Increased Microvascular Density and Enhanced Leukocyte Rolling and Adhesion in the Skin of VEGF Transgenic Mice

Abstract

Vascular endothelial growth factor (VEGF) has been implicated in the pathologic angiogenesis observed in psoriasis and other chronic inflammatory skin diseases that are characterized by enhanced expression of VEGF by epidermal keratinocytes and of VEGF receptors by tortuous microvessels in the upper dermis. To investigate the functional importance of chronic VEGF overexpression in vivo, we used a keratin 14 promoter expression cassette containing the gene for murine VEGF164 to selectively target VEGF expression to basal epidermal keratinocytes in transgenic mice. These mice demonstrated an increased density of tortuous cutaneous blood capillaries with elevated expression levels of the high affinity VEGF receptors, VEGFR-1 and VEGFR-2, most prominently during the neonatal period. In contrast, no abnormalities of lymphatic vessels were detected. In addition, the number of mast cells in the upper dermis was significantly increased in transgenic skin. Intravital fluorescence microscopy revealed highly increased leukocyte rolling and adhesion in postcapillary skin venules that were both inhibited after injection of blocking antibodies against E- and P-selectin. Combined blocking antibodies against intercellular adhesion molecule-1 and lymphocyte function-associated antigen-1 were without effect, whereas an anti-vascular cell adhesion molecule-1/VLA-4 antibody combination almost completely normalized the enhanced leukocyte adhesion in transgenic mice. This study reveals VEGF as a growth factor specific for blood vessels, but not lymphatic vessels, and demonstrates that chronic orthotopic overexpression of VEGF in the epidermis is sufficient to induce cardinal features of chronic skin inflammation, providing a molecular link between angiogenesis, mast cell accumulation, and leukocyte recruitment to sites of inflammation. Vascular endothelial growth factor (VEGF) has been implicated in the pathologic angiogenesis observed in psoriasis and other chronic inflammatory skin diseases that are characterized by enhanced expression of VEGF by epidermal keratinocytes and of VEGF receptors by tortuous microvessels in the upper dermis. To investigate the functional importance of chronic VEGF overexpression in vivo, we used a keratin 14 promoter expression cassette containing the gene for murine VEGF164 to selectively target VEGF expression to basal epidermal keratinocytes in transgenic mice. These mice demonstrated an increased density of tortuous cutaneous blood capillaries with elevated expression levels of the high affinity VEGF receptors, VEGFR-1 and VEGFR-2, most prominently during the neonatal period. In contrast, no abnormalities of lymphatic vessels were detected. In addition, the number of mast cells in the upper dermis was significantly increased in transgenic skin. Intravital fluorescence microscopy revealed highly increased leukocyte rolling and adhesion in postcapillary skin venules that were both inhibited after injection of blocking antibodies against E- and P-selectin. Combined blocking antibodies against intercellular adhesion molecule-1 and lymphocyte function-associated antigen-1 were without effect, whereas an anti-vascular cell adhesion molecule-1/VLA-4 antibody combination almost completely normalized the enhanced leukocyte adhesion in transgenic mice. This study reveals VEGF as a growth factor specific for blood vessels, but not lymphatic vessels, and demonstrates that chronic orthotopic overexpression of VEGF in the epidermis is sufficient to induce cardinal features of chronic skin inflammation, providing a molecular link between angiogenesis, mast cell accumulation, and leukocyte recruitment to sites of inflammation. vascular cell adhesion molecule-1 vascular endothelial cell growth factor VEGF receptor-1 (Flt-1) VEGF receptor-2 (Flk-1) Common chronic inflammatory diseases such as psoriasis and rheumatoid arthritis are characterized by leukocyte infiltration, angiogenesis, and vascular remodelling leading to enhanced tortuosity of blood microvessels (Ryan, 1980Ryan T.J. Microcirculation in psoriasis.Pharmacol Ther (B). 1980; 10: 27Crossref PubMed Scopus (72) Google Scholar;Braverman and Sibley, 1982Braverman I.M. Sibley J. Role of the microcirculation in the treatment and pathogenesis of psoriasis.J Invest Dermatol. 1982; 78: 12-17Abstract Full Text PDF PubMed Scopus (166) Google Scholar;Braverman and Keh-yen, 1986Braverman I.M. Keh-yen A. Three dimensional reconstruction of endothelial cell gaps in psoriatic vessels and their morphologic identity with gaps produced by the intradermal injection of histamine.J Invest Dermatol. 1986; 86: 577-581Crossref PubMed Scopus (39) Google Scholar;Bull et al., 1992Bull R.H. Bates D.O. Mortimer P.S. Intravital video-capillaroscopy for the study of microcirculation in psoriasis.Br J Dermatol. 1992; 126: 436Crossref PubMed Scopus (78) Google Scholar;Fava et al., 1994Fava R. Olsen N. Spencer-green G. et al.Vascular permeability factor/endothelial growth factor (VPF/VEGF): Accumulation and expression in human synovial fluids and rheumatoid synovial tissue.J Exp Med. 1994; 180: 341-346Crossref PubMed Scopus (486) Google Scholar;Koch et al., 1994Koch A.E. Harlow L.A. Haines G.K. et al.Vascular endothelial growth factor. A cytokine modulating endothelial function in rheumatoid arthritis.J Immunol. 1994; 152: 4149-4156PubMed Google Scholar). Previously, we have demonstrated increased expression of vascular endothelial growth factor (VEGF; also known as vascular permeability factor) by epidermal keratinocytes, and of the two high-affinity VEGF receptors, VEGFR-1 (Flt-1) (deVries et al., 1992deVries C. Escobedo J. Ueno H. Houck K. Ferrara N. Williams L.T. The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor.Science. 1992; 255: 989-991Crossref PubMed Scopus (1857) Google Scholar) and VEGFR-2 (KDR) (Terman et al., 1992Terman B.I. Dougher V.M. Carrion M.E. Dimitrov D. Armellino D.C. Gospodarowicz D. Bohlen P. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor.Biochem Biophys Res Commun. 1992; 187: 1579-1586Crossref PubMed Scopus (1371) Google Scholar) by dermal microvessels in psoriasis, contact dermatitis, and several chronic inflammatory bullous skin diseases with enhanced angiogenesis (Detmar et al., 1994Detmar M. Brown L.F. Claffey K.P. et al.Overexpression of vascular permeability factor/vascular endothelial growth factor and its receptors in psoriasis.J Exp Med. 1994; 180: 1141-1146Crossref PubMed Scopus (627) Google Scholar;Brown et al., 1995aBrown L.F. Harrist T.J. Yeo K.-T. et al.Increased expression of vascular permeability factor (vascular endothelial growth factor) in bullous pemphigoid, dermatitis herpetiformis, and erythema multiforme.J Invest Dermatol. 1995 a; 104: 744-749Crossref PubMed Scopus (134) Google Scholar, Brown et al., 1995bBrown L.F. Olbricht S.M. Berse B. et al.Overexpression of vascular permeability factor (VPF/VEGF) and its endothelial cell receptors in delayed hypersensitivity skin reactions.J Immunol. 1995 b; 154: 2801-2807PubMed Google Scholar). Moreover, enhanced VEGF and VEGF receptor expression was a characteristic feature of lesional skin in a recently described mouse model for chronic, psoriasiform skin inflammation (Schön et al., 1997Schön M.P. Detmar M. Parker C.M. Murine psoriasis-like disorder induced by naive CD4+ T cells.Nature Med. 1997; 3: 183-188Crossref PubMed Scopus (138) Google Scholar); however, it has remained unclear whether keratinocyte-secreted VEGF, predominantly the VEGF165 and VEGF121 splice variants (Ballaun et al., 1995Ballaun C. Weninger W. Uthman A. Weich H. Tschachler E. Human keratinocytes express the three major splice forms of vascular endothelial growth factor.J Invest Dermatol. 1995; 104: 7-10Crossref PubMed Scopus (123) Google Scholar;Detmar et al., 1995Detmar M. Yeo K.-T. Nagy J.A. et al.Keratinocyte-derived vascular permeability factor (vascular endothelial growth factor) is a potent mitogen for dermal microvascular endothelial cells.J Invest Dermatol. 1995; 105: 44-50Crossref PubMed Scopus (219) Google Scholar,Detmar et al., 1997Detmar M. Brown L.F. Berse B. Jackman R.W. Elicker B.M. Dvorak H.F. Claffey K.P. Hypoxia regulates the expression of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) and its receptors in human skin.J Invest Dermatol. 1997; 108: 263-268Abstract Full Text PDF PubMed Scopus (225) Google Scholar), can penetrate the epidermal–dermal basement membrane to reach its target cells on dermal microvessels, and whether VEGF itself might be involved in the observed upregulation of its receptors. Previously, we have shown that keratinocyte-derived VEGF is a potent mitogen for human dermal microvascular endothelial cells that express both VEGFR-1 and VEGFR-2 in vitro (Detmar et al., 1995Detmar M. