Regulation of endothelial migration and proliferation by ephrin-A1

    Regulation of endothelial migration and proliferation by ephrin-A1 Elisa Wiedemann, Stefanie Jellinghaus, Georg Ende, Antje Augstein, Ronny Sczech, Ben Wielockx, S¨onke Weinert, Ruth H. Strasser, David M. Poitz PII: DOI: Reference:

S0898-6568(16)30247-9 doi:10.1016/j.cellsig.2016.10.003 CLS 8774

To appear in:

Cellular Signalling

Received date: Revised date: Accepted date:

25 July 2016 9 October 2016 10 October 2016

Please cite this article as: Elisa Wiedemann, Stefanie Jellinghaus, Georg Ende, Antje Augstein, Ronny Sczech, Ben Wielockx, S¨onke Weinert, Ruth H. Strasser, David M. Poitz, Regulation of endothelial migration and proliferation by ephrin-A1, Cellular Signalling (2016), doi:10.1016/j.cellsig.2016.10.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Regulation of endothelial migration and proliferation by ephrin-A1 Elisa Wiedemann1,a, Stefanie Jellinghaus1,a, Georg Ende1,a, Antje

IP

Strasser1, David M. Poitz1

T

Augstein1, Ronny Sczech2, Ben Wielockx3, Sönke Weinert4, Ruth H.

1

SC R

Short title: Ephrin-A1 regulates endothelial migration

Department of Internal Medicine and Cardiology, TU Dresden,

NU

Germany

Lightmicroscopy Facility, BioTec Dresden, Germany

3

Department of Clinical Pathobiochemistry, TU Dresden, Germany

4

Department of Internal Medicine, Division of Cardiology, Angiology and

MA

2

These authors contributed equally to this work

CE P

a

TE

D

Pneumology, Magdeburg University, Germany

Conflict of Interest Disclosures: none

AC

Keywords: Endothelial cells, migration, proliferation, In-stent-restenosis

Total Word count: 6062 ; Abstract word count: 331 ; Number of figures: 9 Correspondence:

Dr. David M. Poitz

TU Dresden, Department of Internal Medicine and Cardiology Fetscherstr. 74, 01307 Dresden Phone:

+49-351-458-6627

Fax:

+49-351-458-6329

eMail:

[email protected] 1

ACCEPTED MANUSCRIPT Abstract Endothelial migration and proliferation are fundamental processes in

T

angiogenesis and wound healing of injured or inflamed vessels. The

IP

present study aimed to investigate the regulation of the Eph/ephrin-

SC R

system during endothelial proliferation and the impact of the ligand ephrin-A1 on proliferation and migration of human umbilical venous (HUVEC) and arterial endothelial cells (HUAEC).

NU

Endothelial cells that underwent contact inhibition showed a massive induction of ephrin-A1. In contrast, an injury to a confluent endothelial

MA

layer, associated with induction of migration and proliferation, showed reduced ephrin-A1 levels. In addition, reducing ephrin-A1 expression by siRNA led to increased proliferation, whereas the overexpression of

D

ephrin-A1 led to decreased proliferative activity. Due to the fact that

TE

wound healing is a combination of proliferation and migration, migration

CE P

was investigated in detail. First, classical wound-healing assays showed increased wound closure in both ephrin-A1 silenced and overexpressing cells. Live-cell imaging enlightened the underlying differences. Silencing

AC

of ephrin-A1 led to a faster but more disorientated migration. In contrast, ephrin-A1 overexpression did not influence velocity of the cells, but the migration was more directed in comparison to the controls. Additional analysis of EphA2-silenced cells showed similar results in terms of proliferation and migration compared to ephrin-A1 silenced cells. Detailed analysis of EphA2 phosphorylation on ligand-dependent phospho-site (Y588) and autonomous activation site (S897) revealed a distinct phosphorylation pattern. Furthermore, the endothelial cells ceased to migrate when they came in contact with an ephrin-A1 coated surface. Using a baculoviral-mediated expression system, ephrin-A1 silencing and overexpression was shown to modulate the formation of focal adhesions. This implicates that ephrin-A1 is involved in changes of 2

ACCEPTED MANUSCRIPT the actin cytoskeleton which explains the alterations in migratory actions, at least in part. In conclusion, ephrin-A1 expression is regulated by cellular density and

T

is itself a critical determinant of endothelial proliferation. According to

IP

current knowledge, ephrin-A1 seems to be remarkably involved in

SC R

elementary processes of endothelial migration like cellular polarization, migratory direction and speed. These data support the notion that ephrin-

AC

CE P

TE

D

MA

NU

A1 plays a pivotal role in basal mechanisms of re-endothelialization.

3

ACCEPTED MANUSCRIPT 1 Introduction The migration and proliferation of endothelial cells represent fundamental

T

events not only during angiogenesis, tumor growth and wound healing

IP

but also in endothelial regeneration after stent-implantation or vascular

SC R

injury [1,2]. Endothelial cell migration in the organism is a complex multistep process which involves attractive and repulsive signals from the surrounding cells and tissue. The migratory movement of the cells

NU

requires stimulation and polarization of the cells, protrusion and adhesion formation and finally forward movement. Therefore, the migratory cells

MA

need direction and limitation. The regulation and interaction of these processes are of great importance and involve a variety of signaling cascades like Src family kinases and Ras/Rho family GTPases [3-5].

D

Proliferation of endothelial cells is required for the growth of new or

CE P

tumor progression.

TE

existing blood vessels in angiogenesis or pathologic processes like

The endothelium lines the interior surface of the vessel lumen and separates the intravasal compartment from the rest of the vessel wall. A

AC

dysfunction of the endothelium disturbs the border between the vessel lumen and the vessel wall, which enables the adhesion and transmigration of blood components into deeper parts of the vessel. This event

is

of

great

importance

for

pathological

processes

like

atherosclerosis [6]. The denudation of the endothelium after stentimplantation in arteries can result in neointima formation by proliferation of smooth-muscle-cells (SMC), release of extracellular matrix and accumulation of inflammatory cells [7]. Late stent-thrombosis, in particular, is caused by a delayed re-endothelialization, a disturbed endothelial function as well as an inhibition of vascular repair mechanisms [8,9]. 4

ACCEPTED MANUSCRIPT The Eph/ephrin-system represents the largest family of receptor tyrosine kinases. The ligands and their receptors are classified into classes A and B depending on their receptor binding affinity and structural similarities.

T

This family consists of 9 EphA and 5 EphB receptors and 5 ephrin-A-

IP

and 3 ephrin-B-ligands. A hallmark of the Eph/ephrin-system is its ability

SC R

to signal bi-directionally, allowing forward-signaling in the receptorexpressing cell and reverse signaling in the ligand-expressing cell to occur. The Eph/ephrin-system influences short-distance cell-to-cell-

NU

communications, which is important for a wide variety of cellular functions in physiology and disease. Among neighbouring cells, the Eph

MA

receptors and ephrin-ligands influence cell proliferation and migration, cell-cell repulsion and adhesion as well as boundary formation [10-12].

D

The Eph/ephrin-system conducts alterations in cell protrusion, adhesion,

TE

segregation and migration by regulating cytoskeletal dynamics [13]. For example, the ligand ephrin-B2 and the receptor EphB4 are fundamentally

CE P

required for angiogenesis and border formation between the arterial and venous vessel system [14-16]. Recently, it was shown that Eph receptors and ephrin-ligands on the surface of exosomes mediate cell-contact

AC

independent Eph/ephrin-signaling in the neuronal system [17], which opens new perspectives on the complex properties of the Eph/ephrinsystem. The complex interplay between Eph receptors and ephrin-ligands on endothelial cell integrity and dysfunction is still not sufficiently understood. We know that different members of the Eph/ephrin-system are regulated by hypoxia and other proangiogenic factors which are important in neovascularization of tumors and inflammatory processes [18]. Previous results from our group showed the importance of ephrinA1-EphA4 forward-signaling in the process of endothelial activation regarding monocyte adhesion [19,20]. It is remarkable that the ephrin-A1 5

ACCEPTED MANUSCRIPT mediated activation of endothelial EphA4 modulates the rearrangement of the actin-cytoskeleton, which is known to be important in the process of migration. The interacting ligand ephrin-A1 and the receptor EphA2

T

are both described to be involved in migratory activity of different cell

IP

types [21-23]. The regulation of this migratory activity was shown to be

SC R

linked to a ligand independent activation of the EphA2 receptor on [24,25].

NU

Based on the knowledge about fundamental functions of Eph/ephrins in the guidance and organization of cell assemblies, proliferation control

MA

and the definition of tissue boundaries, it is presumed that Eph/ephrins are also involved in the process of vascular regeneration. The aim of the current study was to investigate a potential influence of the Eph/ephrin-

D

system on endothelial proliferation and migration as basal mechanisms

CE P

TE

of vascular regeneration or re-endothelialization.

2.1

AC

2 Methods

Cell isolation, culture and siRNA transfection

Venous (HUVEC) and arterial endothelial cells (HUAEC) were isolated from anonymously acquired umbilical cords. Isolation and culturing of the cells was performed as previously described [19,20,26]. SiRNA transfection of HUVEC was done using Turbofect transfection reagent (Thermo Scientific, Waltham, MA, USA) according to the protocol described previously [26]. SiRNAs used in this study are summarized in Table 1 and were purchased from Eurofins MWG Operon (Ebersberg, Germany). 6

ACCEPTED MANUSCRIPT Table 1 Sequences of siRNAs Target

siRNA sequence (5’ to 3’)

siRNA

T

denotation

d (siScr)

si472 ephrin-

antisens

CAAGGCGAUUACACUACCUT

e

T

sense

GGACACAGCUACUACUACATT

antisens

UGUAGUAGUAGCUGUGUCCT

e

T

sense

A1

AC

EphA2

si2181

2.2

T UUCAGCUGCACAUGUAUGGT

e

T

sense

UGACAUGCCGAUCUACAUGT

TE CE P si177

CCAUACAUGUGCAGCUGAAT

antisens

D

si218

SC R

e control

T

NU

siScramble

MA

nonsens

AGGUAGUGUAAUCGCCUUGT

IP

sense

T antisens

CAUGUAGAUCGGCAUGUCAT

e

T

sense

GUACCUGGCCAACAUGAACT T

antisens

GUUCAUGUUGGCCAGGUACT

e

T

Real-time RT-PCR

RNA isolation was conducted using the NucleoSpin-RNA Kit (Machery-Nagel, Düren, Germany). cDNA was synthesized using the Revert AidTMH Minus-cDNA Synthesis Kit (Thermo Scientifc) and 1 to 5 µg total RNA and oligo-dT primers. Quantitative PCR was 7

ACCEPTED MANUSCRIPT performed on CFX96 cycler (BioRad, Hercules, CA, USA) using the MAXIMA SYBR Green qPCR Master Mix in a total volume of 10 µl. Temperature protocol was as follows: initial denaturation at 95°C for

T

8 min followed by 45 amplification cycles, each consisting of 95°C for

IP

20 s, 58°C for 45 s and 72°C for 20 s with a final extension step at

SC R

72°C for 2 min and a subsequent melting point analysis. Primer sets used are summarized in table 2. Gene expression was normalized to HPRT1 after testing different house-keeping genes (HPRT1, HRBP,

NU

EF2, TBP) and confirming the suitability of this gene as house-

MA

keeping gene according to the algorithm of Hellemans [27].

GenBank

Primer

No.

orientation

TE

Gene

D

Table 2 Sequences of primers used for real-time PCR

ephrin-

ephrinA4

NM_004428

AC

A1

CE P

sense

NM_005227

antisense

Sequence (5’ to 3’)

Fragment size

TTG GGT CTG TGC TGC AGT C GAT GTC CAC GTA

123 bp

GTC ATT CAG C sense

CTT CAC ACC CTT CTC CCT CG

antisense

AGA CAC CTG GAG

109 bp

CCT CAA GC sense ephrinA5

NM_001962

ACA TCT CCT CTG CAA TCC CAG

antisense

ATG TAC GGT GTC

167 bp

ATC TGC TGG ephrinB1

sense NM_004429

TCC TGG TTC ACA GTC TCA TGC

antisense

161 bp

ATG AAG GTT GGG 8

ACCEPTED MANUSCRIPT CAA GAT CC sense

B2

NM_004093

GCCATG AAG antisense

GTT GCC GTC TGT

177 bp

T

ephrin-

TGC CAG ACA AGA

IP

GCT AGA ACC

EphA2 NM_004431

GCG ACG TGT GGA

SC R

sense

GCT TTG G antisense

TGG CTT TCA TCA

96 bp

sense

MA

antisense

TE

sense

AC

CE P

EphB1 NM_004441

EphB2 NM_004442

TTT GTC ATC AGC CGG AGA CG

D

EphA4 NM_004438

NU

CCT CGT GG

antisense

CTC TCG CAC TGC

135 bp

TTG GTT GG CAT GGA GAA TGG TGC ATT GG CAG AAT GTT CCT

157 bp

AGC AGC CAG sense

CCG AGC CAC GCA TTT ACA TC

antisense

ATG ATG GAC ACT

160 bp

GCC GAT GG sense EphB4 NM_004444

GCC TCA GGA AGC AGA GCA ATG

antisense

AAC TCA CCT GCA

193 bp

CCA ATC ACC sense HPRT1

NM_000194

TTG CGA CCT TGA CCA TCT TTG

antisense

CTT TGC TGA CCT

96 bp

GCT GGA TTA C 9

Western blotting

SC R

2.3

IP

T

ACCEPTED MANUSCRIPT

Western blot analysis was performed as previously described [28].

NU

The following antibodies were used for this study: polyclonal rabbit anti-ephrin-A1 (Santa Cruz) 1:1000, monoclonal rabbit anti-EphA2

MA

(Cell Signaling) 1:1000, monoclonal rabbit anti-EphA2 (Tyr588) (Cell Signaling) 1:1000, monoclonal rabbit anti-EphA2 (Ser897) (Cell Signaling) 1:1000, monoclonal mouse anti--actin (Santa Cruz)

and

goat

anti-rabbit-HRP

TE

1:10000

D

1:1000, sheep anti-mouse-HRP (Amersham, Piscataway, USA) (Santa

Cruz)

1:2500.

CE P

Quantification of the western blots was done by densitometry using Quantity One (BioRad). BrdU-incorporation ELISA

AC

2.4

Determination of cell proliferation was done using the Cell Proliferation ELISA BrdU (colorimetric) (Roche, Germany). HUVEC were

seeded

in

collagen-coated

96-well

cell

culture

plates

(5x103 cells/well) and cultured for 24 h. Cells were labelled with 10 µM BrdU for 8 h. Quantification of BrdU incorporation was performed according to the manufacture’s protocol. 2.5

Wound-healing assay

To perform wound-healing assays, 2-well culture inserts in µDish35mm high (ibidi, Germany) were used. 2.4x104 HUVEC were seeded into each of the cavities and allowed to adhere overnight. 10

ACCEPTED MANUSCRIPT After removal of the silicon spacer, cells were washed two times with medium to remove cell debris. Afterwards, pictures of the complete wound were acquired every 2 h over a period of 12 h. Quantification

T

of cell-free area was done using ImageJ (EMBL, Germany). In the

IP

case of live-cell-imaging, HEPES-buffered (20 mM) medium was

SC R

used and pictures of a defined area were taken every 5 min. The tracking of cells (10-20 cells per condition and independent experiment) was done manually using ImageJ. The criteria for

NU

tracking the cells consisted of a lack of cell division and complete tracking paths during the whole observation period. Analysis of

MA

migration was done using the Chemotaxis and migration Tool (ibidi, Germany). Analyzed parameters were directness, velocity and the

D

Rayleigh test. Directness is defined as the mean of the ratio of

TE

Euclidian distance (length of the line between start and end point) and accumulated distance (length of the migration path). This

CE P

parameter represents an indicator for how direct the cells migrate to the end point. The Rayleigh test was used to evaluate whether the end points of the cells were distributed randomly or focused in a

2.6

AC

region. Rose plots were generated using R statistical package. Migration assay

To study the effect of migration to ephrin-A1 coated surfaces, 4-well culture inserts in µ-Dishes35mm

high

(ibidi, Germany) were used. One

well was coated with a mixture of fibronectin (0.125 µg/cm2) and IgGFc (1 µg/cm2, control) and the other well was coated with a mixture of fibronectin (0.125 µg/cm2) and ephrin-A1-Fc (1 µg/cm2) overnight. After coating, HUVEC were seeded in the non-coated cavities (7x103 cells/well) and allowed to adhere overnight. The silicon insert was removed and medium was changed. The cells were allowed to grow 11

ACCEPTED MANUSCRIPT over a period of 6-9 days until they reached the borders of the coated wells. Photomicrographs were taken to determine the overgrown area

Baculoviral infection

IP

2.7

T

of the coated wells. Area analysis was performed using ImageJ.

SC R

To get a depiction of the actin-cytoskeleton and focal adhesion in vital endothelial cells CellLight® Talin-RFP, BacMam 2.0 and CellLight® Actin-GFP, BacMam 2.0 (life technologies) were used in combination

NU

with the BacMam Enhancer Kit (life technologies) according to the manufacturer’s specification. HUVEC were infected with both viruses

Data analysis and statistics

D

2.8

MA

simultaneously. The infection dose was 50 particles per cell (PPC).

TE

Data are presented as mean ± standard error. The number of independent experiments with endothelial cells isolated from different

CE P

umbilical cords is given in the figure legends (n-values). Statistical analysis was done using analysis of variance (ANOVA) or paired Student’s t-test. In the case of ANOVA, significant differences

AC

between the groups were identified by Tukey-PostHoc test. P-values below 0.05 were defined as statistically different and the levels of significance were defined as follows: * p<0.05, ** p<0.01, *** p<0.001.

Results 3.1 Cell density is associated with ephrin-A1 expression levels The effect of cell density and proliferation on the expression of different ephrin-ligands and Eph-receptors was investigated. Human venous endothelial cells were seeded in three different cell densities (5x103, 12

ACCEPTED MANUSCRIPT 10x103, 20x103 cells/cm2) and were cultivated for 24, 48 and 72h (Fig. 1A). The mRNA (Fig. 1B and C) and protein expression (Fig. 1D and F) were analyzed. In the first attempt, different ephrin-ligands (ephrin-A1, -

T

A4, -A5, -B1 and –B2) and Eph-receptors (EphA2, -A4, B1, -B2, -B4),

IP

known to be expressed in endothelial cells [29,30], were analyzed on

SC R

mRNA-level by normalization to the house-keeping gene HPRT1. The most pronounced effect was observed in the case of ephrin-A1 which strongly correlates with the cell density (Fig. 1B and 1C). The expression

NU

of the ligand ephrin-B2 showed milder correlation with cell density compared to ephrin-A1 (Fig. 1B). This correlation of ephrin-A1

MA

expression and cell density was also verified at protein level (Fig. 1D). To investigate the consequence of longer cultivation periods, which leads to

D

contact inhibition, cells with an initial cell density of 10000 and 20000

TE

cells/cm2 were cultivated for up to 11 days and mRNA expression of ephrin-A1 was analyzed every 2nd day. After 5 days of cultivation, both

CE P

cultures showed comparable expression of ephrin-A1, which remained unchanged up to 11 days of cultivation (Fig. 1E). The same setting was applied to arterial endothelial cells from the umbilical cord in order to

AC

investigate whether this effect was also apparent in different endothelial cell types (Fig. 2). Comparable to the results in HUVEC, arterial endothelial cells also showed an increased expression of ephrin-A1 in correlation to the cell density on mRNA (Fig. 2A and B) and protein level (Fig. 2C).

The protein expression of the EphA2 receptor was also

monitored in HUVEC growing from different starting cell densities. After 48 h and 72 h, a clear expression pattern was observed: Inversely to the expression of the ligand ephrin-A1, its receptor EphA2 showed highest expression at low cell densities (Fig. 1F). 3.2 Growth arrest and induction of proliferation and migration modulate ephrin-A1 expression 13

ACCEPTED MANUSCRIPT The fact that cell density correlates with the expression of ephrin-A1 raises the question to what extent growth arrest or induction influences ephrin-A1 expression levels. In addition to the results in Fig. 1 and 2, in

T

which growth arrest was mediated by contact inhibition, the effect of

IP

serum-depletion on ephrin-A1 expression was also investigated. As

SC R

outlined in Fig. 3A, culturing endothelial cells with different amounts of serum inversely correlated with the expression of ephrin-A1. The most pronounced effect was seen in completely serum-starved cells (Fig. 3A

NU

top chart). Comparable results were obtained at the protein level (Fig. 3A bottom chart). In contrast to growth inhibition, ephrin-A1 expression was

MA

also monitored in cells, which underwent growth induction by applying different numbers of wounds. Four and eight scratches were applied on a

D

confluent layer of endothelial cells. Ephrin-A1 mRNA (Fig. 3B top chart)

TE

and protein expression (Fig. 3B bottom chart) were measured 24h after the scratch was applied. As shown in Figure 3B, the expression of

CE P

ephrin-A1 diminished in correlation to increasing numbers of wounds after normalization to the housekeeping genes. To elucidate whether modulation of ephrin-A1 levels in injured endothelial cells also influences

AC

the phosphorylation state of the EphA2 receptor, phosphorylation of Y588 (indicative for ligand-dependent activation) and S897 (indicative for autonomous activation) was evaluated 2 h after wounding. This analysis revealed a decreased phosphorylation of the Y588 site, which is known to be activated by binding of ephrin-A1 to EphA2. The phosphorylation site S897, which is described to be activated in migrating cells [24,25], showed increased phosphorylation. The total amount of EphA2 expression was not altered by wounding (Fig. 3C). The phosphorylation state of the EphA2 receptor was also analyzed 24 h after injury. This analysis showed a decreased phosphorylation at Tyr588 by trend and a significantly increased phosphorylation of Ser897 (Fig. 3D). 14

ACCEPTED MANUSCRIPT 3.3 Ephrin-A1 and EphA2 regulate endothelial proliferation Expression of ephrin-A1 correlated with the endothelial cell density and

T

reached a plateau in contact-inhibited HUVEC as well as in serum-

IP

depleted cells. Growth induction in an injured endothelial cell monolayer

SC R

also correlated with decreased ephrin-A1 levels and a modulation of EphA2 phosphorylation. Therefore, the hypothesis was tested whether or not

modulation

of

ephrin-A1

and

EphA2

expression

regulates

NU

proliferation as part of the endothelial wound healing process. This aspect was investigated using a silencing and adenoviral overexpression

MA

approach. Silencing of ephrin-A1 and EphA2 and overexpression of ephrin-A1 are illustrated by the Western Blots (Fig. 4 A-C, upper part). To determine the proliferative activity of HUVEC, BrdU incorporation was

D

measured after 8 h. As shown in Figure 4, silencing of ephrin-A1 (si218

TE

and si472) in endothelial cells increased proliferation compared to cells

CE P

transfected with a nonsense siRNA (siCo) (Fig. 4A). Furthermore, the effect of EphA2-receptor silencing (si177 and si2181) was analyzed and showed that this also led to increased proliferation of endothelial cells

AC

suggesting the importance of ephrin-A1 EphA2 interaction in this process (Fig. 4B). Reversely, ephrin-A1 was overexpressed adenovirally in HUVEC. An adenovirus encoding for the -galactosidase gene was used as

control.

In

ephrin-A1

overexpressing

endothelial

cells,

the

proliferation, measured as BrdU-incorporation after 8 h, was reduced by 18% (Fig. 4C). Similar to the results in Fig. 1 and 2, these experiments demonstrate that ephrin-A1 and its receptor EphA2 are crucial determinants for endothelial cell proliferation.

15

ACCEPTED MANUSCRIPT 3.4 Ephrin-A1 is an important regulator of endothelial migration Aside from the influence of ephrin-A1 and EphA2 on endothelial

T

proliferation, the impact of ephrin-A1 and its receptor EphA2 on

IP

endothelial migration was also analyzed by different approaches. First,

SC R

the migration capacity of endothelial cells was tested in a classical wound-healing assay. The speed of wound closure was measured in ephrin-A1 (si472) or EphA2 (si2181) silenced cells compared to control

NU

cells (siCo) (Fig. 5A and B). These experiments revealed a faster wound closure in ephrin-A1 and EphA2 silenced cells. To test the hypothesis on

MA

whether or not ephrin-A1 is a limiting factor for endothelial migration, the wound closure in ephrin-A1 overexpressing cells was also determined. Interestingly, overexpression of ephrin-A1 led to a faster closure of the

D

wound (Fig. 5C). To gain a deeper insight into the mechanism that led to

TE

these results, a live-cell imaging approach, which allows the tracking of

CE P

single cells, was used. Exemplary videos of the live cell imaging of ephrin-A1 silenced or overexpressing cells can be found in the supplemental data (Supplemental video 1). In Fig. 6 the trajectory plots

AC

of the single cells under silencing of ephrin-A1 (Fig. 6A) or EphA2 (Fig. 6B) and overexpression of ephrinA1 (Fig. 6C) are shown. This analysis clearly revealed that the movement of ephrin-A1 silenced cells (si218 and si472) is much more uncoordinated in comparison to the siCotransfected cells. Similar results were obtained under EphA2-silencing (si177

and

si2181).

In

contrast,

overexpression

of

ephrin-A1

(AdXEFNA1) led to a straight forward migration of the endothelial cells compared to the AdXlacZ-transduced cells. Next, the angle between the start and end point of every single migration step was determined and the accumulated frequency was visualized in rose plots, confirming the impression from the trajectory plots (Fig. 6, rose plots on the right). Migration is a process, which is the sum of directness and velocity. 16

ACCEPTED MANUSCRIPT Therefore, both factors were analyzed, showing that silencing of ephrinA1 significantly increased the velocity of the cells but decreased the directness. This was also true for one of the EphA2-siRNAs. In addition,

T

the significance of directional movement was determined using the

IP

Rayleigh test, which showed that migration was directed towards the

SC R

wound. However, higher values were seen in ephrin-A1 and EphA2 silenced cells compared to siCo-transfected cells, which is indicative for a more directed movement in case of control cells. Interestingly, no

NU

modulation of the velocity was observed in the case of adenoviral overexpression of ephrin-A1. However, overexpression of ephrin-A1 led

MA

to an increased directness and lower values in the Rayleigh test when compared to the control cells, indicating a more directed migration

D

towards the wound.

TE

The influence of ephrin-A1 silencing and overexpression in endothelial

CE P

cells on the activation of its receptor EphA2 was analyzed by immunoblots (Fig. 7). Silencing of ephrin-A1 showed increased EphA2 expression with increased phosphorylation of Ser897 and decreased

AC

phosphorylation of Tyr588 (Fig. 7A). In contrast, overexpression of ephrin-A1 led to a dramatic decrease in total EphA2 protein level. However, phosphorylation of Tyr588 increased whereas Ser897 decreased by overexpression of ephrin-A1 (Fig. 7B).

3.5 Ephrin-A1 modulates focal adhesions and acts as a repulsive factor The above demonstrated data showed an influence of ephrin-A1 on endothelial migration and proliferation, thereby raising questions concerning the underlying mechanisms. Ephrin-A1 might modulate the migratory or proliferative activity of endothelial cells by changing the cytoskeletal dynamics. Therefore, the influence of siRNA-mediated 17

ACCEPTED MANUSCRIPT silencing and adenoviral overexpression of ephrin-A1 on focal adhesion formation was investigated by using baculovirus-mediated expression of Actin-GFP and Talin-RFP to visualize the cytoskeleton and the focal

IP

T

adhesions.

SC R

Ephrin-A1 silenced HUVEC (si218 and si472) showed a slightly increased number of focal adhesions compared to siCo-transfected HUVEC (Figure 8A). Interestingly, cell size of ephrin-A1 silenced HUVEC

NU

also increased in comparison to siCo-transfected HUVEC (data not shown). In line with migration experiments, ephrin-A1 overexpressing

MA

HUVEC also showed slightly increased numbers of focal adhesions (Fig. 8B) whereas the cell size did not differ (data not shown).

D

Ephrin-A1 seems to have an influence on endothelial migration in terms

TE

of sensing the direction of migration. To test whether or not ephrin-A1 acts as a repulsive factor for migrating endothelial cells, a modified

CE P

wound healing assay was performed. As shown in Fig. 9A, two wells of a four-well chamber were coated with proteins (IgG-Fc as a control and ephrin-A1 Fc). In the other two wells, endothelial cells were seeded and

AC

allowed to grow to a confluent monolayer. The silicon spacer was removed and cells were cultivated for up to 9 days. Within this time, the endothelial cells grew and migrated circularly around the wells. Interestingly, the cells stopped to overgrow upon contact with the ephrinA1 coated area. In contrast, the IgG-Fc coated area was rapidly overgrown by the cells (Fig. 9A, middle).

18

ACCEPTED MANUSCRIPT 3 Discussion The results of the present study demonstrate that ligand ephrin-A1 is a

T

crucial determinant for the proliferative and migratory behavior of

IP

endothelial cells.

SC R

Initially we observed a positive correlation between cell density and the expression of ephrin-A1 in HUVEC reaching a maximal expression in confluent endothelial cells (Fig. 1). This effect was also observed in

NU

arterial endothelial cells (HUAEC) (Fig. 2). Aside from this correlation, serum starvation was also identified as an inductor of ephrin-A1

MA

expression (Fig. 3A). The other way round, an injury to the confluent endothelial layer reduces the expression of ephrin-A1 (Fig. 3B) and modulates the phosphorylation of distinct phospho-sites of the EphA2

D

receptor (Fig. 3C). Taken together, these results suggest that ephrin-A1

TE

expression is higher in quiescent endothelial cells compared to

CE P

proliferating endothelial cells. Next, we tested whether or not ephrin-A1 or EphA2 modulation itself can influence endothelial proliferation. In line with the expression data, silencing of ephrin-A1 and EphA2 induced

AC

proliferation, whereas overexpression decreased proliferation. Apart from proliferation, endothelial migration seemed to be also dependent on ephrin-A1. In particular, the mode of migration was regulated by ephrinA1 and its receptor EphA2. A reduction of ephrin-A1 or EphA2 led to an uncoordinated but faster migration whereas the overexpression of ephrin-A1 did not influence velocity but resulted in a more coordinated migration. As a known ephrin-A1 binding receptor, the involvement of the Ephreceptor EphA2 in the proliferative and migratory effect of ephrin-A1 on endothelial cells was further analyzed. The expression level of ephrin-A1 and EphA2 in endothelial cells changes inversely depending on cell density. This distinct expression of high ephrin-A1 in high dense cultures 19

ACCEPTED MANUSCRIPT and low expression EphA2 might be the result of a feedback mechanism as previously described for tumor cells [31,32]. Activation of EphA2 by ephrinA1 leads to the internalization and degradation of the receptor-

T

ligand complex resulting in decreased expression of the receptor as

IP

previously shown [33] and possibly to transcriptional upregulation of the

SC R

ligand.

The activation of the EphA2 receptor can occur in a ligand-dependent but also ligand independent way. Therefore, the phosphorylation status

NU

of EphA2 was analyzed 2 h and 24 h after wounding, which showed dephosphorylation of Tyr588 and an increase in Ser897 phosphorylation

MA

after 2 h, whereas total EphA2 levels were unaltered. 24 h after injury, an increased phosphorylation of Ser897 was detected, whereas Tyr588

D

phosphorylation was only decreased by trend. The here analyzed

TE

phosphorylation site Tyr588 is known to be activated in a ligand dependent manner. The phosphorylation site S897 was described to be

CE P

phosphorylated in a ligand independent manner and is activated in migrating cells [25]. These results demonstrate that injured endothelium show no ligand induced phosphorylation of the EphA2 receptor,

AC

particularly with regard to a decreasing expression profile of ephrin-A1 in these cells.

The influence of the Eph-ephrin-system on proliferation is not entirely new. The interaction between the ligand ephrin-A1 and the receptor EphA2, in particular, has been shown to have an impact on proliferative activity in various cell types [34,35]. However, these studies focused only on the influence of Eph-receptor activity on proliferation by stimulating them with soluble recombinant ligand-proteins. The influence of endothelial ephrin-A1 itself was not analyzed in detail. Ojima and colleagues showed that ephrin-A1 mediated EphA2 activation in bovine retinal endothelial cells (BREC) inhibited VEGFR2-phosphorylation and 20

ACCEPTED MANUSCRIPT its downstream signaling cascade. This resulted in a reduced VEGFinduced migration, tube formation capacity as well as a reduced proliferation of BREC [35]. Similar results were obtained by other groups,

T

who treated endothelial cells with soluble ephrin-A1 and observed a

IP

reduced endothelial proliferation [34,36]. Due to the observed importance

SC R

of ephrin-A1 and EphA2 interaction in the regulation of endothelial proliferation and migration, an involvement of VEGF-receptor signaling might be conceivable. The antiproliferative effect of endothelial ephrin-A1

NU

might be the result from an enhanced and continuous EphA2 receptor activation, which is supported by the data analyzing the phosphorylation

MA

status of EphA2, in particular Y588, in the injury experiments (Fig. 3C and D). In addition, the decreased expression of EphA2 in cultures with

D

higher density is also a sign of ligand induced receptor activation, which

TE

is known to lead to proteosomal degradation of the receptor [33]. The data of Brantley-Sieders et al. stand in contrast to this hypothesis: in

MPMEC

CE P

EphA2-deficiency

(murine

pulmonary

microvascular

endothelial cells) shows no impact on endothelial proliferation, apoptosis and survival in comparison to wildtype cells [37]. However, ephrin-A1

AC

reverse signaling was reported to be antiproliferative in CD4+ T-cells [38]. The influence of ephrin-A1/EphA2 on proliferation seems to be cell-type dependent. Cell density not only influences the proliferative activity of the cells but also determines the speed of migration [39]. Type-A and -B Ephreceptors have already been shown to influence the migratory activity of different cell types after stimulation with ephrin-ligands [40,41]. In regard to tumor progression, the partners, ephrin-A1 and EphA2, were known to influence the migration of tumor and endothelial cells [42]. However, the ligand ephrin-A1 was only tested as a stimulating factor of Eph receptor activation. 21

ACCEPTED MANUSCRIPT The results from our study reveal that endothelial ephrin-A1 itself is involved in the dynamics of migratory behavior of endothelial cells. An endothelial wound healing assay was performed to analyze the impact of

T

ephrin-A1 on migration and proliferation. Endothelial wound healing is

IP

coped partly by migration and partly by proliferation [43]. An induction of

SC R

proliferation and migration by applying scratches to a confluent endothelial monolayer resulted in accordance with the results of the proliferation experiments in reduced ephrin-A1 levels. The lower cell

NU

density in the endothelial defect zone must be the inductor of proliferation and migration by reduced ephrin-A1 levels and additionally

MA

the increased phosphorylation of EphA2 at S897. Interestingly, modulation of endothelial ephrin-A1 expression by siRNA

D

and overexpression, as well as silencing of EphA2 receptor, led to a

TE

faster gap closure, implying a sign of increased migration (Fig. 5). A detailed analysis by live-cell imaging revealed dramatic differences in the

CE P

mode of migration. Silencing of ephrin-A1 resulted in an increased velocity, which was not observed after adenoviral overexpression of ephrin-A1. The mode of migration differed greatly between silencing and

AC

overexpression conditions. Under silencing conditions of ephrinA1, as well as EphA2, cell migration was uncoordinated as indicated by the trajectory plots, rose plots, the directness calculation and Rayleigh test. Striking changes in the mode of migration were observed in ephrin-A1 overexpressing cells, which exhibited an increase in the directness of migration as indicated by the above mentioned parameters. In summary, the here presented results define ephrin-A1 as a crucial determinant for the orientation of the migrating endothelial cells. The signaling cascades possibly involved in this ephrin-A1 mediated regulation of migration can be manifold. Like others, we show that ephrin-A1 can activate the RhoAsystem, which leads to an increased actin-polymerization [20,44,45]. 22

ACCEPTED MANUSCRIPT This might be a possible mechanism that can influence migration properties of endothelial cells. In addition, cell-matrix adhesion and focal adhesion formation has been shown to be influenced by the Eph/ephrin-

T

system, especially EphA4, which is also expressed in endothelial cells

IP

[19,46]. It is known that focal adhesions as an anchoring element also

SC R

influences endothelial barrier and permeability [47], which might be relevant in important processes like inflammatory diseases. Therefore, a disturbed focal adhesion formation, as shown in Fig. 8, might lead to a

NU

modified forward movement.

In the case of the overexpression experiments, a highly focused

MA

migration of ephrin-A1 overexpressing endothelial cells might be the result of an increased repulsive response between endothelial cells and,

D

thus, resulting in a “push” towards the endothelial defect [48]. This would

TE

also be in line with the phenomenon of border definition by ephrin-A1 as shown in Fig. 9A. One possible model is illustrated in Figure 9B. The

CE P

expression gradient of ephrin-A1, which is induced by a lower cell density in the injured areas of the endothelial layer, leads to a territorial orientation of the cells. The leading cells at the wound barrier exhibit a ephrin-A1

AC

lower

expression

possibly

combined

with

a

lower

phosphorylation of Y588 and higher phosphorylation of S897 of the EphA2 receptor and, therefore, develop a migratory activity towards the injured zone. In the retinotectal system, for instance, it has been shown that gradual expression of ephrin-ligands and Eph-receptors represents a base for the regulation of adhesion and repulsion in connection with topographic mapping [49]. In conclusion, the results of the present study show that ephrin-A1 itself together with EphA2 are important players in endothelial proliferation and migration. Therefore, ephrin-A1 should be focused in future studies to clarify

its

potential

as

a

modulating

factor

in

promoting

re23

ACCEPTED MANUSCRIPT endothelialization, which is also an important step during the process of in-stent restenosis. After stent implantation, the denudated inner layer of the vessel needs to be re-endothelialized, however, uncontrolled or

T

overshooting cell proliferation/migration should be avoided. This frail

AC

CE P

TE

D

MA

NU

SC R

could be enabled by the ligand ephrin-A1.

IP

balance of pro- and antiproliferative/migratory activity of the endothelium

24

ACCEPTED MANUSCRIPT Acknowledgement This work was supported by MedDrive grants to S.J. and D.M.P. from the

T

Medical Faculty of the TU Dresden and a stipend to E.W. from the Else-

IP

Kröner-Promotionskolleg Dresden. We want to thank Prof. Triantafyllos

helpful advices.

SC R

Chavakis, a member of the TAC committee of E.W., for his critical and We want to thank Anita Maennel for her excellent

technical assistance. Furthermore, we want to thank Sonia Tien for

NU

proofreading of the manuscript.

MA

Contributions

E.W., S.J., G.E., A.A. and D.M.P. performed the experiments. R.S. and S.W. helped to analyze the data from the migration assays. B.W. and

D

R.H.S. provided critical review to the manuscript and contributed to the

AC

CE P

TE

experimental design. E.W., S.J., G.E. and D.M.P. wrote the manuscript.

25

ACCEPTED MANUSCRIPT Fig. 1 Ephrin-A1 expression is associated to cell density (A) HUVEC were seeded in different cell densities (20000, 10000, 5000

T

cells/cm2) and allowed to grow for 24, 48 and 72 h. The mRNA

IP

expression of different ephrin-ligands and Eph-receptors was evaluated

SC R

by real-time PCR. (B) The heat map shows the expression of important members of the Eph/ephrin-system in relation to the cell density. (C) The most pronounced effect was seen in the case of ephrin-A1 which

NU

correlates highly with the cell density (A-C; n=6). (D) The effect was also detectable on protein level. (n=4) (E) HUVEC were seeded in two

MA

different cell densities (20000 and 10000 cells/cm2) and were cultivated up to 11d in order to reach confluence. Monitoring of ephrin-A1 levels revealed a steady-state expression after 5 days (n=3). (F) The

D

expression of the ephrin-A1 receptor EphA2 was evaluated by Western

TE

Blotting 48 and 72h after seeding of the cells in different densities. This

CE P

analysis showed an inverse correlation of cell density and EphA2 expression (n=6). (§ indicates significant differences to the highest cell

AC

density of the corresponding time point)

26

ACCEPTED MANUSCRIPT Fig. 2 Ephrin-A1 expression in HUAEC HUAEC were seeded in different cell densities (20000, 10000, 5000

T

cells/cm2) and allowed to grow for 24, 48 and 72 h. The mRNA

IP

expression of different ephrin-ligands and Eph-receptors was evaluated

SC R

by real-time PCR. (A) The heatmap shows the expression of important members of the Eph/ephrin-system in relation to cell density. (B) The mRNA expression of ephrin-A1 correlates with the density of HUAEC

AC

CE P

TE

D

MA

NU

(A&B; n=5). (C) The effect is also seen at protein level (n=4).

27

ACCEPTED MANUSCRIPT Fig. 3 Growth arrest and induction modulates ephrin-A1 expression HUVECs were cultivated with different amounts of serum for 4 or 24 h

T

and ephrin-A1 expression was monitored. (A) Serum-starvation induces

IP

ephrin-A1 mRNA expression (upper chart) (n=5). This effect is also

SC R

reflected in the protein level (lower chart) (n=5). To evaluate the consequence of an injury on confluent endothelial cells for the expression of ephrin-A1, different amounts of scratches were applied.

NU

Ephrin-A1 mRNA (B, upper chart) and protein expression (B, lower chart) were measured 24 h after the injury, showing decreased ephrin-A1 levels

MA

in correlation to the size of the wound (mRNA: n=4, protein: n=6). (C) The effect of injury to a confluent endothelial layer on EphA2 phosphorylation was analyzed 2 h after wounding. Injury leads to

D

dephosphorylation of Tyr588 and an increase in Ser897 phosphorylation

TE

whereas total EphA2 levels were unaltered (n=3). (D) 24 h after injury, an

CE P

increased phosphorylation of Ser897 was detected, whereas, Tyr588

AC

phosphorylation was only decreased by trend (n=4).

28

ACCEPTED MANUSCRIPT Fig. 4 Ephrin-A1 expression is linked to proliferation The influence of ephrin-A1 on endothelial proliferation was tested by

T

BrdU incorporation. (A) BrdU incorporation within 8 h was measured in

IP

ephrin-A1 silenced cells (si218 and si472) or endothelial cells transfected

SC R

with a nonsense siRNA (siCo) 48 h after transfection. Silencing of ephrinA1 leads to increased endothelial proliferation (si218 n=4; si472: n=8). Furthermore, the effect of EphA2 silencing on BrdU-incorporation was

NU

measured using two different siRNAs targeting EphA2 (si177, si2181) showing also increased proliferation of EphA2-silenced cells (n=4). (C)

MA

The opposite was observed in cells transduced with an adenovirus overexpressing ephrin-A1 (n=8). Silencing and overexpression efficiency

AC

CE P

TE

EphA2 (B) protein levels.

D

was proven by Western Blot analysis of the ephrin-A1 (A and C) and

29

ACCEPTED MANUSCRIPT Fig. 5 Ephrin-A1 and EphA2 regulate endothelial migration Influence of ephrin-A1 and EphA2 silencing as well as overexpression of

T

ephrinA1 on endothelial migration was studied in a wound-healing assay.

IP

Silencing of ephrin-A1 (A) or EphA2 (B) as well as overexpression of

SC R

ephrin-A1 (C) led to a quicker gap closure, suggesting a faster migration

AC

CE P

TE

D

MA

NU

(n=4).

30

ACCEPTED MANUSCRIPT Fig. 6 Ephrin-A1 and EphA2 regulate the mode of endothelial migration

T

Migratory behavior was investigated in a live-cell imaging approach. (A)

IP

Analysis of the migratory paths of single cells revealed that silencing of

SC R

ephrin-A1 (si218 and si472) led to an uncoordinated movement of the cells in comparison to the control cells (siCo). (B) Similar results were obtained in EphA2-silenced cells (si177 and si2181). (C) In contrast,

NU

overexpression of ephrin-A1 (AdXEFNA1) triggered straight-forward migration in the direction of the gap compared to control-transduced cells

MA

(AdXlacZ). These results were confirmed by analysis of the angle in which each cell moved in each step, which is illustrated in the Rose plots (right part of the figure). Determination of velocity revealed a faster

D

movement of ephrin-A1 silenced cells compared to the control. In

TE

contrast, overexpression had no influence on velocity. The results

CE P

concerning the migratory direction were confirmed by the calculation of directness, which is lower in ephrin-A1 and EphA2 silenced cells and increased in ephrin-A1 overexpression. Furthermore, significances of the

AC

Rayleigh test confirmed the aforementioned observations (n=3, numbers of single cell paths analyzed can be found in the trajectory plots).

31

ACCEPTED MANUSCRIPT Fig. 7 Ephrin-A1 modulation leads to changes in EphA2 phosphorylation The effect of ephrin-A1 silencing and overexpression on EphA2

T

phosphorylation was analyzed by immunoblotting. (A) Silencing of

IP

ephrin-A1 (si218 and si472) leads to an increased EphA2 expression in

whereas

phosphorylation

of

SC R

endothelial cells, phosphorylation of Tyr588 was decreased by trend Ser897

was

increased

(n=4).

(B)

Overexpression led to a dramatic decrease in total EphA2 protein level.

NU

However, phosphorylation of Tyr588 was increased by overexpression of ephrin-A1 whereas the phosphorylation of Ser897 decreased, but in

MA

relation to decreased EphA2 levels, the phosphorylation status of this

AC

CE P

TE

D

site was unaltered (n=3).

32

ACCEPTED MANUSCRIPT Fig. 8 Ephrin-A1 expression modulates focal adhesions Transduction of HUVEC with baculoviruses expressing Actin-GFP and

T

Talin-RFP was used to analyze cytoskeleton and focal adhesions. (A&B)

IP

Exemplary images of actin and talin expression in ephrin-A1 silenced

AdXEFNA1)

and

respective

SC R

cells (A, si218, si472) and ephrin-A1 overexpressing cells (B, controls

(A,

siCo;

B,

AdXlacZ).

Quantification revealed slightly increased formation of focal adhesions

NU

per area in ephrin-A1 silenced cells, which was more pronounced by ephrin-A1 overexpression (n=3 with 6-8 analyzed cells per independent

AC

CE P

TE

D

MA

experiment).

33

ACCEPTED MANUSCRIPT Fig. 9 Ephrin-A1 acts as a repulsive factor for migrating endothelial cells

T

(A) Endothelial cells were seeded in 2 wells of a 4 well chamber,

IP

whereas the other 2 wells were coated with ephrin-A1 Fc or IgG-Fc for

SC R

control. After removal of the spacers, endothelial migration was monitored over a period of 6-9 days to show that EC stop migrating when they reach the ephrin-A1 coated area (n=3). (B) Putative model shows

NU

that ephrin-A1 is crucial for endothelial migration and determination of

AC

CE P

TE

D

MA

the direction of migration.

