used to verify the contribution of PZ to basic angiogenic
processes. The migration assay was performed after knock-down of PZ and
CXCR4 by siRNA targeting (Figure 3 A and B). PZ knock-down in HUVECs
reduced the migratory capacity slightly below control values after 8 and 24 hours.
In addition, CXCR4 knock-down resulted in unstimulated as well as PZ and SDF-
1 exposed cells in migration capacity comparable to that of unstimulated cells at
both time-points (Figure 3 A and B). Performing matrigel angiogenesis assay with
PZ siRNA targeted endothelial cells revealed a significantly reduced capacity in
formation of tubular-like structures (Figure 3 C and D). Comparably, siCXCR4-
targeted cells did not show increased tube formation and did not respond to
stimulation with either PZ or SDF-1 (Figure 3 C and D).
Figure 2. PZ in ischemia-induced neovascularization in vivo.A, Representative pictures of thermal
imaging of mice hindlimbs. B, Quantitative summary of pad temperature differences. On POD 21 PZ
+/+
mice
showed a significant reduction in temperature difference compared to immediately after induction of ischemia
and significant lower temperature difference to PZ
2/2
mice, which offered no changes. Data are given as box
plots indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks repeated measures; * p,0.05
vs. pre OP, 1p,0.05 vs. post OP, #p,0.05 vs. other genotype on POD 21, n511–16. C, After 21 days, PZ
2/
2
mice treated with AdV-PZ showed significant reduction in temperature difference in the ischemic hindlimb.
Data are given as box plots indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks
repeated measures; * p,0.05 vs. pre OP, 1p,0.05 vs. post OP; n510. D, Representative pictures of M.
gastrocnemius muscle tissue after immunofluorescent staining for CD31 (red) or cell nuclei (DAPI, blue).
Intracardiac injection of an endothelial specific lectin confirmed functional perfusion of CD31/DAPI double
positive capillaries (green). E,PZ
+/+
mice exhibited significant higher density of CD31/DAPI-double positive
capillaries per square millimeter of ischemic M. gastrocnemius tissue compared to the non-ischemic PZ
+/+
muscle tissue. In muscle tissue of PZ
2/2
mice, ischemia did not result in remarkable neovascularization. The
number of CD31/DAPI double-positive cells in ischemic PZ
2/2
mice M. gastrocnemius was even significantly
lower compared to ischemic PZ
+/+
mice. Data are given as box plots indicating the median with the 25
th
and
75
th
percentiles. ANOVA on ranks; & p,0.05 vs. non-ischemic, #p,0.05 vs. other genotype with ischemia;
n511–16. F, After exposure of PZ
2/2
mice to AdV-PZ, the compromised angiogenic phenotype of PZ
2/2
mice
after induction of ischemia was significantly reversed. Data are given as box plots indicating the median with
the 25
th
and 75
th
percentiles. ANOVA on ranks; & p,0.05 vs. non-ischemic, #p,0.05 vs. AdV-GFP with
ischemia, n510.
doi:10.1371/journal.pone.0113554.g002
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 10 / 18

Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 11 / 1 8

PZ stimulates endothelial CXCR4 surface expression
In vitro effects of PZ on endothelial cells were comparable to those exerted by
SDF-1, which, together with its canonical receptor CXCR4, plays a central role in
mediating hypoxia-driven neovascularization, by inducing a G-protein-coupled
downstream signaling [25,26]. To further prove the hypothesis, whether CXCR4
could be required for mediating the pro-angiogenic effects of PZ, the functional
response of endothelial cells in the matrigel assay was investigated after G-protein
inhibition by incubation of endothelial cells with PZ in combination with 100 ng/
ml pertussis toxin (PTX). Combined incubation with PZ and PTX abolished the
pro-angiogenic effect of PZ significantly, compared to PZ only treated endothelial
cells (Figure 1 D and E), while adding PTX alone to endothelial cells had no effect
on endothelial tube formation. Stimulation of HUVECs with PZ for 8 and
24 hours resulted in an 1.4-fold and almost 2-fold increase of CXCR4 surface
expression vs. unstimulated ctrl (Figure 4 A–D), as detected by flow cytometric
analysis. Increased expression of CXCR4 is also displayed in representative
confocal laser scanning microscopic images of 8 and 24 hours PZ and SDF-1
stimulated HUVECs as well as unstimulated ctrl, as shown in Figure 4 E.
Application of AMD3100 to AdV-PZ exposed PZ
2/2
mice reverses
the beneficial effects of AdV-PZ
To evaluate if the surface expression of CXCR4 by PZ on stimulated HUVECs is
of relevance in vivo,PZ
2/2
mice were exposed to AdV-PZ or AdV-GFP and
received three single injections of AMD3100. On POD 21, neither AMD3100-
treated PZ
2/2
mice in the AdV-PZ nor in the AdV-GFP treatment group offered
remarkable improvement in reperfusion of the ischemic hindlimb (Figure 4 F and
G). Correspondingly, there was no increase in CD31/DAPI double positive cells in
M. gastrocnemius tissue in both groups above the level detected in non-ischemic
hindlimbs (Figure 4 H and I). In this regard, application of AMD3100 suspended
the beneficial pro-angiogenic effect of AdV-PZ in PZ
2/2
mice, indicating that the
restoration of the PZ
+/+
phenotype by AdV-PZ in PZ
2/2
mice is CXCR4-linked.
Figure 3. Targeting PZ and CXCR4 with siRNA results in decreased migratory capacity and endothelial
tube formation in vitro.PZ and CXCR4 knock-down in unstimulated HUVECs as well as CXCR4 knock-
down in PZ and SDF-1 stimulated HUVECs resulted in migration capacity comparable to that of ctrl after both
8(A) and 24 hours (B). Data are given as box plots indicating the median with the 25
th
and 75
th
percentiles. Ø
corresponds to unstimulated cells. ANOVA on ranks; $ p,0.05 vs. ctrl; n54 independent experiments.
Representative images (C) and quantitative analysis (D) of capillary-like tube formation. HUVECs targeted
with siRNA against PZ or CXCR4 showed significantly reduced capacity for capillary-like tube formation
compared to ctrl. Addition of PZ or SDF-1 to CXCR4 siRNA targeted cells failed to restore endothelial tube
formation. Data are given as box plots indicating the median with the 25
th
and 75
th
percentiles. Ø corresponds
to unstimulated cells. ANOVA on ranks; $ p,0.05 vs. ctrl; n54 independent experiments.
doi:10.1371/journal.pone.0113554.g003
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 12 / 18

Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 13 / 18

Discussion
Despite its characterization in human plasma in 1984 [2], the physiological
function of PZ is not yet completely understood. In line with the prothrombotic
phenotype of PZ deficiency observed in mice [7,8] the meta-analytical study of
Sofi et al. [27] demonstrated that low levels of PZ were associated with an
increased risk for arterial thrombosis, pregnancy complications and venous
thromboembolic diseases. Overall, patients with low levels of PZ showed nearly a
three-fold increased risk of thrombotic events [27]. Beyond the contribution of
PZ in coagulation, there are several clinical studies which show an association
between low PZ levels and the occurrence of coronary and peripheral
atherosclerosis [28,29,11,12] signifying a basic role of PZ in vascular biology,
including angiogenesis. There is a clear link between hypoxia-driven angiogenesis
and coagulation. By activating the hypoxia-inducible transcription factor HIF-1a,
thrombin triggers the expression of several angiogenic molecules and activated
platelets release large stores of angiogenic factors, such as vascular endothelial
growth factor (VEGF), platelet-derived growth factor (PDGF), transforming
growth factor-b(TGF-b), and interleukin-6 (IL-6) [30]. Thus, when coagulation
is initiated, a cascade of angiogenic signals is also generated. Together, coagulation
and angiogenesis could cooperate in stabilizing new blood vessels, repairing
injured vessels, and stimulating the sprouting of new vessels [31].
The purpose of the present study was to unravel the role of PZ in angiogenesis
by means of established in vitro and in vivo assays. Herein, we communicate the
following major findings: (i) In vitro, PZ stimulated migratory activity and
capillary-like tube formation of endothelial cells comparable to SDF-1, implying a
pro-angiogenic potential. (ii) Accordingly, PZ
2/2
mice exhibited diminished
hypoxia-driven neovascularis zation and reperfusion in post-ischemic hindlimbs,
(iii) which was restored by adenoviral gene transfer up to levels seen in PZ
+/+
mice. (iv) The increased surface expression of CXCR4 on PZ-stimulated
endothelial cells and the abrogated restoration of PZ
2/2
mice via AdV-PZ after
concomitant treatment with the CXCR4 antagonist AMD3100 supported the idea
Figure 4. PZ upregulates CXCR4 surface expression on endothelial cells in vitro and mediates its
angiogenic effects via CXCR4 in vivo.Flow cytometric analysis of CXCR4 surface expression on HUVECs
after 8 and 24 hours of incubation with PZ (red line), displayed in representative histograms (Aand C).
Quantitative analysis (Band D) of CXCR4 expression. Stimulation with PZ resulted in an 1.4-fold (B) and
almost 2-fold (D) increase of CXCR4 expression vs. unstimulated cells (ctrl, black line). Data are given as box
plots indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks; p,0.05; $ vs. ctrl; n54–8
independent experiments. ANOVA on ranks; p,0.05; $ vs. ctrl; n54–8 independent experiments. E,
Representative images of confocal laser scanning microscopy of HUVECs stimulated for 8 or 24 hours with
SDF-1 (50 ng/ml) or PZ (3 mg/ml), displaying an increased surface expression of CXCR4 after stimulation with
both substances. F, Representative pictures of thermal imaging of mice hindlimbs. G, Quantitative summary of
pad temperature differences pre OP, post OP and on POD 21. In both groups treated with AMD3100 no
significant change in temperature difference on POD 21 was detectable. Data are given as box plots
indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks repeated measures; * p,0.05 vs.
pre OP; n53–6. H, Representative pictures of M. gastrocnemius after immunofluorescent staining for CD31
(red) or cell nuclei (DAPI, blue). I, Quantitative summary of enumbered CD31/DAPI double positive cells
revealed no significant increase after induction of ischemia in both groups. Data are given as box plots
indicating the median with the 25
th
and 75
th
percentiles. ANOVA on ranks; n53–6.
doi:10.1371/journal.pone.0113554.g004
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 14 / 18

that PZ mediates angiogenesis via a G-protein coupled pathway and involves the
SDF-1/CXCR4 axis. (v) This is underlined by the fact that addition of the G-
protein inhibitor PTX to PZ-stimulated endothelial cells abolished the effect of PZ
on capillary-like tube formation. (vi) Using siRNA-targeting for knock-down of
PZ in in vitro assays resulted in reduced endothelial tube formation. Furthermore,
this addition of PZ to endothelial cells targeted with siCXCR4 failed to restore the
capacity for tube formation.
PZ increased the formation of capillary tube-like structures from endothelial
cells in vitro, comparably as this has been shown for aPC, a serine protease with
central role in physiological coagulation [32]. aPC mediates this effect by binding
to its endothelial protein C receptor (EPCR) and activating protease activated
receptor (PAR-1) [18]. With respect to the structural homology of PZ to aPC,
comparable mechanisms might underlie the action profile of PZ, though not
studied yet in full detail. However, comparably to aPC, which has been shown to
be synthesized in HUVECs [16], endothelial cells also synthesize PZ [20].
The reduced capacity of endothelial cells to migrate and form tubular structures
after knock-down of PZ in vitro is consistent with the reduced angiogenic
response in PZ
2/2
mice employing the hindlimb ischemia model. In line with
this, the reconstitution of physiological plasma concentration of PZ by adenoviral
gene transfer in PZ
2/2
mice, which restored the capability of mice to adequately
respond to hindlimb ischemia with neovascularization, underlines an involvement
of PZ in angiogenic processes.
