Activated Microvessels Express Vascular Endothelial Growth Factor and Integrin α v β 3 During Focal Cerebral Ischemia

Abstract

Both vascular endothelial growth factor (VEGF) and integrin α_vβ₃ play roles in angiogenesis. In noncerebral vascular systems, VEGF can induce endothelial integrin α_vβ₃ expression. However, it is unknown whether VEGF, like integrin α_vβ₃, appears in the initial response of microvessels to focal brain ischemia. Their coordinate expression in microvessels of the basal ganglia after middle cerebral artery occlusion (MCAO) in the nonhuman primate model was examined quantitatively. Cells incorporating deoxyuridine triphosphate (dUTP⁺) by the polymerase I reaction at 1 hour (n = 3), 2 hours (n = 3), and 7 days (n = 4) after MCAO defined the ischemic core (Ic) and peripheral regions. Both VEGF and integrin α_vβ₃ were expressed by activated noncapillary (7.5- to 30.0-μm diameter) microvessels in the Ic region at 1 and 2 hours after MCAO. At 7 days after MCAO, the number of VEGF+, integrin α_vβ₃⁺, or proliferating cell nuclear antigen-positive microvessels had decreased within the Ic region. The expressions of VEGF, integrin α_vβ₃, and proliferating cell nuclear antigen were highly correlated on the same microvessels using hierarchical log-linear statistical models. Also, VEGF and subunit α_v messenger ribonucleic acids were coexpressed on selected microvessels. Here, noncapillary microvessels are activated specifically early during a focal cerebral ischemic insult and rapidly express VEGF and integrin α_vβ₃ together.

Keywords

Cerebral ischemia Microvessel Angiogenesis Vascular endothelial growth factor

The appearance of new blood vessels (angiogenesis) is a relatively late response of the brain to focal ischemia (Krupinski et al., 1994; Chuaqui and Tapia, 1993). The signals, their timing, and sequence of presentation for angiogenesis after the initial ischemic events have not been established (Okada et al., 1996). The appearance of leukocyte adhesion receptors on microvascular endothelium (Okada et al., 1994; Haring et al., 1996) and changes in microvascular integrin expression (Wagner et al., 1997) within the first hours after middle cerebral artery occlusion (MCAO) suggest that signals for microvascular cell activation and angiogenesis may be generated early during ischemia.

Vascular endothelial growth factor (VEGF), also known as vascular permeability factor, has been implicated in endothelial cell proliferation, permeability, and angiogenesis (Dvorak et al., 1995; Senger, 1996). Expression of VEGF and its receptors, flt-1 and KDR, has been observed in developing organs during embryogenesis and in several pathologic conditions, including ischemic injury of the brain (Kovacs et al., 1996; Hayashi et al., 1997) and the myocardium (Li et al. 1996), wound healing (Brown et al., 1992), and in several inflammatory conditions (Detmar et al., 1994; Fava et al., 1994). In vitro studies show that VEGF stimulates the expression of integrin α_vβ₃ in microvascular endothelial cells in addition to tissue factor, osteopontin, and plasminogen activator expression (Clauss et al., 1996; Zucker et al., 1998; Pepper et al., 1991; Senger et al., 1996). In the rodent, VEGF antigen and messenger ribonucleic acid (mRNA) appear from 5 to 24 hours after MCAO, late after the onset of focal ischemia (Kovacs et al., 1996; Hayashi et al., 1997).

Integrin α_vβ₃, a cellular receptor for several Arg-Gly-Asp (RGD)–containing ligands, plays a significant role in the processes of angiogenesis (Brooks et al., 1994). Among the variety of ligands recognized by integrin α_vβ₃ are fibrin(ogen), von Willebrand factor, vitronectin, thrombospondin, osteopontin, and the extracellular matrix proteins laminin and collagen (Cheresh, 1993). In the vasculature, integrin α_vβ₃ is expressed by endothelial cells, smooth muscle cells (SMC), activated leukocytes, and macrophages and has been shown to mediate tumor angiogenesis. Vascular endothelial cell migration involves the ligation of integrin α_vβ₃ with vitronectin, whereas SMC migration involves the interactions of integrin α_vβ₃ with osteopontin and vitronectin (Leavesley et al., 1993). It has been demonstrated that integrin α_vβ₃ promotes adhesion-dependent survival in angiogenic blood vessels (Brooks et al., 1994). Okada et al. demonstrated that integrin α_vβ₃ (but not integrin α_vβ₅, the receptor for vitronectin) appears on noncapillary microvessels in the basal ganglia within 2 hours of MCAO (Okada et al., 1996). Its appearance was significantly associated with fibrin (one of its ligands) deposited within the microvessel lumina at all times after MCAO. Whether this significant association results from increased microvascular permeability, endothelial cell or SMC reactivity, or from separate unrelated events is important for understanding the fate of cerebral microvessels and neurons after MCAO (Okada et al., 1994; Okada et al., 1996; Tagaya et al., 1997).

The hypothesis tested by this study is that microvascular integrin α_vβ₃ expression is increased together with the appearance of VEGF and evidence of endothelial cell and SMC activation within the first moments after the onset of experimental focal brain ischemia. Furthermore, these changes are related to neuron injury. Here, to test the null hypothesis of no co-localization, the number of microvessels expressing proliferating cell nuclear antigen (PCNA), VEGF, and integrin α_vβ₃ within the ischemic and nonischemic regions were tabulated for each subject after MCAO (at 1 hour, 2 hours, and 7 days). Log-linear models then were fitted to pairwise frequencies in multi way tables to determine the significance of any relation among PCNA, VEGF, and α_vβ₃ expression within the subjects. This study is the first to demonstrate that microvascular expression of integrin α_vβ₃ is related to VEGF up-regulation in reactive cerebral microvasculature displaying PCNA in the early moments of focal cerebral ischemia. It also provides evidence of an ordered temporal and topographic connection between microvascular activation and neuron injury during early MCAO.

MATERIALS AND METHODS

Experimental focal cerebral ischemia

Brain tissues from 13 adolescent male baboons (Papio anubis/cynocephalus) were used for this study. All animals were neurologically normal before MCAO and apparently free of infections or inflammation during the experiments. All procedures in this study were approved by the Institutional Animal Research Committee and were performed in accordance with the standards published by the National Research Council (Guide for the Care and Use of Laboratory Animals) and the U.S. Department of Agriculture Animal Welfare Act. In compliance with these standards, efforts were made to ensure that the subjects were free of pain or discomfort. The principal investigator, veterinarians, and primate handling staff were present for all procedures.

In the following experiments, three cohorts of subjects underwent MCAO for 1 hour (n = 3), 2 hours (n = 3), or 7 days (n = 4), compared to a control group (n = 3), which underwent no procedure. The 7-day cohort comprised subjects that underwent MCAO during surgical implantation and displayed persistent unilateral neurologic defects thereafter. Preparation of the awake baboon MCAO stroke model has been described in detail elsewhere (del Zoppo et al., 1986; del Zoppo et al., 1991). In brief, under general anesthesia with isoflurane using a transorbital approach, an inflatable balloon catheter assembly was placed around the MCA proximal to the origin of the lenticulostriate arteries. The MCAO was achieved by inflating the balloon in the awake subject. All experiments were terminated under thiopental sodium anesthesia by transcardiac perfusion with an isosmotic perfusion fluid containing heparin (2000 IU/L), sodium nitroprusside (6.7 μmol/L), and bovine serum albumin (25 g/L). Tissue blocks (1.0 × 1.0 × 0.5 cm) from symmetrically located sites of basal ganglia were excised. Alternate blocks either were embedded in Tissue-Tek OCT compound (Miles, Inc., Elkart, IN, U.S.A.), frozen in isopentane/dry ice, and stored at -80°C until use or were fixed with 2% paraformaldehyde (PFA) for 24 hours and embedded in paraffin for the immunohistochemical and in situ hybridization studies.

Antibodies

Poly clonal antibodies against VEGF (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, U.S.A.) and integrins containing subunit α_v (integrin α_vβ₃ and α_vβ₅) (Chemicon International, Inc., Temecula, CA, U.S.A.), and monoclonal antibodies against PCNA (PC-10) (Sigma BioSciences, St. Louis, MO, U.S.A.) and integrin α_vβ₃ (LM609) (Chemicon Interational, Inc.), were used in this study. The anti-VEGF polyclonal antibody was made against the NH₃-terminal peptide (1 to 20 residues) of VEGF. The specificity of VEGF immunoreactivity was confirmed by disappearance of immunoperoxidase signals on blocking with the VEGF NH₃-terminal peptide (Santa Cruz Biotechnology, Inc.) (data not shown). The antisubunit α_v polyclonal antibody cross-reacted with integrin α_vβ₃ and α_vβ₃. Although the cross-reactivity of this antibody with α_vβ₃ might have led to overestimation of α_vβ₃ expression, a previous study with the same primate model demonstrated that α_vβ₅ was not expressed on micro vessels in the basal ganglia during ischemia (see Results) (Okada et al., 1996). This antibody also cross-reacts with platelet integrin α_vβ₃ in addition to microvascular α_vβ₃; however, early after MCAO, the platelet contribution was expected to be low based on previous data (Okada et al., 1994). Working dilutions of each antibody preparation were optimized in preliminary studies.

Immunohistochemistry

Immunoperoxidase studies were performed on either serial 10-μm frozen sections or serial 3-μm paraffin sections. Frozen sections were fixed with a 1:1 (volume/volume [v/v]) mixture of acetone and methanol for 5 minutes and treated with 10 μmol/L glycine in phosphate-buffered saline (100 mmol/L Na₂HPO₄/NaH₂PO₄ and 150 mmol/L NaCl adjusted to pH 7.4), then incubated with Blotto to block nonspecific antibody reactions (Johnson et al., 1984). The fixed sections then were incubated with the primary antibody at 37°C for 2 hours, followed by incubation with biotinylated horse (or goat) secondary antibodies to anti-mouse (or rabbit) IgG (Vector Laboratories, Burlingame, CA, U.S.A.) at 37°C for 30 minutes. Avidin–biotin complexes were generated with streptavidin–horseradish peroxidase (Vector Laboratories) and detected with 3-amino-9-ethyl carbazole (AEC kit, Biomeda Corp., Foster City, CA, U.S.A.). The sections were then counterstained with Mayer's hematoxylin (Biomeda Corp.) or were left unstained. Paraffin sections were subjected to the same procedures after deparaffinization. For immunoperoxidase studies with the antisubunit α_v polyclonal antibody using paraffin sections, the sections were incubated initially with 1.2 mg/mL trypsin (Sigma BioSciences) at 37°C for 15 minutes to unmask the epitope before the blocking treatment. Routine controls for each experiment included deletion of the primary and secondary antibodies and the use of an irrelevant primary antibody.

DNA scission

Cells displaying DNA fragmentation were detected by incorporation of digoxigenin–deoxyuridine triphosphate (dUTP) with DNA polymerase I as previously described (Tagaya et al., 1997). After fixation with acetone and methanol, frozen sections were incubated with 0.1 U/μL DNA polymerase I (Promega, Madison, WI, U.S.A.) and digoxigenin DNA labeling mixture (Boehringer Mannheim Corp., Indianapolis, IN, U.S.A.) in translation buffer (50 mmol/L Tris-HCl [pH 7.5], 10 mmol/L MgSO₄, and 50 μg/mL bovine serum albumin) at 37°C for 2 hours. The sections were subsequently incubated with horseradish peroxidase–conjugated anti-digoxigenin antibody (Boehringer Mannheim Corp.) for 1 hour. The peroxidase signal was developed with 3-amino-9-ethyl carbazole (AEC kit, Biomeda Corp.). Paraffin sections were subjected to the same procedures after deparaffinization and pretreatment with 2 μg/mL proteinase K for 5 minutes. In each section, regions defined by cells incorporating dUTP (dUTP⁺) and regions peripheral thereto (dUTP⁻) were defined and designated ischemic core (Ic) and peripheral (Ip), respectively. Nonischemic regions were taken from stereotaxically identical sites in the contralateral basal ganglia.

In situ hybridization

In situ hybridization was performed as described previously using ³⁵S-UTP cyclic RNA probes on paraffin sections only (Seiffert et al., 1991). Both sense and antisense probes labeled with ³⁵S-UTP (Amersham Life Sciences, Arlington Heights, IL, U.S.A.) were synthesized from linearized subcloned plasmids using specific RNA polymerases. A 568-base pair (bp) fragment from residue 89 to 657 of mouse VEGF cDNA (GenBank M95200, kindly provided by Dr. Jeffrey Isner, St. Elizabeth's Medical Center, Boston, MA, U.S.A.) was subcloned into pCR II (Invitrogen Corp., San Diego, CA, U.S.A.), linearized with Not I, and transcribed with SP6 RNA polymerase for the antisense probe, or linearized with BamH I and transcribed with T7 RNA polymerase for the sense probe. A 401-bp fragment from residue 3075 to 3476 of human α_v integrin cDNA (GenBank J02826, kindly provided by Dr. Ingrid Stuiver, The Scripps Research Institute, La Jolla, CA, U.S.A.) was subcloned into pGem72f+, linearized with EcoR I, and transcribed with SP6 RNA polymerase for the antisense probe, or linearized with Hind III and transcribed with T7 RNA polymerase for the sense probe.