34

ACCEPTED MANUSCRIPT References

T

[1] L. Lamalice, B.F. Le, J. Huot, Endothelial cell migration during

IP

angiogenesis, Circ. Res 100 (2007) pp. 782-794.

SC R

[2] S. Li, N.F. Huang, S. Hsu, Mechanotransduction in endothelial cell migration, J. Cell Biochem. 96 (2005) pp. 1110-1126.

NU

[3] A. Davy, N.W. Gale, E.W. Murray, R.A. Klinghoffer, P. Soriano, C. Feuerstein, S.M. Robbins, Compartmentalized signaling by GPI-

MA

anchored ephrin-A5 requires the Fyn tyrosine kinase to regulate

D

cellular adhesion, Genes Dev. 13 (1999) pp. 3125-3135. 'Eph'ective signaling: forward,

TE

[4] K.K. Murai and E.B. Pasquale,

CE P

reverse and crosstalk, J. Cell Sci. 116 (2003) pp. 2823-2832. [5] E.B. Pasquale,

Eph-ephrin bidirectional signaling in physiology

and disease, Cell 133 (2008) pp. 38-52.

AC

[6] R. Ross,

Atherosclerosis--an inflammatory disease, N. Engl. J.

Med. 340 (1999) pp. 115-126. [7] M. Nakatani, Y. Takeyama, M. Shibata, M. Yorozuya, H. Suzuki, S. Koba, T. Katagiri,

Mechanisms of restenosis after coronary

intervention: difference between plain old balloon angioplasty and stenting, Cardiovasc. Pathol. 12 (2003) pp. 40-48. [8] T. Inoue, K. Croce, T. Morooka, M. Sakuma, K. Node, D.I. Simon, Vascular

inflammation

endothelialization,

and

restenosis,

repair: and

implications

stent

thrombosis,

for

re-

JACC.

Cardiovasc. Interv. 4 (2011) pp. 1057-1066. 35

ACCEPTED MANUSCRIPT [9] A. Kibos, A. Campeanu, I. Tintoiu, Pathophysiology of coronary artery in-stent restenosis, Acute. Card Care 9 (2007) pp. 111-119.

T

[10] J. Egea and R. Klein, Bidirectional Eph-ephrin signaling during

SC R

IP

axon guidance, Trends Cell Biol. 17 (2007) pp. 230-238. [11] J.P. Himanen, N. Saha, D.B. Nikolov, Cell-cell signaling via Eph

NU

receptors and ephrins, Curr. Opin. Cell Biol. 19 (2007) pp. 534-542. [12] E.B. Pasquale, Eph receptor signalling casts a wide net on cell

MA

behaviour, Nat Rev Mol. Cell Biol. 6 (2005) pp. 462-475. [13] M.G. Coulthard, M. Morgan, T.M. Woodruff, T.V. Arumugam, S.M. Eph/Ephrin

D

Taylor, T.C. Carpenter, M. Lackmann, A.W. Boyd,

TE

signaling in injury and inflammation, Am. J. Pathol. 181 (2012) pp.

CE P

1493-1503.

[14] R.H. Adams, G.A. Wilkinson, C. Weiss, F. Diella, N.W. Gale, U. Deutsch, W. Risau, R. Klein, Roles of ephrinB ligands and EphB

AC

receptors

in

cardiovascular

development:

demarcation

of

arterial/venous domains, vascular morphogenesis, and sprouting angiogenesis, Genes Dev. 13 (1999) pp. 295-306. [15] S.S. Gerety and D.J. Anderson, Cardiovascular ephrinB2 function is essential for embryonic angiogenesis, Development 129 (2002) pp. 1397-1410. [16] Y. Wang, M. Nakayama, M.E. Pitulescu, T.S. Schmidt, M.L. Bochenek, A. Sakakibara, S. Adams, A. Davy, U. Deutsch, U. Luthi, A. Barberis, L.E. Benjamin, T. Makinen, C.D. Nobes, R.H.

36

ACCEPTED MANUSCRIPT Adams,

Ephrin-B2 controls VEGF-induced angiogenesis and

lymphangiogenesis, Nature 465 (2010) pp. 483-486.

T

[17] J. Gong, R. Korner, L. Gaitanos, R. Klein, Exosomes mediate cell

IP

contact-independent ephrin-Eph signaling during axon guidance, J.

SC R

Cell Biol. 214 (2016) pp. 35-44.

[18] M.M. Vihanto, J. Plock, D. Erni, B.M. Frey, F.J. Frey, U. Huynh-Do,

NU

Hypoxia up-regulates expression of Eph receptors and ephrins in

MA

mouse skin, FASEB J. 19 (2005) pp. 1689-1691. [19] G. Ende, D.M. Poitz, E. Wiedemann, A. Augstein, J. Friedrichs, S. Giebe, S. Weinert, C. Werner, R.H. Strasser, S. Jellinghaus, TNF-

D

alpha-mediated adhesion of monocytes to endothelial cells-The

TE

role of ephrinA1, J. Mol. Cell Cardiol. 77 (2014) pp. 125-135.

CE P

[20] S. Jellinghaus, D.M. Poitz, G. Ende, A. Augstein, S. Weinert, B. Stutz, R.C. Braun-Dullaeus, E.B. Pasquale, R.H. Strasser, EphrinA1/EphA4-mediated adhesion of monocytes to endothelial cells,

AC

Biochim. Biophys. Acta 1833 (2013) pp. 2201-2211. [21] P. Goichberg, Y. Bai, D. D'Amario, J. Ferreira-Martins, C. Fiorini, H. Zheng, S. Signore, M.F. del, S. Ottolenghi, D.A. D'Alessandro, R.E. Michler, T. Hosoda, P. Anversa, J. Kajstura, M. Rota, A. Leri, The ephrin A1-EphA2 system promotes cardiac stem cell migration after infarction, Circ. Res. 108 (2011) pp. 1071-1083. [22] N. Kaplan, A. Fatima, H. Peng, P.J. Bryar, R.M. Lavker, S. Getsios, EphA2/Ephrin-A1 signaling complexes restrict corneal epithelial cell migration, Invest Ophthalmol. Vis. Sci. 53 (2012) pp. 936-945. 37

ACCEPTED MANUSCRIPT [23] E.B. Pasquale, Eph receptors and ephrins in cancer: bidirectional signalling and beyond, Nat. Rev. Cancer 10 (2010) pp. 165-180.

cell

repulsion

by

increasing

EphA2

S897

SC R

mediated

Protein kinase A can block EphA2 receptor-

IP

E.B. Pasquale,

T

[24] A. Barquilla, I. Lamberto, R. Noberini, S. Heynen-Genel, L.M. Brill,

phosphorylation, Mol. Biol. Cell 27 (2016) pp. 2757-2770.

NU

[25] H. Miao, D.Q. Li, A. Mukherjee, H. Guo, A. Petty, J. Cutter, J.P. Basilion, J. Sedor, J. Wu, D. Danielpour, A.E. Sloan, M.L. Cohen,

MA

B. Wang, EphA2 mediates ligand-dependent inhibition and ligandindependent promotion of cell migration and invasion via a

D

reciprocal regulatory loop with Akt, Cancer Cell 16 (2009) pp. 9-20.

Schmeisser,

TE

[26] A. Augstein, D.M. Poitz, R.C. Braun-Dullaeus, R.H. Strasser, A. Cell-specific and hypoxia-dependent regulation of

CE P

human HIF-3alpha: inhibition of the expression of HIF target genes in vascular cells, Cell Mol. Life Sci. 68 (2011) pp. 2627-2642.