Even though a possible down-stream signalling effect of PZ remains unknown,
the immunolocalization of PZ in the endothelium of arterial and venous vessel
sections implies the binding of PZ to a postulated, but so far not identified
endothelial receptor [20], which could be the crucial point in PZ-mediated effects.
Addition of PTX to PZ stimulated cells reverted the pro-angiogenic effect of PZ
significantly, which implies that PZ mediates its action via a G-protein coupled
pathway. Antagonizing CXCR4 by AMD3100 resulted in an insufficient
restoration of vascularization in AdV-PZ-pretreated PZ
2/2
mice. This observa-
tion and the flow cytometric data of the PZ-induced increase of CXCR4 surface
expression on endothelial cells allow to assume that PZ affects CXCR4 surface
expression via a G-protein coupled pathway. However, PZ only stimulates CXCR4
surface expression and seems not to be a ligand of CXCR4 like SDF-1, as PZ
stimulation did not activate the CXCR4-associated signal cascade, namely pERK
and pAKT (data not shown).
In conclusion, our results confirm the importance of PZ in vascular
homeostasis and extend the current knowledge by demonstrating a relevant
contribution of PZ in ischemia-induced angiogenesis. The mode of PZ action with
the underlying molecular mechanisms remains to be clarified. Nevertheless, PZ
might offer future potential as a novel approach for therapeutic angiogenesis.
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 15 / 18

Supporting Information
Figure S1. Genotyping of PZ mice. All animals were genotyped for presence or
absence of PZ by PCR using genomic DNA isolated from the tail tip (aqua dest.
served as negative control; M, marker).
doi:10.1371/journal.pone.0113554.s001 (TIF)
Figure S2. Kinetics of PZ expressing adenovirus in vitro.A, Representative
pictures of phase contrast microscopy of H1299 cells infected with PZ expressing
adenovirus (AdV-PZ), only GFP expressing adenovirus (AdV-GFP) or non
infected cells (Mock). 1006magnification B, Representative fluorescence
microscopy pictures of H1299 cells infected with PZ expressing adenovirus (AdV-
PZ), only GFP expressing adenovirus (AdV-GFP) or non-infected cells (Mock).
Original magnification 6100. C, Representative Western Blot of H1299 cells
exposed to AdV-PZ or AdV-GFP or non-infected cells (Mock) displaying a band
at 62 kDa only in cells infected with AdV-PZ. b-actin served as loading control. D,
PZ concentrations in the supernatant of H1299 cells measured by ELISA, data are
given in mean ¡SEM; n53 independent experiments; n.d., not detectable.
doi:10.1371/journal.pone.0113554.s002 (TIF)
Figure S3. Kinetic of PZ expressing adenovirus in vivo.A, Representative
intravital fluorescence microscopy images of liver displaying GFP-fluorescent
hepatocytes of PZ
2/2
mice exposed to AdV-PZ or AdV-GFP over a period of
96 hours. Original magnification 650. B, Representative immuno-histochemical
images of hepatic tissue stained for GFP in PZ
2/2
mice exposed to AdV-PZ or
AdV-GFP. Original magnification 6100. C, Representative immunohistochem-
ical images of hepatic tissue stained for PZ in PZ
2/2
mice exposed to AdV-PZ or
AdV-GFP. Original magnification 6100. D, PZ plasma concentrations measured
by ELISA in PZ
2/2
mice exposed to AdV-PZ or AdV-GFP; Data are given in
mean ¡SEM; n53.
doi:10.1371/journal.pone.0113554.s003 (TIF)
Material and Methods S1. Further information about genotyping of PZ mice,
adenovirus vector production and adenovirus kinetic studies can be found in
the supporting information.
doi:10.1371/journal.pone.0113554.s004 (DOCX)
Acknowledgments
We thank Prof. B. M. Pu
¨tzer and A. Stoll (Institute for Experimental Gene
Therapy and Cancer Research, University of Rostock) for adenovirus production
and technical assistance. We also thank B. Blendow, D. Frenz, M. Nerowski and E.
