DARC Regulates Angiogenesis by Mediating Vascular Endothelial Cell Migration

Background: DARC (The Duffy antigen receptor for chemokines) is a kind of glycosylated membrane protein that binds to members of the CXC chemokine family associated with angiogenesis and has recently been reported to be implicated in diverse normal physiologic processes. This study aimed to investigate the involvement of DARC in angiogenesis, which is known to generate new capillary blood vessels from preexisting ones. Methods: HDMECs (Human dermal microvascular endothelial cells) were divided into two groups (DARC overexpression group, and control group). We used Brdu staining to detect cell proliferation, and wound healing assay to detect cell migration. Then tube formation assay were observed. Also, western blot and immunouorescent staining were used to estimate the relationship between DARC and RhoA (Ras homolog gene family, member A). Results: HDMECs proliferation, migration, and tube formation were inhibited signicantly when DARC was overexpressed intracellular. DARC impaired microlament dynamics and intercellular connection in migrating cells, and RhoA activation underlay the effect of DARC on endothelial cell. Furthermore, DARC inhibited the formation of new capillaries in vitro. Conclusion: Our ﬁ ndings revealed the role of DARC in the angiogenic process and provided a novel mechanism for RhoA activation during endothelial cell migration and angiogenesis.


Backgroud
It is accepted that angiogenesis is the formation of new capillary blood vessels from preexisting ones, and it is a balanced process regulated by many angiogenic factors and anti-angiogenesis factors [1,2].
The process of generating new blood vessels is closely related to physiological and pathological processes such as tissue development, wound repair, and tumor growth [3][4][5]. Among them, the growth and development of tissues require the new blood vessels to provide nutrients, and the preparation of the wound bed also requires the new blood vessels to provide nutrients and various growth factors [6,7]. This process is highly controlled by the intricate balance of both proangiogenic and antiangiogenic factors, and involves a cascade of events of which migration of capillary endothelial cells is an essential component [8,9].
The Duffy antigen receptor for chemokines (DARC) is a glycosylated membrane protein that belongs to the subfamily of silent chemokine receptors, which is expressed mainly on erythrocytes and endothelial cells [10]. However, the function of DARC expressed on endothelial cells remains unclear. Previous studies have indicated that DARC is a negative regulator of breast carcinoma and inhibits tumor growth by inhibiting tumorigenic chemokine, and DARC overexpression in breast cancer patients is associated with a better prognosis, lower metastatic potential, and less lymphocytic in ltrate [11][12][13]. Similar experimental results were reported in the lung [14] and laryngeal squamous cell carcinomas [15]. Another study showed that tumors from DARC-de cient mice had higher intra-tumor concentrations of angiogenic chemokines, larger tumor vessel density, and they had greatly augmented prostate tumor growth [16].
The negative involvements of DARC in the development of prostate cancer [16] and breast cancer [13] suggested that this protein might play a negative regulatory role in angiogenesis, of which vascular endothelial cell proliferation and migration are essential components. Our study was undertaken to test this hypothesis directly. There has been a growing body of evidence that coordinated action of microtubules and actin laments is critical for cell migration [17,18]. Microtubule dynamics are precisely regulated by various microtubule-binding proteins and the Rho family [18][19][20]. Micro laments can also regulate the activity of Rho GTPases, and the interaction between cytoskeleton and Rho GTPases is required for e cient cell migration and intercellular connection [20]. We hypothesized that the negative regulation of DARC may be related to its effect on the cytoskeleton and thus inhibiting cell migration.
However, the precise role of DARC in angiogenesis and the exact molecular mechanisms of DARC in the endothelial cells are still not clear. Therefore, the current study was to explore the involvement of the DARC in angiogenesis in vitro and also in vivo respectively.
Adenoviruses that over-express DARC (Ad-DARC, experimental group) and empty vector of adenovirus without target gene (Ad-eGFP, control group) were made by the Cyagen Biotechnology Co., Ltd. All the adenoviruses were labeled with green uorescent tags. HDMECs were infected with Ad-DARC or Ad-eGFP at 1 × 10 6 PFU, in a minimal volume of serum-free medium, and the infection e ciency of HDMECs was con rmed to be above 80% by preliminary experiments.

Cell proliferation and viability assay
Cells were grown on glass coverslips and cultured with anti-bromodeoxyuridine (BrdU) antibody (1:500, ThermoFisher #B35128) at 4℃ through the night. After the coverslips were washed with phosphate buffer saline (PBS, Gibco, catalog number 20012027), the cells were xed with 4% paraformaldehyde for 15 minutes at room temperature. Then the coverslips were placed in 2N HCl for 30 minutes and washed with PBS. After that, the cells were incubated with antibodies against BrdU (Abcam, ab6326), and Cy3conjugated secondary antibodies, and then unspeci c bindings were blocked with PBS containing 5% goat serum. Finally, the cells were stained with 4',6-diamidino-2-phenylindole (DAPI) for 10 minutes, and then examined by uorescence microscopy.
CellTiter-Glo Luminescent Cell Viability Assay Kit (Cat# G7570, Promega) was used according to the manufacturer's protocol. Brie y, cells were cultured in 96-well plates and infected with Ad-DARC as well as Ad-eGFP in the following day. Equal volumes of medium and CellTiter-Glo Reagent were added at 1st day, 3rd day, and 5th day respectively, and the results were analyzed. Luminescence was detected by a multiwell scanning spectrophotometer.

Cell migration assay
Oris™ Cell Migration Assay Kit (PLATYPUS Technologies, Product No: CMA1.101) was used for cell migration detection according to the manufacturer's protocol. Brie y, cells were infected with Ad-DARC as well as Ad-eGFP and then seeded in each test well through one of the side ports of the Oris™ Cell Seeding Stopper. When cells had grown to be a con uent monolayer, we removed all the stoppers. Cells were imaged kinetically for up to 24 hours in a Cytation TM 3(BioTek Instruments, Inc.) microplate imager with incubation at 37℃ and a gas control module set to 5% CO 2 using the settings outlined. Then photos were taken at 6-hour intervals with 3 random locations to examine the extent of wound closure. The gap area was calculated with Image-Pro Plus version 6.0 (Media Cybernetics, Inc., Warrendale, PA). The analysis relied on the determination of the area of the detection zone of post-migration wells in comparison with pre-migration wells in which no migration occurred to calculate post-migration percent closure using imaging data.

Tube formation assay
Matrigel (BD Biosciences) was added to 96-well plates with a volume of 50 µl per well, and HDMECs (2 × 10 4 cells/100 µl) were cultured after the matrigel had solidi ed at 37℃. To observe tube formation, photographs were taken 24 hours later with the uorescence microscope, and the degree of tube formation was quanti ed by measuring the branch points with the use of Image J software.

Cytoskeleton staining
Cells were grown on glass coverslips and xed with 4% paraformaldehyde for 15 minutes at room temperature. Then they were incubated with 10 µg/ml phalloidine for 30 minutes. After that, the cells were washed with PBS and stained with DAPI for 10 minutes. Finally, the cells were observed under the Olympus IX71 uorescent microscope.
Immuno uorescence staining Cells on glass coverslips were xed with 4% paraformaldehyde for 15 minutes at room temperature and then permeabilized using 0.1% Triton X-100 for 5 minutes. After blocking in 5% goat serum for 60 minutes, the samples were immunostained with primary antibodies against connexin 43(1:5000, ab11369, Abcam,), RhoA(10 µg/ml, ab86297, Abcam), followed by incubation with the goat anti-mouse IgG Alexa Fluor-cy3-conjugated and the goat anti-rabbit IgG Alexa Fluor-488-conjugated secondary antibodies(1:200, Invitrogen). A confocal laser scanning microscopy with an Olympus FV 1000 device was used to observe cellular uorescence associated with the cell connection.

Western blot analysis
After being washed twice with PBS, the cells were lysed for 30 minutes in ice-cold lysis buffer containing 50 mM Tris, pH 7.5, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 1 mM sodium orthovanadate, 5 mM sodium uoride, 1 mM ethylene diamine tetraacetic acid (EDTA), and the Protease Inhibitors Cocktail (Sigma). Then the samples were boiled at 100℃ for 10 minutes and were placed in ice immediately. Cell lysates were subjected to western blot analysis using rabbit antibodies against RhoA (1 µg/ml, ab86297, Abcam) and mouse antibodies against GAPDH (1 µg/ml, ab9484, Abcam) as the primary antibodies. The signi cance of the protein expression differences was calculated at last.