AC

[27] J. Hellemans, G. Mortier, P.A. De, F. Speleman, J. Vandesompele, qBase

relative

quantification

framework

and

software

for

management and automated analysis of real-time quantitative PCR data, Genome Biol. 8 (2007) p.R19. [28] D.M. Poitz, A. Augstein, C. Gradehand, G. Ende, A. Schmeisser, R.H. Strasser, Regulation of the Hif-system by micro-RNA 17 and 20a - role during monocyte-to-macrophage differentiation, Mol. Immunol. 56 (2013) pp. 442-451. [29] S.D. Funk, A. Yurdagul, Jr., P. Albert, J.G. Traylor, Jr., L. Jin, J. Chen, A.W. Orr, EphA2 activation promotes the endothelial cell 38

ACCEPTED MANUSCRIPT inflammatory response: a potential role in atherosclerosis, Arterioscler. Thromb. Vasc. Biol. 32 (2012) pp. 686-695.

T

[30] A. Sakamoto, Y. Sugamoto, Y. Tokunaga, T. Yoshimuta, K.

IP

Hayashi, T. Konno, M.A. Kawashiri, Y. Takeda, M. Yamagishi,

SC R

Expression profiling of the ephrin (EFN) and Eph receptor (EPH) family of genes in atherosclerosis-related human cells, J. Int. Med.

NU

Res 39 (2011) pp. 522-527.

[31] M. MacRae, R.M. Neve, P. Rodriguez-Viciana, C. Haqq, J. Yeh, C. A conditional feedback loop

MA

Chen, J.W. Gray, F. McCormick,

regulates Ras activity through EphA2, Cancer Cell 8 (2005) pp.

D

111-118.

Debinski,

TE

[32] J. Wykosky, E. Palma, D.M. Gibo, S. Ringler, C.P. Turner, W. Soluble monomeric EphrinA1 is released from tumor

CE P

cells and is a functional ligand for the EphA2 receptor, Oncogene 27 (2008) pp. 7260-7273.

AC

[33] O. Sabet, R. Stockert, G. Xouri, Y. Bruggemann, A. Stanoev, P.I. Bastiaens, Ubiquitination switches EphA2 vesicular traffic from a continuous safeguard to a finite signalling mode, Nat Commun. 6 (2015) p.8047. [34] H. Miao, B.R. Wei, D.M. Peehl, Q. Li, T. Alexandrou, J.R. Schelling, J.S. Rhim, J.R. Sedor, E. Burnett, B. Wang, Activation of EphA receptor tyrosine kinase inhibits the Ras/MAPK pathway, Nat Cell Biol. 3 (2001) pp. 527-530. [35] T. Ojima, H. Takagi, K. Suzuma, H. Oh, I. Suzuma, H. Ohashi, D. Watanabe, E. Suganami, T. Murakami, M. Kurimoto, Y. Honda, N. 39

ACCEPTED MANUSCRIPT Yoshimura, EphrinA1 inhibits vascular endothelial growth factorinduced

intracellular

signaling

and

suppresses

retinal

neovascularization and blood-retinal barrier breakdown, Am. J.

IP

T

Pathol. 168 (2006) pp. 331-339.

Ephrin-A1 is up-

SC R

[36] Y. Song, X.P. Zhao, K. Song, Z.J. Shang,

regulated by hypoxia in cancer cells and promotes angiogenesis of HUVECs through a coordinated cross-talk with eNOS, PLoS. One.

NU

8 (2013) p.e74464.

MA

[37] D.M. Brantley-Sieders, J. Caughron, D. Hicks, A. Pozzi, J.C. Ruiz, J. Chen, EphA2 receptor tyrosine kinase regulates endothelial cell migration and vascular assembly through phosphoinositide 3-

TE

pp. 2037-2049.

D

kinase-mediated Rac1 GTPase activation, J. Cell Sci. 117 (2004)

CE P

[38] J.G. Wohlfahrt, C. Karagiannidis, S. Kunzmann, M.M. Epstein, W. Kempf, K. Blaser, C.B. Schmidt-Weber,

Ephrin-A1 suppresses

AC

Th2 cell activation and provides a regulatory link to lung epithelial cells, J. Immunol. 172 (2004) pp. 843-850. [39] R. Riahi, J. Sun, S. Wang, M. Long, D.D. Zhang, P.K. Wong, Notch1-Dll4 signalling and mechanical force regulate leader cell formation during collective cell migration, Nat Commun. 6 (2015) p.6556. [40] H.S. Hjorthaug and H.C. Aasheim, Ephrin-A1 stimulates migration of CD8+CCR7+ T lymphocytes, Eur. J. Immunol. 37 (2007) pp. 2326-2336.

40

ACCEPTED MANUSCRIPT [41] J.J. Steinle, C.J. Meininger, R. Forough, G. Wu, M.H. Wu, H.J. Granger,

Eph B4 receptor signaling mediates endothelial cell

migration and proliferation via the phosphatidylinositol 3-kinase

Ephs and ephrins in cancer:

SC R

[42] A. Beauchamp and W. Debinski,

IP

T

pathway, J. Biol. Chem. 277 (2002) pp. 43830-43835.

ephrin-A1 signalling, Semin. Cell Dev. Biol. 23 (2012) pp. 109-115.

NU

[43] A.W. Clowes, T.R. Kirkman, M.A. Reidy, Mechanisms of arterial graft healing. Rapid transmural capillary ingrowth provides a

MA

source of intimal endothelium and smooth muscle in porous PTFE prostheses, Am. J. Pathol. 123 (1986) pp. 220-230.

D

[44] M. Sahin, P.L. Greer, M.Z. Lin, H. Poucher, J. Eberhart, S.

TE

Schmidt, T.M. Wright, S.M. Shamah, S. O'connell, C.W. Cowan, L. Hu, J.L. Goldberg, A. Debant, G. Corfas, C.E. Krull, M.E.

CE P

Greenberg, Eph-dependent tyrosine phosphorylation of ephexin1 modulates growth cone collapse, Neuron 46 (2005) pp. 191-204.

AC

[45] S.M. Shamah, M.Z. Lin, J.L. Goldberg, S. Estrach, M. Sahin, L. Hu, M. Bazalakova, R.L. Neve, G. Corfas, A. Debant, M.E. Greenberg, EphA receptors regulate growth cone dynamics through the novel guanine nucleotide exchange factor ephexin, Cell 105 (2001) pp. 233-244. [46] C. Bourgin, K.K. Murai, M. Richter, E.B. Pasquale, The EphA4 receptor regulates dendritic spine remodeling by affecting beta1integrin signaling pathways, J. Cell Biol. 178 (2007) pp. 1295-1307. [47] M.H. Wu,

Endothelial focal adhesions and barrier function, J.

Physiol 569 (2005) pp. 359-366. 41

ACCEPTED MANUSCRIPT [48] J. Cayuso, Q. Xu, D.G. Wilkinson,

Mechanisms of boundary

formation by Eph receptor and ephrin signaling, Dev. Biol. 401

T

(2015) pp. 122-131.

IP

[49] D.D. O'Leary and D.G. Wilkinson, Eph receptors and ephrins in

AC

CE P

TE

D

MA

NU

SC R

neural development, Curr. Opin. Neurobiol. 9 (1999) pp. 65-73.

42

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

43

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

44

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

45

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

46

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

47

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

48

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

49

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

50

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

51

ACCEPTED MANUSCRIPT Regulation of endothelial migration and proliferation by ephrin-A1 Elisa Wiedemann1,a, Stefanie Jellinghaus1,a, Georg Ende1,a, Antje

SC R

IP

Strasser1, David M. Poitz1

T

Augstein1, Ronny Sczech2, Ben Wielockx3, Sönke Weinert4, Ruth H.

NU

Highlights:

AC

CE P

TE

D

MA

1.) EphrinA1 positively correlates with cell density in HUVEC. 2.) EphrinA1 regulates cell proliferation in HUVEC. 3.) EphrinA1 modulates the mode of migration of endothelial cells. 4.) EphrinA1 is a stop-signal for migrating endothelial cells.

52