Lorbeer-Rehfeldt (Institute for Experimental Surgery, University of Rostock) for
excellent technical assistance.
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 16 / 18

Author Contributions
Conceived and designed the experiments: NW BV. Performed the experiments:
AB BG. Analyzed the data: AB NW BG BV. Contributed reagents/materials/
analysis tools: AB NW BG BV. Wrote the paper: AB NW BG BV. Performed the
histological staining of kinetic analysis of the adenovirus: BG.
References
1. Fujimaki K, Yamazaki T, Taniwaki M, Ichinose A (1998) The gene for human protein Z is localized to
chromosome 13 at band q34 and is coded by eight regular exons and one alternative exon. Biochemistry
37: 6838–6846.
2. Broze GJ Jr, Miletich J (1984) Human Protein Z. J Clin Invest 73: 933–938.
3. Prowse CV, Esnouf MP (1977) The isolation of a new warfarin-sensitive protein from bovine plasma.
Biochem Soc Trans 5: 255–256.
4. Broze GJ Jr. Protein Z-dependent regulation of coagulation (2001) Thromb Haemost 86: 8–13.
5. Han X, Fiehler R, Broze GJ Jr (1998) Isolation of a protein Z-dependent plasma protease inhibitor. Proc
Natl Acad Sci U S A 95: 9250–9255.
6. Han X, Huang ZF, Fiehler R, Broze GJ Jr (1999) The protein Z-dependent protease inhibitor is a serpin.
Biochemistry 38: 11073–11078.
7. Yin ZF, Huang ZF, Cui J, Fiehler R, Lasky N, et al. (2000) Prothrombotic phenotype of protein Z
deficiency. Proc Natl Acad Sci U S A 97: 6734–6738.
8. Zhang J, Tu Y, Lu L, Lasky N, Broze GJ Jr (2008) Protein Z-dependent protease inhibitor deficiency
produces a more severe murine phenotype than protein Z deficiency. Blood 111: 4973–4978.
9. Santacroce R, Sarno M, Cappucci F, Sessa F, Colaizzo D, et al. (2006) Low protein Z levels and risk
of occurrence of deep vein thrombosis. J Thromb Haemost 4: 2417–2422.
10. Vasse M, Guegan-Massardier E, Borg JY, Woimant F, Soria C (2001) Frequency of protein Z
deficiency in patients with ischaemic stroke. Lancet 357: 933–934.
11. Sofi F, Cesari F, Pratesi G, Cellai AP, Pulli R, et al. (2007) Low protein Z levels in patients with
peripheral arterial disease. Thromb Haemost 98: 1114–1117.
12. Sofi F, Cesari F, Tu Y, Pratesi G, Pulli R, et al. (2009) Protein Z-dependent protease inhibitor and
protein Z in peripheral arterial disease patients. J Thromb Haemost 7: 731–735.
13. Aihara K, Azuma H, Takamori N, Kanagawa Y, Akaike M, et al. (2004) Heparin cofactor II is a novel
protective factor against carotid atherosclerosis in elderly individuals. Circulation 109: 2761–2765.
14. Ikeda Y, Aihara K, Yoshida S, Iwase T, Tajima S, et al. (2012) Heparin cofactor II, a serine protease
inhibitor, promotes angiogenesis via activation of the AMP-activated protein kinase-endothelial nitric-
oxide synthase signaling pathway. J Biol Chem 287: 34256–34263.
15. Whitmont K, Fulcher G, Reid I, Xue M, McKelvey K, et al. (2013) Low circulating protein C levels are
associated with lower leg ulcers in patients with diabetes. Biomed Res Int. 2013: 719570.