In vivo matrigel angiogenesis model
Eight-week-old nude mice were purchased from the Animal Center of the Air Forth Medical University. All animal procedures were performed in accordance with the Animal Care and Use Committee at the University of Air Forth Medical. Matrigel (BD Bioscience, catalog number 354248) was diluted with serumfree Dulbecco's modi ed Eagle's medium(DMEM, Gibco, catalog number 11965092) to 10 mg/ml. All the nude mice were weighed and anesthetized by 60 µl/20 g bodyweight injection of 1% sodium pentobarbital. The dorsal skin was disinfected with 70% ethanol, then tissue engineering chambers were lled with matrigel and then implanted subcutaneously into the dorsal of athymic nude mice. Sixteen nude mice were implanted with matrigel and randomly divided into four groups: the positive control group (n = 4), tissue engineering chambers that were lled with matrigel containing angiogenic factors (heparin and vascular endothelial growth factor); the negative control group (n = 4), tissue engineering chambers that were lled with matrigel without angiogenic factors; the DARC group (n = 4), tissue engineering chambers that were lled with matrigel containing angiogenic factors and Ad-DARC adenovirus; the control group (n = 4), tissue engineering chambers that were lled with matrigel containing angiogenic factors and Ad-eGFP adenovirus. The tissue engineering chambers were removed 6 days after implanted and Dextran was injected into nude mice via the tail vein before removed to mark the blood vessels. New blood capillaries in the matrigel were observed by uorescence microscopy. To quantify the neovascularization in vivo, the cells were recovered from the matrigel by dispase (Corning, Cat. No. 354235) digestion and cell recovery solution (Corning, Cat. No. 354253). Then the recovered cells were stained with FITC-lectin for statistical analysis .

Overexpression of DARC impairs angiogenesis in vitro
To investigate the potential role of DARC in angiogenesis, we rst examined the effect of DARC overexpression on proliferation, viability, migration, and apoptosis in vitro. Two adenoviruses were used to infect HDMECs. Ad-DARC was able to overexpress DARC expression effectively in HDMECs (Fig. 1A). The cells were stained with BrdU to investigate the proliferation of DARC overexpression in the cells. The result showed that DARC overexpression inhibited HDMECs proliferation and also inhibited the proliferation caused by 20 ng/ml VEGF (Fig. 1B). To evaluate the effects of DARC on the cell viability, cell viability assay was performed on HDMECs. Overexpression of DARC resulted in signi cant inhibition of the cell viability at the 3rd day and the 5th day respectively (Fig. 1C). Similarly, cell migration was also inhibited in the DARC overexpression group observed by the Oris™ cell migration assay (Fig. 1D).

DARC regulates angiogenesis by inhibiting tube formation
To further investigate the function of DARC in angiogenesis, we examined the effect of DARC overexpression on vascular endothelial tube formation in vitro. We observed the branch point structures at 24 hours after plating cells onto matrigel. The results showed that overexpressed DARC remarkably impaired tube formation compared with the control group, and also impaired the tube formation caused by VEGF (Fig. 2). By counting the branch points, we found that overexpressed DARC in HDMECs inhibited HDMECs tube formation by 67.16% and 60.92%, respectively, normal cultured group and VEGF added group.

DARC modulates micro lament dynamics and intercellular connection in migrating cells
We conducted the cytoskeleton staining. According to the Phalloidine staining result, the cytoskeleton protein degraded and reconstructed in the DARC overexpression group, also the intercellular connection decreased signi cantly (Fig. 3A), and cell spreading was inhibited. On the basis of the nding that cells gap junctions were decreased, we then detected connexin 43. The result obviously showed that DARC overexpression inhibited the expression level of connexin 43 (Fig. 3B). These data suggested that DARC might regulate the migration of vascular endothelial cells through modulation of micro lament dynamics.

RhoA activation impairs the effects of DARC on endothelial cell migration
To gain more mechanistic insight into how DARC mediates endothelial cell migration, we detected the activities of RhoA, which is known mainly involved in cell polarization and migration primarily through interplay with the cytoskeleton [20,21]. According to the immuno uorescence staining result, the RhoA expression decreased signi cantly in the DARC overexpression group (Fig. 4A). The western-blot result showed that DARC overexpression signi cantly decreased the intracellular RhoA expression (Fig. 4B). To examine whether RhoA activation contributes to the action of DARC in endothelial cell migration, endothelial cells were transfected with pSUPER-RhoA and then wound healing assay was performed. We found that overexpression of RhoA could remarkably rescue DARC overexpression-induced migration defects (Fig. 4C), indicating that RhoA activation underlies the effect of DARC on endothelial cell migration.

DARC inhibits angiogenesis in vivo
To further con rm the in vitro data indicating the importance of DARC in angiogenesis, we studied the effect of DARC on the angiogenic response in nude mice. Tissue engineering chambers were lled with matrigel containing heparin and vascular endothelial growth factor and were implanted into the dorsal area of athymic nude mice (Fig. 5A ).
We found apparent vascular structures in the tissue engineering chambers 6 days after implantation ( Fig. 5B Positive control), and no vascular structures were observed in tissue engineering chambers that were lled with matrigel without heparin and vascular endothelial growth factor (Fig. 5B Negative  control). It is very important that the tissue engineering chamber with Ad-DARC addition obviously inhibited vascular growth which was induced by the angiogenic factors. On the contrary, the control group did not affect vascular growth into the tissue engineering chamber (Fig. 5B). By laser confocal microscope, obvious vascular structures were found in the control group as in the DARC group, and this was consistent with the results of the positive and negative controls (Fig. 5C).
In order to quantify the neovascularization in vivo, cells were recovered from the matrigel by dispase digestion and stained with FITC-lectin. As shown in Fig. 5D, the uorescence intensity of cells recovered from the DARC group was dramatically decreased compared with the control group. These results con rmed the important role of DARC in angiogenesis in vivo.

