Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation Research
Search: search_blue_button Advanced Search
Circulation Research. 2007;100:1308-1316
Published online before print April 5, 2007, doi: 10.1161/01.RES.0000266662.98355.66
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
100/9/1308    most recent
01.RES.0000266662.98355.66v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Staniszewska, I.
Right arrow Articles by Marcinkiewicz, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Staniszewska, I.
Right arrow Articles by Marcinkiewicz, C.
Related Collections
Right arrow Endothelium/vascular type/nitric oxide
Right arrow Angiogenesis
Right arrow Cell biology/structural biology
Right arrow Cell signalling/signal transduction
(Circulation Research. 2007;100:1308.)
© 2007 American Heart Association, Inc.


Molecular Medicine

Interaction of {alpha}9ß1 Integrin With Thrombospondin-1 Promotes Angiogenesis

Izabela Staniszewska, Shachi Zaveri, Luis Del Valle, Isabela Oliva, Vicki L. Rothman, Sidney E. Croul, David D. Roberts, Deane F. Mosher, George P. Tuszynski, Cezary Marcinkiewicz

From the From the Department of Neuroscience (I.Z., S.Z., L.D.V., I.O., V.L.R., G.P.T., C.M.), Center for Neurovirology, Temple University, School of Medicine, Philadelphia, Pa; Department of Medicine and Pathobiology (S.E.C.) University of Toronto, Canada; Laboratory of Pathology (D.D.R.), National Cancer Institute, NIH, Bethesda, Md; Department of Medicine (D.F.M.), University of Wisconsin-Madison.

Correspondence to Cezary Marcinkiewicz: Temple University, School of Medicine, Department of Neuroscience, Center for Neurovirology, 1900 N.12th Street, Philadelphia, PA 19122. E-mail cmarcink{at}temple.edu


*    Abstract
up arrowTop
*Abstract
down arrowIntroduction
down arrowMaterials and Methods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Thrombospondin-1 is a multifunctional protein interacting with several cell surface receptors including integrins. We found that it is a ligand for {alpha}9ß1 integrin, and has an integrin binding site within its N-terminal domain (NoC1). Interaction of thrombospondin-1 and its recombinant NoC1 domain with {alpha}9ß1 integrin was confirmed in ELISA and cell adhesion assays. Binding of NoC1 to cells expressing {alpha}9ß1 integrin activated signaling proteins such as Erk1/2 and paxillin. Blocking of this integrin by monoclonal antibody and the met-leu-asp-disintegrin inhibited dermal human microvascular endothelial cell proliferation and NoC1-induced migration of these cells. Immunohistochemical studies revealed that {alpha}9ß1 is expressed on microvascular endothelium in several organs including skin, lung, heart and brain. NoC1 induced neovascularization in an experimental quail chorioallantoic membrane system and Matrigel plug formation assay in mice. This proangiogenic activity of NoC1 in vivo was inhibited by {alpha}9ß1 inhibitors. In summary, our results revealed that {alpha}9ß1 integrin expressed on microvascular endothelial cells interacts with thrombospondin-1, and this interaction is involved in modulation of angiogenesis.


Key Words: integrins • thrombospondin • angiogenesis


*    Introduction
up arrowTop
up arrowAbstract
*Introduction
down arrowMaterials and Methods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Integrins are a large family of cell surface receptors that are important in forming and physiological functioning of all human organs. These receptors in the active form are heterodimeric complexes pairing 2 {alpha} and ß subunits. Currently, 18 {alpha} and 8 ß subunits have been identified and they may be combined in a restricted manner to form at least 24 heterodimers.1 Despite having structural function, integrins also participate in signal transduction from "outside to inside" and "inside to outside" of the cell. This phenomenon is crucial for cell physiology including cytoskeleton reorganization resulting in shape change, adhesion and migration, regulation of cell proliferation, and the cell survival/death process.2,3 The {alpha} subunits of integrins may be composed of 1, or 2 polypeptide chains heavy and light, which are linked by disulfide bounds. The {alpha}9 integrin subunit belongs to the single polypeptide chain family and has been identified as an integrin heterodimer, only in association with the ß1 subunit.4 This {alpha}9ß1 integrin is widely distributed through out the human body and is expressed on many types of cells including epithelial cells, muscle cells, neutrophils, and endothelial cells.4,5

The wide distribution of {alpha}9ß1 integrin could be related to its high cross-reactivity with a variety of endogenous ligands such as VCAM-1,6 tenascin-C,7 and osteopontin.8 Binding of snake venom disintegrins by the MLD motif to this integrin was also described.6,9 Most recently, {alpha}9ß1 integrin was reported to directly bind VEGF-C and VEGF-D confirming its significance in the development of the lymphatic system.5

Thrombospondin-1 (TSP-1), the most abundant member of the five member thrombospondin gene family, is composed of 3 identical disulfide-linked chains each consisting of 1,152 amino acids composed primarily of domains consisting of repeating homologous amino acid sequences. These distinct domains interact with different cell surface receptors and mediate a variety of cellular processes including cell attachment, migration, proliferation, and differentiation.10 The various integrins belong to a separate group of TSP-1 receptors.11 The first discovered and the most characterized is {alpha}vß3 integrin that binds to the RGD-containing type 3 repeat of the TSP-1 molecule,12 whereas the N-terminus of TSP-1, also called NoC1, was shown to bind the 3 ß1 integrins {alpha}3ß1, {alpha}4ß1, and {alpha}6ß1.11 Within this domain, {alpha}3ß1 integrin recognizes the NVR motif, {alpha}4ß1 integrin binds to the LDVP sequence, and interaction of {alpha}6ß1 integrin is sensitive to Glu90mutation.

TSP-1 has been reported as a modulator of angiogenesis.13 This protein, and some of its fragments, directly inhibit endothelial cell functions such as proliferation and migration, but fragments containing the N-module stimulate the same endothelial cells responses.14–16 In some cases treatment of cultured endothelial cells by these compounds induces apoptosis.17 Integrins expressed on endothelial cells appear to be important components for their interaction with TSP-1 in this process.15,16 In light of our finding that {alpha}9ß1 integrin is highly expressed on the endothelium of microvasculature of variety of tissues, we assume that it may participate in angiogenesis. In this article we present data to clarify the role of this integrin in neovascularization as a receptor for TSP-1. Particularly, the interaction of {alpha}9ß1 with NoC1 may explain TSP-1’s variable ability to induce angiogenesis.


*    Materials and Methods
up arrowTop
up arrowAbstract
up arrowIntroduction
*Materials and Methods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Integrin Ligands and Snake Venom Disintegrins
TSP-1 was purified from human blood.18 The recombinant NoC1 domain (residues 1 to 356) and delNo1 domain (residues 312 to 1170) were prepared as described previously.19,20 Snake venom disintegrins VLO5 and VLO4 were purified from Vipera lebetina obtusa venom (Latoxan, Valence, France) using 2 steps of reverse phase HPLC as described previously.9

Cell Lines
{alpha}9- and mock-transfected SW480 cells were provided by Dr D. Sheppard (University of California, San Francisco, Calif). {alpha}9K562 cells were provided by Dr P. Weinreb (Biogen Inc.) and control K562 cells were purchased from ATCC (Manassas, Va). Primary adult dermal human microvascular endothelial cells (dHMVEC), pulmonary human microvascular endothelial cells (pHMVEC) and cardiac human microvascular endothelial cells (cHMVEC) were purchased from Cambrex. Primary brain human microvascular endothelial cells (bHMVEC) were purified from fetal tissue according to procedure described earlier.21 Passages of primary endothelial cells between 5 and 8 were used for experiments.

Cell Adhesion Studies
Adhesion studies of cultured cells labeled with 5-chloromethyl fluorescein diacetate (CMFDA) (Invitrogen) were performed as described previously.9

ELISA Assay
ELISA assay with purified {alpha}9ß1 and {alpha}5ß1 integrins was performed according to procedure described in the data supplement available at http://circres.ahajournals.org.

Cell Migration and Proliferation Assays
Cell migration assay is described in Data Supplement.

Cell proliferation assay was performed using BrdUrd kit according to the manufacturer’s instruction (Roche, Mannheim, Germany).

Angiogenesis In Vivo
The shell less Japanese quail CAM (chorioallantoic membrane) angiogenesis assay was performed according to the procedure described previously.22 Mouse Matrigel plug assay and immunohistochemistry are described in online data supplement available at http://circres.ahajournals.org.


*    Results
up arrowTop
up arrowAbstract
up arrowIntroduction
up arrowMaterials and Methods
*Results
down arrowDiscussion
down arrowReferences
 
Interaction of TSP-1 and NoC1 Domain With {alpha}9ß1 Integrin in an Adhesion Assay
The experiments showing {alpha}9ß1 integrin-dependent cell adhesion to TSP-1 were performed using 2 cell lines transfected with human {alpha}9 integrin. {alpha}9SW480 (Figure 1A) and {alpha}9K562 (Figure 1B) cell lines revealed potent, dose-dependent adhesive properties to immobilized TSP-1 and to the N-terminal recombinant fragment of this molecule, NoC1. However, {alpha}9K562 required the addition of 1 mmol/L Mn2+ (Figure 1B) to enable adhesion, whereas the use of 1 mmol/L Ca2+ and Mg2+, as in the case of {alpha}9SW480 cells, showed no adhesion of {alpha}9K562 cells to these ligands. Both control cells, which do not express {alpha}9ß1 integrin, displayed no adhesion to NoC1 domain, although mock-SW480 cells adhered to immobilized TSP-1, but to a significantly lower extent than {alpha}9SW480 cells. Specific interaction of {alpha}9ß1 with TSP-1 and NoC1 was confirmed by competitive cell adhesion studies (supplemental Figure I in the online data supplement). Adhesion of both {alpha}9SW480 and {alpha}9K562 cell lines was completely blocked by Y9A2 and VLO5, whereas inhibitors of the {alpha}5ß1 integrin were not active. Interestingly, partial inhibition was observed for adhesion of {alpha}9SW480 cells to TSP-1 in the presence of Y9A2 and Lia1/2. This may suggest that other than ß1-integrins are involved in this binding. This is probably {alpha}vß3 integrin, which has a binding site in the type three repeats of the TSP-1 molecule, and was detected by us on SW480 cells.


Figure 1
View larger version (9K):
[in this window]
[in a new window]

 
Figure 1. Interaction of {alpha}9ß1 integrin with TSP-1 and recombinant NoC1 domain in adhesion assay. A, Adhesion of {alpha}9SW480 (filled symbols) and mock-SW480 (open symbols) transfected cells to immobilized TSP-1 (circles) and NoC1 (triangles). TSP-1 or NoC1 were immobilized on the 96-well microplate in PBS by overnight incubation at 4°C in PBS. Experiment was performed as described in Fig. S1. B, Adhesion of {alpha}9K562 cells (filled symbols) and control nontransfected K562 cells (open symbols) to immobilized TSP-1 (circles) and NoC1 (triangles). The experiment was performed according to the same conditions as SW480 cells, however, adhesion buffer contained 1 mmol/L of Mn2+ in TBS. The error bars represent the standard deviation from three independent experiments.

Binding of {alpha}9ß1 Integrin With TSP-1 and NoC1 in a Solid Phase Assay
{alpha}9ß1 integrin was purified from {alpha}9SW480 and {alpha}9K562 cells using VLO5 affinity chromatography. These 2 {alpha}9ß1 integrin preparations were tested for their interaction with TSP-1 and the recombinant NoC1 domain in an ELISA assay. In this assay, {alpha}9ß1 integrin was immobilized on a plate and bound TSP-1 or it recombinant fragment NoC1 was detected by an immune reaction with a polyclonal antibody against TSP-1. Binding of TSP-1 and NoC1 was observed for {alpha}9ß1 isolated either from {alpha}9SW480 (Figure 2A) or from {alpha}9K562 cells (Figure 2B). However, similar to our adhesion studies, purified {alpha}9ß1 from {alpha}9K562 cells required stimulation with 1 mmol/L Mn2+ to bind its ligands. Interestingly, adding other integrin activators such as PMA or monoclonal antibody (TS2/16) had no stimulatory effect on of {alpha}9ß1 interaction with its ligands in ELISA, as well as in the adhesion assay.