16. Tanabe S, Sugo T, Matsuda MJ (1991) Synthesis of protein C in human umbilical vein endothelial cells.
J Biochem 109: 924–928.
17. Uchiba M, Okajima K, Oike Y, Ito Y, Fukudome K, et al. (2004) Activated protein C induces endothelial
cell proliferation by mitogen-activated protein kinase activation in vitro and angiogenesis in vivo. Circ
Res. 95: 34–41.
18. Minhas N, Xue M, Fukudome K, Jackson CJ (2010) Activated protein C utilizes the angiopoietin/Tie2
axis to promote endothelial barrier function. FASEB J 24: 873–881
19. Xue M, Chow SO, Dervish S, Chan YK, Julovi SM, et al. (2011) Activated protein C enhances human
keratinocyte barrier integrity via sequential activation of epidermal growth factor receptor and Tie2. J Biol
Chem. 286: 6742–6750.
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 17 / 18

20. Vasse M, Denoyelle C, Corbie`re C, Litzler PY, Legrand E, et al. (2006) Human endothelial cells
synthesize protein Z, but not the protein Z dependent inhibitor. Thromb Haemost 95: 519–523.
21. Heida NM, Leifheit-Nestler M, Schroeter MR, Mu¨ller JP, Cheng IF, et al. (2010) Leptin enhances the
potency of circulating angiogenic cells via src kinase and integrin (alpha)vbeta5: implications for
angiogenesis in human obesity. Arterioscler Thromb Vasc Biol. 30: 200–206.
22. Limbourg A, Korff T, Napp LC, Schaper W, Drexler H, et al. (2009) Evaluation of postnatal
arteriogenesis and angiogenesis in a mouse model of hind-limb ischemia. Nat Protoc.;4: 1737–46.
23. Wagner NM, Bierhansl L, No¨ldge-Schomburg G, Vollmar B, Roesner JP (2013) Toll-like receptor 2-
blocking antibodies promote angiogenesis and induce ERK1/2 and AKT signaling via CXCR4 in
endothelial cells. Arterioscler Thromb Vasc Biol. 33: 1943–51.
24. Carmeliet P (2000) Mechanisms of angiogenesis and arteriogenesis. Nat Med. 6: 389–395.
25. Ceradini DJ, Kulkarni AR, Callaghan MJ, Tepper OM, Bastidas N, et al. (2004) Progenitor cell
trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med. 108: 858–864.
26. Ho TK, Tsui J, Xu S, Leoni P, Abraham DJ, et al. (2010) Angiogenic effects of stromal cell-derived
factor-1 (SDF-1/CXCL12) variants in vitro and the in vivo expressions of CXCL12 variants and CXCR4 in
human critical leg ischemia. J Vasc Surg. 51: 689–699.
27. Sofi F, Cesari F, Abbate R, Gensini GF, Broze G Jr, et al. (2010) A meta-analysis of potential risks of
low levels of protein Z for diseases related to vascular thrombosis. Thromb Haemost. 103: 749–756.
28. Fedi S, Sofi F, Brogi D, Tellini I, Cesari F, et al. (2003) Low protein Z plasma levels are independently
associated with acute coronary syndromes. Thromb Haemost. 90: 1173–1178.
29. Sofi F, Cesari F, Vigiani S, Fatini C, Marcucci R, et al. (2005) Protein Z plasma levels in different
phases of activity of coronary atherosclerosis. J Thromb Haemost. 3: 2254–2258.
30. Carmeliet P (2003) Angiogenesis in health and disease. Nat Med. 9: 653–60.
31. Carmeliet P (2001) Biomedicine. Clotting factors build blood vessels. Science. 293: 1602–1604.
32. Gramling MW, Beaulieu LM, Church FC (2010) Activated protein C enhances cell motility of endothelial
cells and MDA-MB-231 breast cancer cells by intracellular signal transduction. Exp Cell Res. 316: 314–
328.
Protein Z in Angiogenesis
PLOS ONE | DOI:10.1371/journal.pone.0113554 December 4, 2014 18 / 18