Discussion
In the past few decades, signi cant progress has been achieved in identifying angiogenic factors and inhibitors. Although lots of new inhibiting angiogenesis molecules have been discovered, the precise molecular mechanisms remain to be elucidated. In our study, we have found that DARC had the negative effects not only for endothelial cell migration and for intercellular connection in vitro but also for angiogenesis in vivo. Our results were partly consistent with certain previous studies that DARC was involved in reducing angiogenesis in tumorigenesis [16,22,23], and also DARC was highly expressed on endothelial cells post-capillary venules and endothelial cell membrane [24,25]. In these studies, Xu et al. have further investigated the mechanism, and they have shown that DARC on endothelial cells attenuated the angiogenic activity by causing senescence [26]. However, our study provides the evidence showing a novel possibility in DARC-induce angiogenesis inhibition, and we propose that DARC inhibited the migration of endothelial cells by inhibiting the expression of RhoA, resulting in the inhibition of angiogenesis.
In the current studies, we show that HDMECs proliferation was inhibited signi cantly when DARC was overexpressed intracellular. Meanwhile, the proliferation of HDMECs induced by VEGF was also inhibited obviously (Fig. 1B). Our data are also supported by the cell viability assay showing that the ATP concentration was inhibited by DARC both normal cultured cells and VEGF induced cells (Fig. 1C). As shown by overexpressing DARC and performing Oris™ cell migration assay, we revealed the effects of DARC on cell migration, that is, the inhibitory effect of DARC on cell migration became more signi cant with the increase of time (Fig. 1D). These results suggest that DARC has the effect of reducing the angiogenic potential.
Cell migration plays a central role during tube formation and cell movement in angiogenesis, and cellular movement is a complex, tightly regulated multistep process [27]. Through tube formation assay and cytoskeleton staining, we have found that DARC inhibited vascular endothelial cell migration by impairing intercellular connections that were related to micro lament dynamics.
The micro lament cytoskeleton is known to undergo dramatic rearrangement in response to signals that stimulate cell migration, and such property of micro laments relies on precise control of micro lament dynamics in cells. Micro laments can also regulate the activity of Rho GTPases, and the interaction between cytoskeleton and Rho GTPases is required for e cient cell migration and intercellular connection [21]. Interestingly, we clearly showed that the inhibitory effect of DARC on the intercellular connection was induced by RhoA, which is well known to play an important role in cell movement, cytoskeleton recombination, and cell proliferation [28][29][30]. Our data supported that RhoA expression decreased with the overexpression of DARC by showing the results of uorescence staining and protein expression (Fig. 4A, 4B). Through the results of the wound healing experiment after the overexpression of DARC and RhoA simultaneously, we found that overexpression of RhoA could remarkably rescued DARC overexpression-induced migration defects (Fig. 4C), indicating that RhoA activation underlies the effect of DARC on endothelial cell migration.
In vitro studies have demonstrated that DARC signi cantly inhibited endothelial cell growth and tube formation. Similarly, by the matrigel angiogenesis model in vivo, we found that the control group showed a large number of new capillary vessels, but only a very few new capillary vessels could be found in the DARC group. These data suggested that DARC inhibited the formation of new capillaries which induced by angiogenic factors in vivo. The in vivo results were consistent with our previous nding that DARC inhibited angiogenesis in vitro.
In conclusion, our data showing dramatic decreases in new capillary vessels formation provided strong evidence that the DARC had an important regulatory function in this process, and micro lament was involved in the regulation of this process. Our nding that RhoA inhibition was involved in DARC's effects on endothelial cell migration and angiogenesis is very interesting. Considering the interplay between RhoA activity and micro laments in migrating cells, it is conceivable that the action of DARC in Rac1 down-regulation might be an event downstream of the modulation of microtubule dynamics by DARC. However, it is still unclear that how DARC regulates micro lament through RhoA exactly and what the speci c signal transduction pathways it affects. This will be important to investigate in the future.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was supported by grants from the National Natural Science Foundation of China (Grant No: 81701909) Authors' contributions X.W., J.L., and X.L. designed the research; X.W. and Y.B. wrote the paper; and X.W., Y.L., J.L., and C.Z. performed the research and analyzed the data.