Figure 2
View larger version (10K):
[in this window]
[in a new window]

 
Figure 2. Binding of TSP-1 and NoC1 to purified {alpha}9ß1 and {alpha}5ß1 integrins in ELISA assay. {alpha}9ß1 integrin was isolated from {alpha}9SW480 cells (A), or from {alpha}9K562 cells (B), whereas {alpha}5ß1 (C), was purchased from Chemicon. Experiment was performed as described in the online data supplement. Interaction of integrins with TSP-1 is presented as filled circles and NoC1 with open circles. The error bars represent the standard deviation from 3 independent experiments.

Binding of NoC1 Activates {alpha}9ß1 Integrin-Dependent Pathway of Cell Signaling
Previously published studies revealed that {alpha}9ß1 integrin is involved in activation of a signaling pathway that includes phosphorylation of MAPK Erk1/2 and paxillin.5,23 Figure 3A shows that adhesion of {alpha}9SW480 cells to immobilized NoC1 increased phosphorylation of Erk1/2 when compared with control mock-SW480 cells. The activation of this MAPK pathway was inhibited by specific blockers of {alpha}9ß1 integrin such as a monoclonal antibody Y9A2 and the heterodimeric disintegrin VLO5. Phosphorylation of paxillin was also inhibited by Y9A2 and VLO5 (Figure 3B).


Figure 3
View larger version (9K):
[in this window]
[in a new window]

 
Figure 3. {alpha}9ß1 integrin-dependent signaling induced by NoC1. A, {alpha}9SW480 (lines 2 to 5) or mock-SW480 (line 1) cells were allowed to adhere to immobilized NoC1 (20 µg/mL) or BSA (line 2) for 30 minutes in the absence or presence of {alpha}9ß1 integrin inhibitors, Y9A2 (10 µg/mL) or VLO5 (10 µg/mL). Cell lysates were obtained and equal amounts of protein were separated under reducing conditions on 10% SDS-PAGE. The proteins from the gel were electro-transferred into PVDF membrane and incubated with primary anti-phospho-Erk1/2 (Thr202/Tyr204) and anti-Erk1/2 polyclonal antibodies. The bands were visualized using chemiluminescent Western detection kit. The numbers above the bands represent value of average pixels, reflecting intensity of bands, digitalized using Un-Scan-It software. B, Paxillin phosphorylation in {alpha}9SW480 cells induced by NoC1. In the experiment the anti-phospho-paxillin (Tyr31) and anti-paxillin polyclonal antibodies were used.

Effect of Inhibition of {alpha}9ß1 Integrin on Adhesion to NoC1 and Migration of dHMVEC
dHMVEC express variety of integrins. The adhesion microarray with immobilized monoclonal antibodies (Figure 4A) revealed that dHMVEC expressed commonly known integrins for microvascular endothelial cells isolated from other organs such as lung (pHMVEC), heart (cHMVEC), or brain (bHMVEC). The main difference was in the expression of {alpha}9ß1 that is only present on dHMVEC. Moreover, interaction of this integrin with NoC1 was confirmed in cell adhesion competitive experiment (supplemental Figure II in the online data supplement). In this assay only inhibitors of {alpha}9ß1, Y9A2 and VLO5 blocked adhesion of dHMVEC to this TSP-1 fragment. Immunostaing of paraffin embedded tissues from related organs with an anti-{alpha}9 antibody, showed significant expression of this integrin on blood capillaries endothelium (Figure 4B). This suggests that during isolation and culturing of the majority of microvascular endothelial cells {alpha}9ß1 integrin disappears from their surface, except those isolated from skin. Based on this, we tested the effect of TSP-1 and its NoC1 fragments on proangiogenic activity of dHMVEC including migration and proliferation. The migratory activity of dHMVEC through the membrane containing the immobilized NoC1 domain of TSP-1 was dependent on {alpha}9ß1 integrin (Figure 5). Addition of {alpha}9ß1 inhibitors, such as the monoclonal antibody Y9A2 or MLD-disintegrin VLO5, completely abolished migration of dHMVEC, even below random migration. On the other hand, specific inhibitors for {alpha}5ß1 integrin, also found on endothelial cells, such as monoclonal antibody SAM-1 and disintegrin VLO4, showed no inhibitory effect on migration. The inhibitory activity of these compounds approached the maximal point after 2 hours, although control samples without inhibitors maximal migration levels obtained after 4 hours of incubation.


Figure 4
View larger version (61K):
[in this window]
[in a new window]

 
Figure 4. Detection of {alpha}9ß1 integrin on endothelial cells from different tissues. A, Monoclonal antibody adhesion array of HMVEC isolated from skin, lung, heart and brain. Adhesion assay was performed to immobilized monoclonal antibodies (5 µg/mL PBS) as described in supplemental Figure I. B, Images (x400 magnification) of immunohistochemistry on paraffin sections of human skin, lung, heart and brain with anti-{alpha}9 polyclonal antibody demonstrated strong labeling of endothelial cells.


Figure 5
View larger version (10K):
[in this window]
[in a new window]

 
Figure 5. Effect of VLO5 and Y9A2 on transmigration of dHMVEC through the membrane with immobilized NoC1. Cells labeled with calceine were preincubated with Y9A2 (10 µg/mL) (open triangles), VLO5 (1 µmol/L) (filled triangles), SAM-1 (anti-{alpha}5) (10 µg/mL) (open squares) and VLO4 (1 µmol/L) (filled squares) for 30 minutes in EBM-2 at room temperature and applied to the upper chamber, whereas to the lower chamber chemoattractant (2% FBS) was applied. Random migration is indicated as open circles, whereas control migration of cells without inhibitors is presented as filled circles. Error bars represent S.D. from triplicated experiments.

Effect of Inhibition of {alpha}9ß1 Integrin on Proliferation of Endothelial Cells
The data presented in Figure 6 show that VLO5 is the most potent inhibitor of proliferation of dHMVEC in a BrdUrd assay comparable to that observed with vincristin used as a positive control. However, in bHMVEC VLO5 blocked proliferation to the much lower extent, although still significant in comparison with the control. TSP-1 as well as Y9A2 monoclonal antibody significantly inhibited proliferation of dHMVEC only under low concentration of FBS (0.1%). Higher concentrations of FBS protected the cells against TSP-1–induced inhibition of proliferation. Y9A2 also had no significant effect on proliferation of dHMVEC in the presence of 2% FBS and bHMVEC under both concentrations of FBS. The effect of NoC1 was not observed for dHMVEC and bHMVEC.


Figure 6
View larger version (9K):
[in this window]
[in a new window]

 
Figure 6. Effect of antagonizing {alpha}9ß1 integrin on proliferation of dHMVEC and bHMVEC. Cells were grown on 96-well plate up to 70% confluence and then treated for 48 hours with Y9A2 (10 µg/mL), VLO5 (1 µmol/L), NoC1 (10 µg/mL), TSP-1 (10 µg/mL) and vincristine (50 µg/mL) in the EBM-2 medium containing 2% FBS (filled bars) or 0.1% FBS (open bars). BrdUrd color development assay was performed according to manufacturer’s instruction (Roche). Error bars represent S.D. from triplicated experiments. *, significant difference in comparison to control (P<0.05).

Involvement of {alpha}9ß1 Integrin in NoC1 Induced Angiogenesis In Vivo
Recombinant NoC1 domain increased the vascularization ratio in the quail CAM to the same range as growth factors such as VEGF and bFGF (supplemental Figure III). delNo1 domain, which is complementary to the rest of TSP-1 molecule, as well as TSP-1 isolated from human blood, had no effect on CAM endogenous angiogenesis (supplemental Figure IIId and IIIf). The NoC1-induced angiogenesis was potently inhibited by Y9A2 and VLO5 (Figure 7). On the other hand, the effect of Y9A2 and VLO5 on endogenously-induced embryonic vascularization level was not observed (supplemental Figure IIIb and IIIc). The combination of NoC1 and delNo1 domains showed a significant decrease in angiogenesis as opposed to NoC1 alone. However, this level was still significantly higher than native TSP-1 isolated from blood, suggesting a different conformational organization of recombinant domains and native TSP-1. The presence of {alpha}9ß1 integrin in quail CAM vasculature was confirmed by a Western blot of CAM lysate (supplemental Figure IVA) and fluoro-immunostaing (supplemental Figure IVB). This integrin was also detected in CAM lysate by immunoprecipitation using Y9A2 as a primary antibody (supplemental Figure IVC).


Figure 7
View larger version (14K):
[in this window]
[in a new window]

 
Figure 7. Effect of growth factors and {alpha}9ß1 integrin antagonists on NoC1-induced angiogenesis in CAM assay of quail system. Experiments were performed as described in supplemental Figure III. Tested compounds include: control, PBS (a); 5 µg NoC1 (b); 5 µg NoC1+5 µg Y9A2 (c); 5 µg NoC1+5 µg VLO5 (d); 5 µg NoC1+15 µg delNoC1 (e); 5 µg NoC1+0.5 µg VEGF (f); 5 µg NoC1+0.5 µg bFGF (g). Representative binary images of mid-arterial end points fragments of CAMs dissected from embryos are presented on the left panels and graphic values of angiogenesis index as a fractal dimension (Df) are presented on the right plots. Error bars represent standard deviation from analysis of 8 to 10 embryos per group. *, significant difference in comparison to NoC1-treted (b) group (P<0.05).

NoC1 also potently induced the vascularization level in Matrigel plugs implanted into mice (Figure 8). The number of vessels significantly increased following injection of Matrigel in the presence of NoC1 compare to the control. However, this effect was abolished in the presence of disintegrin VLO5, which blocks {alpha}9ß1 integrin. Using immunohistochemistry, expression of this integrin was detected on endothelial cells that formed the vasculature in Matrigel (Figure 8A).


Figure 8
View larger version (77K):
[in this window]
[in a new window]

 
Figure 8. Effect of VLO5 on NoC1-induced angiogenesis in Matrigel plug assay in mouse model. A, Images (x400 magnification) of paraffin sections of Matrigel plugs stained with H&E, anti-von Willebrand factor and anti-{alpha}9 polyclonal antibodies. B, Comparison of the number of vessels counted by microscopic field. *, significant difference between control and treated groups (P<0.001); **, significant difference between NoC1 and NoC1 + VLO5 (P<0.001).


*    Discussion
up arrowTop
up arrowAbstract
up arrowIntroduction
up arrowMaterials and Methods
up arrowResults
*Discussion
down arrowReferences
 
In the presented study we found that {alpha}9ß1 integrin is a receptor for TSP-1 and the binding site is localized within the NoC1 domain of this extracellular matrix protein. TSP-1 has been characterized as a ligand for many cell surface receptors including integrins, whereas {alpha}9ß1 integrin is a multifunctional receptor interacting with several ligands including ECM proteins, VCAM-1, and ADAMs family members. However, there are no reports showing an interaction of TSP-1 with this integrin, and our work is the first to characterize the biological consequence of this interaction. Reactivity of TSP-1 with {alpha}9ß1 integrin was confirmed in cell adhesion assays using cell lines expressing this integrin and in an ELISA with purified receptor. TSP-1 or the NoC1 domain bound very efficiently to recombinant {alpha}9ß1 integrin expressed on the SW480 cell line in an adhesion assay, and this binding was inhibited by specific inhibitors of this integrin such as Y9A2 and VLO5.

The binding of TSP-1 and NoC1 to {alpha}9SW480 cells and to {alpha}9ß1 integrin purified from these cells does not require any integrin activators including activating monoclonal antibodies (TS2/16) or PMA. The same {alpha}9ß1 integrin-dependent adhesive ability of TSP-1 and NoC1 to cells naturally expressing this integrin such as dHMVEC was also observed in the absence of integrin activators. These data suggest that under physiological conditions, interaction of {alpha}9ß1 integrin with TSP-1 does not require activation of the integrin. This finding may increase the physiological significance of the {alpha}9ß1/TSP-1 interaction, especially in light of other integrins such as {alpha}4ß1 and {alpha}6ß1 that have also been identified as TSP-1 receptors.16,24 However, these integrins required activation to bind efficiently to TSP-1. Alternatively, we used K562 cells transfected with {alpha}9 integrin subunits, and confirmed that {alpha}9ß1 integrin is a receptor for TSP-1. However, these cells showed no reactivity with TSP-1 or NoC1 under physiological concentrations of calcium and magnesium. Adhesion of {alpha}9K562 cells to these ligands was only observed in the presence of Mn2+ cations. Many investigators consider Mn2+ as a cell activating compound or "cross-linker" enhancer of ligands to certain integrins.25,26 However, in our case the lack of binding in the absence of Mn2+ is probably because of improper conformation of {alpha}9ß1 integrin after recombinant expression of {alpha}9 subunit on the K562 cell surface. Any other activatory compounds such as TS2/16 or PMA have no effect on increasing the binding of TSP-1 to {alpha}9K562 cells (data not shown). Moreover, the integrin purified from these cells on a VLO5-affinity column also required the presence of Mn2+ to bind to TSP-1 or NoC1 in ELISA assay. These cations can probably convert the conformation of {alpha}9ß1 integrin’s binding pocket making it accessible to a large molecule such as TSP-1 in a manner independent of any activation "out-in-out" pathways of cell signaling. Other ligands for {alpha}9ß1 integrin such as VCAM-1 also required Mn2+ for binding to {alpha}9K562 cells.27 Interestingly, VLO5 interacts with {alpha}9K562 cells in the absence of Mn2+ (data not shown), probably because of the high binding affinity and low size of the molecule that may enter the integrin’s binding pocket without any conformational preferences.

Based on the high homology between {alpha}4 and {alpha}9 subunits of integrins, we anticipated that {alpha}4ß1 and {alpha}9ß1 integrins may have the same binding sites on TSP-1 within the NoC1 domain. However, our experiments did not clearly confirm this hypothesis. The previously identified region of binding of {alpha}4ß1 integrin that contains the AELDVP sequence showed D162 as the most sensitive amino acid for the ligand/receptor interaction.28 We performed series of experiments with mutated D162A NoC1 domain and synthetic peptides AELDVP or AELAVP in adhesion and ELISA assays (data not shown). However, our results were not conclusive and could not confirm that this region of TSP-1 is a binding site for {alpha}9ß1 integrin. It is common that despite having high homology and sharing some common ligands like VCAM-1, these 2 integrins also have separate binding activities to specific ligands such as tenascin-C for {alpha}9ß129 and CS-1 fibronectin fragment for {alpha}4ß1.30 Further studies are required to identify this binding site.

The interaction of {alpha}9SW480 cells with the ligands for {alpha}9ß1 without any activator of integrins suggested that this transfected cell line contains integrins in the physiologically functional stage. Indeed, the previously performed studies showed that {alpha}9ß1 integrin ligands such as VCAM-1 or tenascin-C, following binding to these cells, activated signaling molecules important for cell physiology.5,23 Our results show that interaction of the NoC1 domain with these cells led to phosphorylation of Erk1/2 and paxillin, indicating that TSP-1 is a ligand for {alpha}9ß1 integrin and may modulate cell proliferation and motility. These 2 activities are characteristic for endothelial cells during angiogenesis and {alpha}9ß1 may be considered as another integrin that plays a role in this neovascularization process. To evaluate the potential involvement of this integrin in angiogenesis, we performed a systematic characterization of the integrin content in primary capillary endothelial cells isolated from different organs. We designed an anti-integrin monoclonal antibody adhesion microarray for microvascular endothelial cells isolated from skin, lung, heart and brain, and we found that {alpha}9ß1 integrin is only expressed on primary dHMVEC. Surprisingly, this integrin was detected on the microvascular endothelial cells in all considered tissues by immunohistochemistry. This discrepancy may be because {alpha}9ß1 integrin is very sensitively anchored to the surface of endothelial cells, and during an isolation of cells, is degraded. Moreover, further cell culturing and passaging do not recover integrin expression, and pHMVEC, cHMVEC and bHMVEC show its absence on the cell surface. In spite of this, {alpha}9ß1 integrin is preserved on dHMVEC during isolation and culturing procedures. dHMVEC are invaluable to investigate interaction of TSP-1 and {alpha}9ß1 integrin in angiogenesis-related in vitro experiments. Although data received in these in vitro experiments directly characterizes skin angiogenesis, it would be appropriate to correlate them with other organs, because the in vivo presence of this integrin on endothelial cells is common and not related to tissue specificity.

The isolated microvascular endothelial cells express a variety of integrins and some of them like {alpha}vß3 (vitronectin receptor), or {alpha}1ß1 and {alpha}2ß1 (collagen receptors), were well described as regulators of cell proliferation and migration involved in angiogenesis.31,32 Interestingly, dHMVEC also contain {alpha}4ß1 integrin, but in comparison with {alpha}9ß1, expression of this integrin is very low. Adhesion of these primary endothelial cells to NoC1 was inhibited by Y9A2 and VLO5, whereas blocking monoclonal antibodies against other integrins, or the RGD-disintegrin VLO4 were not effective. These results confirmed previous data showing a lack of participation of {alpha}4ß1 integrin in adhesion of dHMVEC to NoC1, and this integrin trends to function rather for large vessel endothelial cells, such as human umbilical endothelial cells (HUVEC).16

TSP-1 has several binding sites for various integrins that may have overlapping functions in the regulation of angiogenesis. Previous reports suggested that {alpha}3ß1 and {alpha}4ß1 integrins expressed on endothelial cells are receptors for NoC1 domain that mediate the proangiogenic activity of TSP-1.15,16 Our results revealed that {alpha}9ß1 integrin plays a similar role as a receptor for TSP-1. Experiments performed with cultured primary dHMVEC as well as experiments in vivo indicated participation of {alpha}9ß1 integrin in NoC1-induced angiogenesis. Although proliferation of dHMVEC was not effected by NoC1, this TSP-1 domain was a potent enhancer of cell migration. The motility of endothelial cells was inhibited by specific {alpha}9ß1 integrin inhibitors. This observation suggests that the proangiogenic activity of NoC1 is related to increasing ability of endothelial cells to migrate in the surrounding matrix rather than to enhance cell proliferation. The participation of {alpha}9ß1 integrin in promotion of cell migration was observed before, although the mechanism and cell signaling that are involved in this process are still unknown.23 The inhibitory activity of anti-{alpha}9ß1 integrin compounds in dHMVEC proliferation suggests that antagonizing of this integrin may be important in cell survival/death process. Especially, potent antiproliferative effects of VLO5 in dHMVEC indicate that this disintegrin following binding to {alpha}9ß1, induces a signaling cascade affecting the cell cycle. Consistent with this idea is induction of apoptosis by VLO5 following binding to integrins. The involvement of {alpha}vß3 integrin in triggering pro-apoptotic signal through activation of caspase 8 pathway has been reported.33 Moreover, RGD-containing snake venom disintegrins were also reported as apoptosis inducers in endothelial cells.34 Although a very potent, anti-proliferative effect of VLO5 was observed for dHMVEC that express {alpha}9ß1 integrin, the much lower but still significant effect of this disintegrin is present for bHMVEC that do not express this integrin (Figure 6). This phenomenon may be explained by interaction of VLO5 with {alpha}4ß1 integrin,9 which is present on all endothelial cells. However, expression of {alpha}4ß1 is very low (Figure 4A) and in the absence of {alpha}9ß1 integrin the effect of VLO5 is less significant. In conclusion, the ability of VLO5 to block endothelial cell proliferation and in the consequence angiogenesis, appears to be highly dependent to its interaction with {alpha}9ß1 integrin. This conclusion is supported by our immunohistochemistry data that showed high expression of this integrin on endothelial cells from several organs.

The quail CAM and mouse Matrigel plug assays appear to be appropriate for investigation of involvement of {alpha}9ß1 integrin in angiogenesis in vivo because endothelial cells of both species highly express this integrin. NoC1 induced angiogenesis in these systems to a level comparable with growth factors such as VEGF and FGF. Simultaneous treatment of CAM embryos with NoC1 and these growth factors, showed no differences in the vascularization ratio (Figure 7). This finding differs from experiments performed in other avian CAM assay such as chicken,16 where NoC1 significantly increased the angiogenic response either alone or in the presence of growth factors and was inhibited by {alpha}4ß1 integrin antagonists. These discrepancies may be explained by differences in the avian species (quail versus chicken) used in experiments, as well as concentrations of compounds used for angiogenesis induction.

The blocking effect of {alpha}9ß1 integrin inhibitors on angiogenesis induced by NoC1 clearly indicates that this integrin present on blood capillaries significantly participates in the neovascularization process. In this respect, proangiogenic effects of TSP-1 appear to be directly correlated with activation of endothelial cells by interaction with cell membrane receptors such as {alpha}9ß1 integrin. The dual activity of TSP-1 in modulating angiogenesis has been reported in the past.35 However, angiogenesis induced by the 25 kDa proteolytic fragment of TSP-1 that binds heparin,36 as well as the entire TSP-1 molecule used in lower concentrations,37 is mediated by the upregulation of MMPs. We cannot exclude that this kind of indirect proangiogenic effect of NoC1/{alpha}9ß1 interaction may also occur, but the data presented in this article show that the predominant role of these molecules is for a promigratory effect on endothelial cells. Recently, the enzymatic cleavage of the TSP-1 molecule has been observed by ADAMTS1.38 Two peptides (36 kDa and 110 kDa) were identified as products of this degradation. The fragment with lower molecular weight corresponds to the NoC1 domain that may contribute to upregulation of angiogenesis. On the other hand, the larger fragment may participate in the inhibition of angiogenesis by interaction with {alpha}vß3 integrin. It is likely, that the conformation of the nondegraded TSP-1 exposes an active site only for negative regulation of angiogenesis, whereas the N-terminally located proangiogenic integrin binding sites are not accessible for endothelial cell receptors. Enzymatic digestion of this molecule that may occurs under certain pathophysiological conditions releases this N-terminal domain that may start to act as a positive regulator of angiogenesis.


*    Acknowledgments
 
Sources of Funding

This work was supported in part by NIH grants 5R01CA100145 (C.M.), R01CA88931 (G.P.T.), R01HL54462 (D.F.M.) and AHA grant 0230163N (C.M.).

Disclosures

None.


*    Footnotes
 
Original received July 12, 2006; resubmission received January 29, 2007; revised resubmission received March 23, 2007; accepted March 28, 2007.


*    References
up arrowTop
up arrowAbstract
up arrowIntroduction
up arrowMaterials and Methods
up arrowResults
up arrowDiscussion
*References
 

  1. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell. 2002; 110: 673–687.[CrossRef][Medline] [Order article via Infotrieve]
  2. Hood JD, Cheresh DA. Role of integrins in cell invasion and migration. Nat Rev Cancer. 2002; 2: 91–100.[CrossRef][Medline] [Order article via Infotrieve]
  3. Stupack DG, Cheresh DA. Get a ligand, get a life: integrins, signaling and cell survival. J Cell Sci. 2002; 115: 3729–3738.[Abstract/Free Full Text]
  4. Palmer EL, Ruegg C, Ferrando R, Pytela R, Sheppard D. Sequence and tissue distribution of the integrin {alpha} 9 subunit, a novel partner of ß 1 that is widely distributed in epithelia and muscle. J Cell Biol. 1993; 123: 1289–1297.[Abstract/Free Full Text]
  5. Vlahakis NE, Young BA, Atakilit A, Sheppard D. The lymphangiogenic vascular endothelial growth factors VEGF-C and -D are ligands for the integrin {alpha}9ß1. J Biol Chem. 2005; 280: 4544–4552.[Abstract/Free Full Text]
  6. Marcinkiewicz C, Taooka Y, Yokosaki Y, Calvete JJ, Marcinkiewicz MM, Lobb RR, Niewiarowski S, Sheppard D. Inhibitory effects of MLDG-containing heterodimeric disintegrins reveal distinct structural requirements for interaction of the integrin {alpha} 9ß 1 with VCAM-1, tenascin-C, and osteopontin. J Biol Chem. 2000; 275: 31930–31937.[Abstract/Free Full Text]
  7. Yokosaki Y, Matsuura N, Higashiyama S, Murakami I, Obara M, Yamakido M, Shigeto N, Chen J, Sheppard D. Identification of the ligand binding site for the integrin {alpha}9 ß1 in the third fibronectin type III repeat of tenascin-C. J Biol Chem. 1998; 273: 11423–11428.[Abstract/Free Full Text]
  8. Smith LL, Cheung H-K, Ling LE, Chen J, Sheppard D, Pytela R, Giachelli CM. Osteopontin N-terminal domain contains a cryptic adhesive sequence recognized by {alpha}9ß1 integrin. J Biol Chem. 1996; 271: 28485–28491.[Abstract/Free Full Text]
  9. Bazan-Socha S, Kisiel DG, Young B, Theakston RD, Calvete JJ, Sheppard D, Marcinkiewicz C. Structural requirements of MLD-containing disintegrins for functional interaction with {alpha} 4 ß 1 and {alpha} 9 ß1 integrins Biochemistry. 2004; 43: 1639–1647.[CrossRef][Medline] [Order article via Infotrieve]
  10. Sargiannidou I, Zhou J, Tuszynski GP. The role of thrombospondin-1 in tumor progression. Exp Biol Med. 2001; 226: 726–733.[Abstract/Free Full Text]
  11. Calzada MJ, Roberts DD. Novel integrin antagonists derived from thrombospondins. Curr Pharm Design. 2005; 11: 815–827.[CrossRef][Medline] [Order article via Infotrieve]
  12. Lawler J, Weinstein R, Hynes RO. Cell attachment to thrombospondin: the role of ARG-GLY-ASP, calcium, and integrin receptors. J Cell Biol. 1988; 107: 2351–2361.[Abstract/Free Full Text]
  13. Lawler J. Thrombospondin-1 as an endogenous inhibitor of angiogenesis and tumor growth. J Cell Mol Med. 2002; 6: 1–12.[Medline] [Order article via Infotrieve]
  14. Nikosia RF, Tuszynski GP. Matrix-bound thrombospondin promotes angiogenesis in vitro J Cell Biol. 1994; 124: 183–193.[Abstract/Free Full Text]
  15. Chandrasekaran L, He CZ, Al-Barazi H, Krutzsch HC, Iruela-Arispe ML, Roberts DD. Cell contact-dependent activation of {alpha}3ß1 integrin modulates endothelial cell responses to thrombospondin-1. Mol Biol Cell. 2000; 11: 2885–2900.[Abstract/Free Full Text]
  16. Calzada MJ, Zhou L, Sipes JM, Zhang J, Krutzsch HC, Iruela-Arispe ML, Annis DS, Mosher DF, Roberts DD. {alpha}4ß1 integrin mediates selective endothelial cell responses to thrombospondins 1 and 2 in vitro and modulates angiogenesis in vivo. Circ Res. 2004; 94: 462–470.[Abstract/Free Full Text]
  17. Guo N, Krutzsch HC, Inman JK, Roberts DD. Thrombospondin 1 and type I repeat peptides of thrombospondin 1 specifically induce apoptosis of endothelial cells. Cancer Res. 1997; 57: 1735–1742.[Abstract/Free Full Text]
  18. Tuszynki GP, Srivastava S, Switalska HI, Holt JC, Cierniewski CS, Niewiarowski S. The interaction of human platelet thrombospondin with fibrinogen. Thrombospondin purification and specificity of interaction. J Biol Chem. 1985; 260: 12240–12245.[Abstract/Free Full Text]
  19. Calzada MJ, Annis DS, Zeng B, Marcinkiewicz C, Banas B, Lawler J, Mosher DF, Roberts DD. Identification of novel ß1 integrin binding sites in the type 1 and type 2 repeats of thrombospondin-1. J Biol Chem. 2004; 279: 41734–41743.[Abstract/Free Full Text]
  20. Annis DS, Murphy-Ullrich JE, Mosher DF. Function-blocking antithrombospondin-1 monoclonal antibodies. J Thromb Haemost. 2006; 4: 459–468.[CrossRef][Medline] [Order article via Infotrieve]
  21. Abbott NJ, Hughes CC, Revest A, Greenwood J. Development and characterization of a rat brain capillary endothelial culture: towards an in vitro blood-brain barrier. J Cell Sci. 1992; 103: 23–37.[Abstract/Free Full Text]
  22. Parsons-Wingerter P, Lwai B, Yang MC, Elliot KE, Milaninia A, Redlitz A, Clark JI, Sage H. A novel assay of angiogenesis in the quail chorioallantoic membrane: stimulation by bFGF and inhibition by angiostatin according to fractal dimension and grid intersection. Microvasc Res. 1998; 55: 201–214.[CrossRef][Medline] [Order article via Infotrieve]
  23. Young BA, Taooka Y, Liu S, Askins KJ, Yokosaki Y, Thomas SM, Sheppard D. The cytoplasmic domain of the integrin {alpha}9 subunit requires the adaptor protein paxillin to inhibit cell spreading but promotes cell migration in a paxillin-independent manner. Mol Biol Cell. 2001; 12: 3214–3225.[Abstract/Free Full Text]
  24. Calzada MJ, Sipes JM, Krutzsch HC, Yurchenco PD, Annis DS, Mosher DF, Roberts DD. Recognition of the N-terminal modules of thrombospondin-1 and thrombospondin-2 by {alpha}6ß1 integrin. J Biol Chem. 2003; 278: 40679–40687.[Abstract/Free Full Text]
  25. Bazzoni G, Shih DT, Buck CA, Hemler ME. Monoclonal antibody 9EG7 defines a novel ß 1 integrin epitope induced by soluble ligand and manganese, but inhibited by calcium. J Biol Chem. 1995; 270: 25570–25577.[Abstract/Free Full Text]
  26. Byzova TV, Plow TF. Activation of {alpha}vß3 on vascular cells controls recognition of prothrombin. J Cell Biol. 1998; 143: 2081–2092.[Abstract/Free Full Text]
  27. Pepinsky BR, Mumford RA, Chen LL, Leone D, Amo SE, Riper GV, Whitty A, Dolinski B, Lobb RR, Dean DC, Chang LL, Raab CE, Si Q, Hagmann WK, Lingham RB. Comparative assessment of the ligand and metal ion binding properties of integrins {alpha}9ß1 and {alpha}4ß1. Biochemistry. 2002; 41: 7125–7141.[CrossRef][Medline] [Order article via Infotrieve]
  28. Li Z, Calzada MJ, Sipes JM, Cashel JA, Krutzsch HC, Annis DS, Mosher DF, Roberts DD. Interactions of thrombospondins with {alpha}4ß1 integrin and CD47 differentially modulate T cell behavior. J Cell Biol. 2002; 157: 509–519.[Abstract/Free Full Text]
  29. Yokosaki Y, Palmer EL, Prieto AL, Crossin KL, Bourdon MA, Pytela R, Sheppard D. The integrin {alpha}9ß1 mediates cell attachment to a non-RGD site in the third fibronectin type III repeat of tenascin. J Biol Chem. 1994; 269: 26691–26696.[Abstract/Free Full Text]
  30. Chan BM, Elices MJ, Murphy E, Hemler ME. Adhesion to vascular cell adhesion molecule 1 and fibronectin. Comparison of {alpha} 4 ß 1 (VLA-4) and {alpha} 4 ß 7 on the human B cell line JY. J Biol Chem. 1992; 267: 8366–8370.[Abstract/Free Full Text]
  31. Senger DR, Perruzzi CA, Streit M, Koteliansky VE, de Fougerolles AR, Detmar M. The {alpha}1ß1 and {alpha}2ß1 integrins provide critical support for vascular endothelial growth factor signaling, endothelial cell migration, and tumor angiogenesis. Am J Pathol. 2002; 160: 195–205.[Abstract/Free Full Text]
  32. Stupack DG, Cheresh DA. Integrins in angiogenesis Curr Top Dev Biol. 2004; 64: 207–238.[CrossRef][Medline] [Order article via Infotrieve]
  33. Stupack DG, Cheresh DA. Apoptotic clues from the extracellular matrix: regulators of angiogenesis. Oncogene. 2003; 22: 9022–9029.[CrossRef][Medline] [Order article via Infotrieve]
  34. Yeh CH, Peng HC, Huang TF. Accutin, a new disintegrin, inhibits angiogenesis in vitro and in vivo by acting as integrin {alpha}vß3 antagonist and inducing apoptosis. Blood. 1998; 92: 3268–3276.[Abstract/Free Full Text]
  35. Tolsma SS, Volpert OV, Good DJ, Frazier WA, Polverini PJ, Bouck N. Peptides derived from two separate domains of the matrix protein thrombospondin-1 have anti-angiogenic activity. J Cell Biol. 1993; 122: 497–511.[Abstract/Free Full Text]
  36. Taraboletti G, Morbidelli L, Donnini S, Parenti A, Granger HJ, Giavazzi R, Ziche M. The heparin binding 25 kDa fragment of thrombospondin-1 promotes angiogenesis and modulates gelatinase and TIMP-2 production in endothelial cells. FASEB J. 2000; 14: 1674–1676.[Free Full Text]
  37. Qian X, Wang TN, Rothman VL, Nikosia RF, Tuszynski GP. Thrombospondin-1 modulates angiogenesis in vivo by up-regulation of matrix metalloproteinase-9 in endothelial cells. Exp Cell Res. 1997; 235: 403–412.[CrossRef][Medline] [Order article via Infotrieve]
  38. Lee NV, Sato M, Annis DS, Loo JA, Wu L, Mosher DF, Iruela-Arispe ML. ADAMTS1 mediates the release of antiangiogenic polypeptides from TSP1 and 2. EMBO J. 2006; 25: 5270–5283.[CrossRef][Medline] [Order article via Infotrieve]



This article has been cited by other articles:


Home page
Physiol. Rev.Home page
A. P. Lillis, L. B. Van Duyn, J. E. Murphy-Ullrich, and D. K. Strickland
LDL Receptor-Related Protein 1: Unique Tissue-Specific Functions Revealed by Selective Gene Knockout Studies
Physiol Rev, July 1, 2008; 88(3): 887 - 918.
[Abstract] [Full Text] [PDF]


Home page
J. Biol. Chem.Home page
K. Tan, M. Duquette, J.-h. Liu, K. Shanmugasundaram, A. Joachimiak, J. T. Gallagher, A. C. Rigby, J.-h. Wang, and J. Lawler
Heparin-induced cis- and trans-Dimerization Modes of the Thrombospondin-1 N-terminal Domain
J. Biol. Chem., February 15, 2008; 283(7): 3932 - 3941.
[Abstract] [Full Text] [PDF]


Home page
J. Cell Sci.Home page
I. Staniszewska, I. K. Sariyer, S. Lecht, M. C. Brown, E. M. Walsh, G. P. Tuszynski, M. Safak, P. Lazarovici, and C. Marcinkiewicz
Integrin {alpha}9{beta}1 is a receptor for nerve growth factor and other neurotrophins
J. Cell Sci., February 15, 2008; 121(4): 504 - 513.
[Abstract] [Full Text] [PDF]


Home page
J. Biol. Chem.Home page
A. V. Shinde, C. Bystroff, C. Wang, M. G. Vogelezang, P. A. Vincent, R. O. Hynes, and L. Van De Water
Identification of the Peptide Sequences within the EIIIA (EDA) Segment of Fibronectin That Mediate Integrin {alpha}9{beta}1-dependent Cellular Activities
J. Biol. Chem., February 1, 2008; 283(5): 2858 - 2870.
[Abstract] [Full Text] [PDF]


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
100/9/1308    most recent
01.RES.0000266662.98355.66v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Staniszewska, I.
Right arrow Articles by Marcinkiewicz, C.