Deparaffinized PFA-fixed sections were incubated in 0.2 mol/L HCl for 10 minutes and digested with 5 μg/mL proteinase K for 10 minutes. After incubation in prehybridization buffer (50% formamide [v/v], 0.3 mol/L NaCl, 20 mmol/L Tris-HCl [pH 8.0], 5 mmol/L ethylenediamine tetraacetic acid [EOTA], 1% Oenhardt's solution, 10% dextran sulfate (weight/volume), and 10 mmol/L dithiothreitol] at 42°C for 3 hours, the sections were hybridized with 1 × 10⁶ cpm of ³⁵S-labeled riboprobe at 55°C for 18 hours. After hybridization, the sections were washed twice in 2 × sodium chloride/sodium citrate (SSC)/2 mmol/L EOTA, treated with RNase A (5 to 20μg/mL) for 30 minutes, and washed twice again in 2 × SSC/2 mmol/L EDT A. A high-stringency wash of 0.1 × SCC/%bT–mercaptoethanol/EDTA was performed at 55°C for 2 hours followed by washing four times in 0.5 × SSC. The sections were coated with NTB2 emulsion (Kodak, New Haven, CT, U.S.A.) and exposed in a sealed box at 4°C for 4 to 6 weeks. After development, the sections were counterstained with Richardson solution.

Quantitative analysis

The absolute number and size (the minimum transverse diameter) of microvessels that demonstrated signals by immunohistochemical study or by in situ hybridization were quantified with the aid of computerized video imaging microscopy (del Zoppo et al., 1991). A matrix of 100 nonoverlapping microscopic fields at 400× magnification comprised each 6.0-mm² region of interest. Two identical regions of interest were chosen within the Ic and Ip region, and one region of interest was chosen within the nonischemic region and in the left basal ganglia of control subjects. These regions of interest were superimposed in consecutive sections (immunostaining or in situ hybridization) and used to assess coexpression of the epitopes of interest on single microvessels or in groups of microvessels.

Statistical analysis

All values are expressed as the mean ± SO. Within each animal, microvessels were cross-classified according to the presence or absence of PCNA, VEGF, and integrin α_vβ₃ antigen or transcript using appropriate probes at 1 hour, 2 hours, or 7 days after MCAO. The corresponding counts were tabulated in multi way frequency tables. Hierarchical log-linear models, a standard statistical technique for categorical data analysis (Bishop et al., 1975), were used to analyze relations among these factors, and, in particular, to assess the null hypothesis that PCNA, VEGF, and α_vβ₃ expression were independent of one another. The quality of the fit of the log-linear models was assessed with deviance and chi-square statistics, with larger (nonsignificant) P values connoting adequate fits. Calculations were performed in Systat 6.0 for Windows (Wilkinson, 1996). Otherwise, significance was set at 2P < 0.05.

RESULTS

Patterns of nuclear DNA scission after middle cerebral artery occlusion

Cellular dUTP incorporation defined the regions of ischemic injury in the ipsilateral basal ganglia of all MCAO animals. Neurons, astrocytes, and microvascular cells labeled by digoxigenin-dUTP appeared in topographic patterns similar to those previously reported (Tagaya et al., 1997). At 1 and 2 hours after MCAO, dUTP was incorporated into nonvascular cells in clusters but into few microvascular cells within the Ic region. From a total of 66.1 ± 25.8 dUTP⁺ cells/mm2 in the Ic region at 2 hours after MCAO, only 0.8 ± 0.2 dUTP⁺ cells/mm² were associated with microvascular structures. By 7 days after MCAO, the number of dUTP⁺ microvascular cells had increased to 5.2 ± 2.4 dUTP⁺ cells/mm². At any time point, dUTP incorporation was associated predominantly with capillaries and small noncapillary microvessels with a mean diameter of 8.4 ± 3.9 μm (Fig. 1).

FIG. 1.

Microvascular distribution of deoxyuridine triphosphate (dUTP), proliferating cell nuclear antigen (PCNA), vascular endothelial growth factor (VEGF) and integrin α_vβ₃ by transverse diameter (μm) in the ischemic core (Ic) and peripheral (Ip) regions of ischemic subjects at various periods after middle cerebral artery occlusion (MCAO) (1 hour [n = 3], 2 hours [n = 3], and 7 days [n = 4]). Deoxyuridine triphosphate and PCNA were expressed by endothelial cells, smooth muscle cells, or both. VEGF and α_vβ₃ were seen in the myointima. □, 4.0 to 7.5 μm; ■, 7.5 to 30.0 μm; 30.0 to 50.0 μm; more than 50 μm; C, control.

PCNA expression in microvessels after middle cerebral artery occlnsion

Among all experiments, no detectable evidence of PCNA, integrin α_vβ₃, or VEGF was seen in any non ischemic basal ganglia. However, PCNA was found within the ischemic basal ganglia at both 1 or 2 hours after MCAO, predominantly in the Ic region. The PCNA⁺ nonvascular cells were clustered predominantly within the border zone of the Ic region, whereas in noncapillary microvessels PCNA was observed in the endothelium and the medial SMC, where it co-localized equally with VEGF or integrin α_vβ₃ in the Ic region (Fig. 1, Table 1). The number of microvessels displaying PCNA increased from 1 to 2 hours after MCAO (2.1 ± 1.9 to 6.0 ± 2.4 vessels/mm², 2P = 0.144), but this difference did not reach significance (Table 2). In the Ip region, only 0.3 ± 0.3 microvessels/mm² displayed PCNA immunoreactivity at 2 hours after MCAO. By 7 days after MCAO, the number of PCNA immunoreactive microvessels in both the Ic and Ip regions was low (0.4 ± 0.4 and 0.7 ± 0.4 vessels/mm², respectively).

TABLE 1.

Colocalization of PCNA with VEGF or integrin α_v_>β₃_>

	Microvascular cells

Colocalization	EC(%)	SMC(%)	SMC + EC(%)
PCNA + integrin α_vβ₃ (subunit α_v)	43.9	50.5	5.6
PCNA + VEGF	40.2	54.3	5.5

Over all time points and subjects after middle cerebral artery occlusion. EC, endothelial cells; PCNA, proliferating cell nuclear antigen; SMC, smooth muscle cells; VEGF, vascular endothelial growth factor.

TABLE 2.

Density of microvessels expressing dUTP, PCNA, VEGF, and integrin α_v_>β₃_>

		Density of detectable microvessels (mm⁻²)

Duration	n	dUTP (EC)	dUTP (SMC)	PCNA (EC)	PCNA (SMC)	VEGF	α_vβ₃
Control	3	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0
MCAO
Ic Region
1 hour	3	0.3 ±0.0	0.1 ± 0.1	1.2 ± 1.0	1.0 ± 0.9	1.5 ± 1.3	2.9 ± 1.7
2 hour	3	0.7 ± 0.2	0.2 ± 0.0	3.9 ± 2.2	3.2 ± 1.1	3.9 ± 1.0	13.7 ± 5.8
7 day	4	3.4 ± 1.7	0.5 ± 0.3	0.2 ± 0.2	0.2 ± 0.3	0.4 ± 0.6	1.5 ± 0.7
Ip Region
1 hour	3	0.0 ± 0.0	0.0 ± 0.0	0.1 ± 0.1	0.2 ± 0.2	0.0 ± 0.0	1.2 ± 0.5
2 hour	3	0.0 ± 0.0	0.0 ± 0.0	0.2 ± 0.2	0.1 ± 0.1	0.2 ± 0.3	1.6 ± 0.9
7 day	4	0.0 ± 0.0	0.0 ± 0.0	0.5 ± 0.3	0.2 ± 0.1	0.3 ± 0.1	2.2 ± 0.6
N Region
1 hour	3	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0
2 hour	3	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0
7 day	4	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0

All values are mean ± SD. EC, endothelial cells; MCAO, middle cerebral artery occlusion; PCNA, proliferating cell nuclear antigen; SMC, smooth muscle cells; VEGF, vascular endothelial growth factor.

Integrin α_vβ₃ expression in microvessels after middle cerebral artery occlusion

Because LM609 consistently cross-reacts with integrin α_vβ₃ only in frozen tissue, a polyclonal antibody against α_v, which cross-reacts with α_vβ₃ (and α_vβ₅), was used against the PFA-fixed tissue. A significant cross-correlation between the number of microvessels identified by both antibodies in frozen tissues (n = 10, r = 0.9421, 2P < 0.0001) and between LM609 (frozen) and the polyclonal antibody (PFA-fixed) (n = 10, r = 0.8940, 2P < 0.0001) was shown in preliminary experiments. It was concluded that the polyclonal antibody could identify microvascular α_vβ₃ expression in PFA-fixed tissue in a nearly equivalent manner to LM609 in the frozen preparations (see Materials and Methods).

No microvessel-associated α_v antigen was detected in nonischemic tissues. In the ischemic basal ganglia at 1 and 2 hours after MCAO, subunit α_v antigen appeared in the media of 7.5- to 30.0-μm diameter microvessels as previously shown (Okada et al., 1996) (Fig. 1). In the Ic region, the number of α_v-immunoreactive microvessels increased from 1 to 2 hours after MCAO (2.9 ± 1.7 vessels/mm² to 13.7 ± 5.8 vessels/mm2, 2P = 0.065) (Table 2). By comparison, a few microvessels expressing α_v antigen also were found in the Ip region at 2 hours after MCAO. By 7 days after MCAO, the number of subunit α_v-immunoreactive microvessels in the Ic region also had decreased but was not different from those in the surrounding Ip region (Table 2).

VEGF expression in ischemic microvessels after middle cerebral artery occlusion

At 1 and 2 hours after MCAO, VEGF also appeared in the media of 7.5- to 30.0-μm diameter microvessels in both the Ic and Ip regions (Fig. 1). Within the Ic region, the area of VEGF⁺ microvessels was completely encompassed by the area of microvessels expressing integrin α_vβ₃ at all time points (Fig. 2). Here, microvascular VEGF expression increased from 1 to 2 hours MCAO (1.5 ± 1.3 to 3.9 ± 1.0 microvessels/mm², 2P = 0.105) (Table 2), but there were far fewer VEGF⁺ microvessels in the Ip region at 2 hours after MCAO (0.2 ± 0.3 microvessels/mm²). By 7 days after MCAO, the number of VEGF-expressing microvessels in the Ic region had decreased to 0.4 ± 0.6 vessels/mm².

FIG. 2.

Distribution of microvessel-associated immunoreactive VEGF (black) and integrin α_v (gray) in the ischemic basal ganglia of each subject after MCAO. Notice the larger fields of distribution of subunit α_v (integrin α_vβ₃) than VEGF. The numbers below each basal ganglia designate the individual subjects. () α_vβ₃; (■) VEGF; () Overlap.

The VEGF antigen also appeared in nonvascular cells at 1 and 2 hours after MCAO, including small clusters of astrocytes in the Ip region and in the Ic/Ip border zone. Only within the Ic region, polymorphonuclear (PMN) leukocytes within or adjacent to microvessels or associated with hemorrhage were another nonvascular source of VEGF antigen (data not shown).

Co-localization of PCNA, VEGF, and integrin α_vβ₃

Proliferating cell nuclear antigen, VEGF, and integrin α_vβ₃ were detected together on selected microvessels followed through multiple serial sections only within the ischemic basal ganglia (Fig. 3). To determine the strength of the interaction of the three antigens, log-linear models of the three pairs of antigens appearing on the same microvessels in the fixed samples were prepared. With data from the Ic region, it was found that the log-linear models with subjects, and PCNA, VEGF, or α_vβ₃ expression constituting first-order effects, together with second-order interactions (pairwise interactions between the first-order factors) fit the cell frequencies remarkably well (Table 3). In this context, the significance of pairwise interactions connotes the lack of independence of PCNA, VEGF, and integrin α_vβ₃ expression. Furthermore, each of the three models (Table 3) was parsimonious in that deletion of any of the second-order interactions (or first-order effects) resulted in a significantly poorer fit.

FIG. 3.

Co-localization of PCNA, VEGF, and α_v antigens in a representative 25-μm diameter microvessel at 2 hours after MCAD in serial 3-μm sections (transverse view). The PCNA immunoreactivity (A), VEGF antigen (B), and integrin α_v antigen (C) are evident in the media. (D) No dUTP incorporation is apparent in the same microvessel.

TABLE 3.

Pairwise interactions

Comparisons	χ²₉	2p
Subject vs PCNA vs VEGF	8.42	0.49
Subject vs PCNA vs integrin α_vβ₃	8.85	0.45
Subject vs VEGF vs integrin α_vβ₃	10.78	0.29

The χ² statistics are deviances, with nonsignificant values indicative of better fits. PCNA, proliferating cell nuclear antigen; VEGF, vascular endothelial growth factor.

Despite substantial animal-to-animal heterogeneity of expression, subjects constituted a highly significant main effect in this log-linear modeling. The subjects could not be placed into relatively homogeneous subgroups by time (1 hour, 2 hours, or 7 days), as might have been anticipated at the beginning of the study, however. On the other hand, the subjects were found to cluster into three relatively homogeneous subgroups independent of time based on joint PCNA, VEGF, and integrin α_vβ₃ expression in the Ic region (Fig. 4). Notably, in the Ip region, there was no significant relation among the microvascular expressions of PCNA, VEGF, and integrin α_vβ₃.

FIG. 4.

Co-localization of PCNA, VEGF, and integrin α_v in microvessels of the Ic region. Microvessels were cross-classified according to the presence or absence of the three antigens. The 10 subjects were distributed into three subgroups. Group I (relatively high expression of PCNA, VEGF, and integrin α_vβ₃) consisted of five subjects that underwent 1 hour (n = 2) or 2 hours (n = 3) of MCAO. Group II (intermediate group between groups I and III) consisted of two subjects at 7 days of MCAO. Group III (almost no expression of any of PCNA, VEGF, and integrin α_vβ₃) consisted of three subjects at 1 hour (n = 1) and 7 days (n = 2) of MCAO.

Expression of subunit α_v and VEGF mRNA

No VEGF or subunit α_v mRNA transcripts were detected in the nonischemic regions of any subject. In the ischemic basal ganglia, VEGF and subunit α_v mRNA were found on the same microvessels followed through multiple serial sections (Fig. 5). Primarily in the Ic region, subunit α_v and VEGF transcripts were detected together in the media of noncapillary microvessels of 7.5- to 30.0-μm diameter, like α_vβ₃ antigen (Okada et al., 1996) (Fig. 6). From 1 to 2 hours after MCAO, a modest increase in the number of microvessels displaying subunit α_v and VEGF mRNA was seen only in the Ic region (Table 4), which paralleled antigen expression. As with group I, which displayed significant co-localization of PCNA, VEGF, and integrin α_vβ₃ antigens (Fig. 4), the relation of α_v subunit to VEGF transcripts in the vascular media of each subject also was significant (2P < 0.05).

FIG. 5.

Co-localization of mRNA transcripts of the integrin α_v subunit and VEGF in a 22-μm diameter microvessel at 2 hours' MCAO in serial transverse sections. VEGF transcripts (A), and α_v transcripts (B) were detected in the media of the same microvessel.

FIG. 6.

Distribution of microvessel-associated integrin subunit α_v and VEGF mRNA transcripts by transverse diameter (μm) in the Ic and Ip regions of ischemic subjects after MCAD (1 hour [n = 3], 2 hours [n = 3], and 7 days [n = 4]). □, 4.0 to 7.5 μm; ■, 7.5 to 30.0 μm; , 30.0 to 50.0 μm; , more than 50 μm; C, control.

TABLE 4.

Density of microvessels expressing VEGF and subunit α_vtranscripts

	Density of detectable micro vessels (mm⁻²)

Duration	n	VEGF	α_v
Control	3	0.0 ± 0.0	0.0 ± 0.0
MCAD
lc Region
1 hour	3	2.3 ± 1.7	2.5 ± 2.0
2 hour	3	5.6 ± 0.7	3.8 ± 1.2
7 day	4	1.0 ± 0.5	0.4 ± 0.5
lp Region
1 hour	3	0.1 ± 0.1	0.2 ± 0.2
2 hour	3	0.7 ± 0.4	0.1 ± 0.2
7 day	4	1.1 ± 0.8	0.2 ± 0.2
N Region
1 hour	3	0.0 ± 0.0	0.0 ± 0.0
2 hour	3	0.0 ± 0.0	0.0 ± 0.0
7 day	4	0.0 ± 0.0	0.0 ± 0.0

All values are mean ± SD. MCAO, middle cerebral artery occlusion; VEGF, vascular endothelial growth factor.

The VEGF mRNA transcripts also were expressed by astrocytes and PMN leukocytes in the ischemic basal ganglia in a pattern similar to VEGF antigen. At 1 and 2 hours after MCAO only, astrocytes in the Ip region and the Ic/Ip border zone showed VEGF mRNA transcripts.

Interestingly, a relatively strong VEGF mRNA signal also was associated with cells next to some microvessels in the Ic region displaying VEGF antigen in the media. These VEGF-expressing cells had the morphologic features of PMN leukocytes that had penetrated the vascular wall (Fig. 7) or histiocyte-appearing cells localized within the vascular wall (Kato et al., 1996).

FIG. 7.

Cells with the morphologic makeup of polymorphonuclear (PMN) leukocytes displaying VEGF transcripts adjacent to a 14.6-μm diameter microvessel in the Ic region at 2 hours' MCAO.

DISCUSSION

The association of integrin α_vβ₃ with the microvascular deposition of fibrin after MCAD demonstrated by Okada et al. (1996) highlights one rapid response of the cerebral microvasculature to ischemia. Whether and how the presentation of the ligand, fibrin(ogen) in this instance, stimulates microvascular integrin α_vβ₃ expression is not apparent. One explanation is that integrin α_vβ₃ up-regulation is one event associated with microvascular activation stimulated by ischemia, and that ligand presentation occurs by independent but parallel processes. Alternatively, integrin α_vβ₃ presentation may be stimulated by an activator that also increases endothelial cell permeability, exposing the plasma to the tissue factor surrounding the blood vessel (del Zoppo et al., 1992). Senger et al. (Senger, 1996) have shown that exogenous VEGF can induce integrin α_vβ₃ expression in microvascular endothelial cells in vitro. Also, VEGF is known to increase endothelial cell permeability. Furthermore, both integrin α_vβ₃ and VEGF up-regUlation have been shown to prevent endothelial cell apoptosis in vitro (Spyridopoulos et al., 1997; Stromblad et al., 1996). Those observations suggest that a specific relation between VEGF and integrin α_vβ₃ could exist in the responses of microvessels to ischemia in the brain. If true, VEGF and integrin α_vβ₃ should coincide with evidence of microvascular activation in the initial response of the microvasculature to focal ischemia, and these responses could be time dependent. In this study, the corresponding null hypothesis of no co-localization of PCNA, VEGF, and integrin α_vβ₃ expression was effectively refuted by the log-linear modeling. Analysis of pairwise associations in individual microvessels demonstrated that a highly significant and coordinated expression of VEGF, integrin α_vβ₃, and PCNA was provoked in the microvessels of the Ic region beginning within 1 hour after MCAO.

The appearance of PCNA, a 29-kd accessory protein for DNA polymerases δ and ε (Shivji et al., 1995), in microvascular endothelium and SMC suggests a significant immediate microvascular cell response to focal ischemia. Expression of PCNA in the nucleoplasm clearly reflects proliferative responses of cells to growth factors, DNA injury or repair, and other initiators of cell activation during G1 and S phase (Kelman, 1997; Kurki et al., 1988). Expression of PCNA, then, is closely related to cell cycle initiation and could reflect an activated cell proliferative response after MCAO. There has been no report of microvascular or cellular presentation of this antigen in acute cerebral ischemia. Integral to chromosomal DNA replication (polymerase δ) and repair (polymerase ε) (Shivji et al., 1995), PCNA expression may be involved in the processes of DNA repair (Kelman, 1997). As VEGF and integrin α_vβ₃ expression may be involved in suppression of the cell death pathway (Spyridopoulos et al., 1997; Strombladra et al, 1996), their expression with PCNA may contribute to the preservation of cerebral vessels under ischemic conditions. In contrast to the absence of PCNA in microvascular endothelial cells and pericytes before 2 days in a rodent middle cerebral artery ligation model (Liu, 1994 and Chen, 1994), the current experiments confirm that PCNA expression is one of the earliest cerebral microvascular endothelial cell and SMC responses to ischemia.

Vascular endothelial growth factor acts directly on endothelial cells to stimulate their replication (Connolly et al., 1989). Whereas VEGF did not increase proliferation of aortic SMC in culture (Weatherford et al., 1996) or in large-artery injury models (Asahara et al., 1996), exogenous VEGF has been shown to increase endothelial cell and SMC proliferation in 20 to 100 μm collateral arterioles of the ischemic lower limb in the rabbit (Takeshita et al., 1995; Tsurumi et al., 1996). The cellular responses to VEGF in the skeletal muscle collaterals are consistent with the finding of VEGF in endothelial cells and SMC in PCNA⁺ cerebral microvessels after MCAO.

Whereas the expressions of VEGF and integrin α_vβ₃ antigens on microvessels in the Ic region were significantly related, the number and area of distribution of microvessels expressing integrin α_vβ₃ consistently exceeded those expressing VEGF antigen (Fig. 2, Table 2). This pattern may result from differences in antibody sensitivity, contributions of VEGF from extravascular cellular sources, or the effects of stimulation of integrin α_vβ₃ by non-VEGF agonists such as basic fibroblast growth factor. Fibroblast growth factor is reported to increase cell surface expression of the integrin α_v and β₃ subunits in cultured endothelial cells (Sepp et al., 1994).

The consistently larger area of expression of integrin α_vβ₃ than VEGF also suggests that the two vascular elements may reflect differential responses to one or more products of ischemia. Induction of VEGF, and, perhaps, integrin α_vβ₃ on microvessels probably results from local hypoxia. Under hypoxic conditions, increased VEGF expression has been detected in cultured astrocytes (Ijichi et al., 1995), tumor cell lines (Minchenko et al., 1994; Berse et al., 1992), and arterial SMC (Brogi et al., 1994). The VEGF and subunit (v genes both contain a hypoxia-inducible factor-1 (HIF-1) response element in their promoter regions, which is activated by the binding of HIF-1 under hypoxic conditions in vivo (Forsythe et al., 1996; Gleadle and Ratcliffe, 1997). The presence of the HIF-1 target sequence 5′-GACGTGGA-3′ (bp 791 to 798 of α_v integrin promoter cDNA, GenBank U07375) in both genes implies that induction of VEGF and perhaps subunit α_v may be independently regulated by an HIF-1 response. Further proof of the dependence of integrin α_vβ₃ on VEGF expression requires detailed examination of HIF-1 expression in this model with the use of specific inhibitors.

We failed to find any time relation of joint microvascular PCNA, VEGF, and integrin α_vβ₃ expression in the Ic region. The known relation of VEGF expression to HIF-1 induction hints at a basis for the lack of temporal relation in these studies. It is likely that in the injured territory of each animal subject, subregions within Ic experience ischemic injury at various times based on the changes or shunting of blood flow during MCAO. The resulting topographic heterogeneity of expression means that the appearance of all three elements by 1 to 2 hours MCAO in a given subject would be “averaged” so that any time dependence of expression, if present, would be obscured.

Two previous studies of VEGF expression in experimental focal cerebral ischemia in rodents differ from the current results. Kovacs et al. describe the appearance of VEGF on macrophages, neurons, glial cells, and endothelial cells in the central ischemic region (from 18 hours to 5 days) and in the peripheral area (from 18 hours to 14 days) after irreversible MCAO in the rat (Kovacs et al., 1996). Separately, Hayashi et al. describe up-regulation of VEGF antigen and mRNA by neurons and glial cells at 1 to 3 hours reperfusion after 1-hour MCAO (Hayashi et al., 1997). The more rapid appearance of microvascular VEGF in the primate basal ganglia is consistent with the increased susceptibilities of neurons to ischemic injury (Tagaya et al., 1997). Other forms of injury to the CNS may trigger VEGF. After cold injury and stab injury in rodent models, VEGF expression in astrocytes was maximal at 3 to 4 days (Papavassiliout et al., 1997), whereas VEGF appeared at various times in PMN leukocytes (1 hour to 1 day), astrocytes (3 to 8 days), macrophages (3 to 8 days), and vascular smooth muscle cells (6 hours to 10 days) (Nag et al., 1997). The reasons for these temporal differences in VEGF appearance in individual cell groups of nonprimate species are not known but may represent differences in species, model preparation, and injury type. In any event, the early coexpression of VEGF and integrin α_vβ₃ in the primate represents a prompt response of noncapillary cerebral microvessels to focal ischemia that may have a broad temporal presentation with persistent cerebral injury.

One possible effect of VEGF and integrin α_vβ₃ expression is the stimulation of angiogenesis downstream. However, there is limited information about the late vascular responses to focal cerebral ischemia. Tsutsumi et al. demonstrated capillary bud formation by 7 days after MCAO in the infarct zone in a canine stroke model (Tsutsumi, 1986; Tsutsumi et al., 1986). Chuaqui and Tapia subsequently demonstrated that in the infarcted zone, whereas most brain microvessels disappeared by 7 days after stroke onset, a subgroup of small normal-appearing muscular arterioles was frequently preserved within the first 4 weeks, followed by an increase in their number (Chuaqui and Tapia, 1993). In stroke patients who survived from 5 to 92 days, the number of microvessels within the infarcted hemisphere was seen to increase (Krupinski et al., 1994). In the current study, there was no clear evidence that microvessel density increased in either the Ic or the Ip region of the ischemic basal ganglia by 7 days MCAO (data not shown). However, the presence of VEGF and integrin α_vβ₃ within Ic suggests that signals associated with vascular proliferation are generated early. The precise mechanism of VEGF and integrin α_vβ₃ up-regulation and of their relation to neovascularization after focal cerebral ischemia are not known.

This report indicates that the early appearance of integrin α_vβ₃ and VEGF correlate with microvascular activation and the activation of intravascular coagulation, leading to fibrin formation (Okada et al., 1996). Thrombin generation may provide a link between microvascular reactivity and permeability, and the local cleavage of fibrinogen (Aschner et al., 1997; Garcia et al., 1996). It has been reported that osteopontin, a ligand for integrin α_vβ₃, is up-regulated late (more than 5 days) after MCAO in the cerebral parenchyma of the rat (Wang et al., 1998). Ligation of osteopontin to integrin α_vβ₃ is enhanced when osteopontin is cleaved by thrombin (Senger et al., 1996). Although osteopontin may induce migration of both endothelial cells and SMC, no relation to the early microvascular responses to ischemia has been reported.

The rapid appearance of PCNA, VEGF, and integrin α_vβ₃ on microvessels in the ischemic core also coincides with evidence of neuron injury in the nonhuman primate (Tagaya et al., 1997). Possible relations that may determine whether the activation of the cerebral microvessels are linked to neuron injury have yet to be explored. However, a highly significant correlation between neuron injury (or size of ischemic injury) and the appearance of pro-MMP-2 beginning within 1 hour after MCAO in this model has been shown (Heo et al., 1999). Zucker et al. have shown that exogenous VEGF stimulates endothelial cells to generate MMP-2 (Zucker et al., 1998). In the presence of coagulation factors, VEGF also may generate thrombin, known to activate pro-MMP-2 (Zucker et al., 1998). It has been proposed that the effects of MMP-2 on the microvascular basal lamina may be related to neuron injury (Heo et al., 1999). With the association of TIMP-1 and MMP-1 expression to VEGF stimulation in both tumor and endothelial cell culture systems (Unemori et al., 1992; Yoshiji et al., 1998; Zucker et al., 1998), additional pathways to microvascular remodeling and thrombin generation may contribute to neuron injury.

To explore the downstream events from integrin α_vβ₃ expression, interventional approaches using peptides containing the RGD sequence, which effectively block integrin α_vβ₃ function in tumor angiogenesis, may be tried (Brooks et al., 1994). Similarly, functional inhibitors of VEGF may prove useful to specifically define its relation to integrin α_vβ₃ expression and microvascular responses to ischemia. Ultimately, it is of interest whether the coexpression of VEGF and integrin α_vβ₃ on activated cerebral microvessels leads to changes in microvascular integrity during ischemia, to microvascular remodeling and the generation of new vessels, and to alterations in neuron integrity.

Footnotes

Abbreviations used

References

Asahara

Chen

Tsurumi

Kearney

Rossow

Passeri

Symes

(1996) Accelerated restitution of endothelial integrity and endothelium-dependent function after phVEGF16 gene transfer. Circulation 94(12):3291–3302

Aschner

Lum

Fletcher

Malik

(1997) Bradykinin- and thrombin-induced increases in endothelial permeability occur independently of phospholipase C but require protein kinase C activation. J Cell Physiol 173:387–396

Berse

Brown

Van De Water

Dvorak

Senger

(1992) Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages, and tumors. Mol Biol Cell 3:211–220

Bishop

YMM

Fienberg

Holland

(1975) Discrete Multivariate Analysis: Theory and Practice, Cambridge, McGraw-Hill

Brogi

Namiki

Isner

(1994) Indirect angiogenic cytokines upregulate VEGF and bFGF gene expression in vascular smooth muscle cells, whereas hypoxia upregulates VEGF expression only. Circulation 90:649–652

Brooks

Clark

RAF

Cheresh

(1994) Requirement of vascular integrin α_vβ₃ for angiogenesis. Science 264:569–571

Brooks

Montgomery

Rosenfeld

Reisfeld

Klier

(1994) Integrin α_vβ₃ antagonists promote tumor regression by promoting apoptosis of angiogenic blood vessels. Cell 79:1157–1164

Brown

Yeo

K-T

Berse

Yeo

T-K

Senger

Dvorak

(1992) Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med 176:1375–1379

Cheresh

(1993) Integrins: Structure, function and biological properties. Adv Mol Cell Biol 6:225–252

10.

Chuaqui

Tapia

(1993) Histologic assessment of the age of recent brain infarcts in man. J Neuropathol Exp Neurol 52(5):481–489

11.

Clauss

Grell

Fangmann

Fiers

Scheurich

Risau

(1996) Synergistic induction of endothelial tissue factor by tumor necrosis factor and vascular endothelial growth factor: functional analysis of the tumor necrosis factor receptors. FEBS Lett 390:334–338

12.

Connolly

Heuvelman

Nelson

Olander

Eppley

Delfino

(1989) Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis. J Clin Invest 84(5):1470–1478

13.

del Zoppo

Copeland

Harker

Waltz

Zyroff

Hanson

(1986) Experimental acute thrombotic stroke in baboons. Stroke 17:1254–1265

14.

del Zoppo

Schmid-Schönbein

Mori

Copeland

Chang

C-M

(1991) Polymorphonuclear leukocytes occlude capillaries following middle cerebral artery occlusion and reperfusion in baboons. Stroke 22:1276–1283

15.

del Zoppo

J-Q

Copeland

Thomas

Schneiderman

Morrissey

(1992) Tissue factor localization for non-human primate cerebral tissue. Thromb Haemost 68:642–647

16.

Detmar

Brown

Claffey

Yeo

K-T

Kocher

Jackman

(1994) Overexpression of vascular permeability factor/vascular endothelial growth factor and its receptors in psoriasis. J Exp Med 180:11421–11426

17.

Dvorak

Brown

Detmar

Dvorak

(1995) Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 146: 1029–1039

18.

Fava

Olsen

Spencer-Green

Yeo

K-T

Yeo

T-K

Berse

(1994) Vascular permeability factor/vascular endothelial growth factor (VPF/VEGF): accumulation and expression in human synovial fluids and rheumatoid synovial tissue. J Exp Med 180:341–346

19.

Forsythe

Jiang

Iyer

Agani

Leung

Koos

(1996) Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol 16:4604–4613

20.

Garcia

Verin

Schaphorst

(1996) Regulation of thrombinmediated endothelial cell contraction and permeability. Semin Thromb Hemost 22:309–315

21.

Gleadle

Ratcliffe

(1997) Induction of hypoxia-inducible factor-1, erythropoietin, vascular endothelial growth factor, and glucose transporter-1 by hypoxia: evidence against a regulatory role for Src kinase. Blood 89:503–509

22.

Haring

H-P

Berg

Tsurushita

Tagaya

del Zoppo

(1996) E-selectin appears in non-ischemic tissue during experimental focal cerebral ischemia. Stroke 27: 1386–1392

23.

Hayashi

Abe

Suzuki

ltoyama

(1997) Rapid induction of vascular endothelial growth factor gene expression after transient middle cerebral artery occlusion in rats. Stroke 28:2039–2044

24.

Heo

Lucero

Abumiya

Koziol

Copeland

del Zoppo

(1999) Matrix metalloproteinases increase very early during experimental focal cerebral ischemia. J Cereb Blood Flow Metab 19:624–633

25.

Ijichi

Sakuma

Tofilon

(1995) Hypoxia-induced vascular endothelial growth factor expression in normal rat astrocyte cultures. Glia 14:87–93

26.

Johnson

Gautseh

Sportsman

Elder

(1984) Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acids transferred to nitrocellulose. Gene Anal Tech 1:3–8

27.

Kato

Okada

Takashima

Ohmori

Sunaga

(1996) Granule-laden perivascular cells observed in rabbits with experimental cerebral atherosclerosis. Angiology 47:343–347

28.

Kelman

(1997) PCNA: Structure, functions and interactions. Oncogene 14(6):629–640

29.

Kovacs

Ikezaki

Samoto

Inamura

Fukui

(1996) VEGF and flt. Expression time kinetics in rat brain infarct. Stroke 27:1865–1872

30.

Krupinski

Kaluza

Kumar

Wang

(1994) Role of angiogenesis in patients with cerebral ischemic stroke. Stroke 25:1794–1798

31.

Kurki

Ogata

Tan

(1988) Monoclonal antibodies to proliferating cell nuclear antigen (PCNA)/cyclin as probes for proliferating cells by immunofluorescence microscopy and flow cytometry. J Immunol Methods 109(1):49–59

32.

Leavesley

Schwartz

Rosenfeld

Cheresh

(1993) Integrin β₁- and β₃-mediated endothelial cell migration is triggered through distinct signaling mechanisms. J Cell Biol 121: 163–170

33.

Brown

Hibberd

Grossman

Morgan

Simons

(1996) VEGF, flk-1, and flt-1 expression in a rat myocardial infarction model of angiogensis. Am J Physiol 270:H1803–H1811

34.

Liu HMChen

(1994) Correlation between fibroblast growth factor expression and cell proliferation in experimental brain infarct: studied with proliferating cell nuclear antigen immunohistochemistry. J Neuropathol Exp Neurol 53(2): 118–126

35.

Minchenko

Bauer

Salceda

Caro

(1994) Hypoxic stimulation of vascular endothelial growth factor expression in vitro and in vivo. Lab Invest 71:374–379

36.

Nag

Takahashi

Kilty

(1997) Role of vascular endothelial growth factor in blood–brain barrier breakdown and angiogenesis in brain trauma. J Neuropathol Exp Neurol 56(8):912–921

37.

Okada

Copeland

Fitridge

Koziol

del Zoppo

(1994) Fibrin contributes to microvascular obstructions and parenchymal changes during early focal cerebral ischemia and reperfusion. Stroke 25:1847–1854

38.

Okada

Copeland

Hamann

Koziol

Cheresh

del Zoppo

(1996) Integrin α_vβ₃ is expressed in selected microvessels following focal cerebral ischemia. Am J Pathol 149:37–44

39.

Okada

Copeland

Mori

Tung

M-M

Thomas

del Zoppo

(1994) P-selectin and intercellular adhesion molecule-l expression after focal brain ischemia and reperfusion. Stroke 25:202–211

40.

Papavassiliou

Gogate

Proescholdt

Heiss

Walbridge

Edwards

(1997) Vascular endothelial growth factor (vascular permeability factor) expression in injured rat brain. J Neurosci Res 49(4):451–60

41.

Pepper

Ferrara

Orci

Montesano

(1991) Vascular endothelial growth factor (VEGF) induces plasminogen activators and plasminogen activator inhibitor-1 in microvascular endothelial cells. Biochem Biophys Res Commun 181:902–906

42.

Seiffert

Keeton

Eguchi

Sawdey

Loskutoff

(1991) Detection of vitronectin mRNA in tissues and cells of the mouse. Proc Natl Acad Sci USA 88:9402–9406

43.

Senger

(1996) Molecular framework for angiogenesis: a complex web of interactions between extravasated plasma proteins and endothelial cell proteins induced by angiogenic cytokines. Am J Pathol 149:1–7

44.

Senger

Ledbetter

Claffey

Papdopoulos-Sergiou

Peruzzi

Detmar

(1996) Stimulation of endothelial cell migration by vascular permeability factor/vascular endothelial growth factor through cooperative mechanisms involving the α_vβ₃ integrin, osteopontin, and thrombin. Am J Pathol 149:293–305

45.

Sepp

L-J

Lee

Brown

Caughman

Lawley

(1994) Basic fibroblast growth factor increases expression of the α_vβ₃ integrin complex on human microvascular endothelial cells. J Invest Dermatol 103:295–299

46.

Shivji

Podust

Hubscher

Wood

(1995) Nucleotide excision repair DNA synthesis by DNA polymerase epsilon in the presence of PCNA, RFC, and RPA. Biochemistry 34(15):5011–5017

47.

Spyridopoulos

Brogi

Kearney

Sullivan

Cetrulo

Isner

(1997) Vascular endothelial growth factor inhibits endothelial cell apoptosis induced by tumor necrosis factor-α: balance between growth and death signals. J Mol Cell Cardil 29:1321–1330

48.

Stromblad

Becher

Yebra

Brooks

Cheresh

(1996) Suppression of p⁵³ activity and p21^WAF1/CIP1 expression by vascular cell integrin α_vβ₃ during angiogenesis. J Clin Invest 98:426–433

49.

Tagaya

Liu

K-F

Copeland

Seiffert

Engler

Garcia

(1997) DNA scission following focal brain ischemia: temporal differences in two species. Stroke 28: 1245–1254

50.

Takeshita

Rossow

Kearney

Zheng

Bauters

Bunting

(1995) Time course of increased cellular proliferation in collateral arteries after administration of vascular endothelial growth factor in a rabbit model of lower limb vascular insufficiency. Am J Pathol 147(6):1649–1660

51.

Tsurumi

Takeshita

Chen

Kearney

Rossow

Passeri

(1996) Direct intramuscular gene transfer of naked DNA encoding vascular endothelial growth factor augments collateral development and tissue perfusion. Circulation 94:3281–3290

52.

Tsutsumi

(1986) Experimental cerebral infarction in the dog: scanning electron microscopy with vascular endocasts of the microvessels in the ichemic brain. Neurol Med Chir 26:595–600

53.

Tsutsumi

Shibata

Inoue

Mori

(1986) Experimental cerebral infarction in the dog: ultrastructural study of microvessels in subacute cerebral infarction. Neurol Med Chir 27:73–77

54.

Unemori

Ferrara

Bauer

Amento

(1992) Vascular endothelial growth factor induces interstitial collagenase expression in human endothelial cells. J Cell Physiol 153:557–562

55.

Wagner

Tagaya

Koziol

Quaranta

del Zoppo

(1997) Rapid disruption of an astrocyte interaction with the extracellular matrix mediated by α₆β₄ during focal cerebral ischemia/reperfusion. Stroke 28:858–865

56.

Wang

Louden

Yue

Ellison

Barone

Solleveld

(1998) Delayed expression of osteopontin after focal stroke in the rat. J Neurosci 18(6):2075–2083

57.

Weatherford

Sackman

Reddick

Freeman

Stevens

Goldman

(1996) Vascular endothelial growth factor and heparin in a biologic glue promotes human aortic endothelial cell proliferation with aortic smooth muscle cell inhibition. Surgery 120(2):433–439

58.

Wilkinson

(1996) Systat 6.0 for Windows: Statistics, Chicago, SPSS, Inc.

59.

Y oshiji

Harris

Raso

Gomez

Lindsay

Shibuya

(1998) Mammary carcinoma cells over-expressing tissue inhibitor of metalloproteinases-1 show enhanced vascular endothelial growth factor expression. Int J Cancer 75:81–87

60.

Zucker

Mirza

Conner

Lorenz

Drews

Bahou

(1998) Vascular endothelial growth factor induces tissue factor and matrix metalloproteinase production in endothelial cells: conversion of prothrombin to thrombin results in progelatinase A activation and cell proliferation. Int J Cancer 75:780–786

Activated Microvessels Express Vascular Endothelial Growth Factor and Integrin α v β 3 During Focal Cerebral Ischemia

Abstract

Keywords

MATERIALS AND METHODS

Experimental focal cerebral ischemia

Antibodies

Immunohistochemistry

DNA scission

In situ hybridization

Quantitative analysis

Statistical analysis

RESULTS

Patterns of nuclear DNA scission after middle cerebral artery occlusion

PCNA expression in microvessels after middle cerebral artery occlnsion

Integrin αvβ3 expression in microvessels after middle cerebral artery occlusion

VEGF expression in ischemic microvessels after middle cerebral artery occlusion

Co-localization of PCNA, VEGF, and integrin αvβ3

Expression of subunit αv and VEGF mRNA

DISCUSSION

Footnotes

Abbreviations used

References

Integrin α_vβ₃ expression in microvessels after middle cerebral artery occlusion

Co-localization of PCNA, VEGF, and integrin α_vβ₃

Expression of subunit α_v and VEGF mRNA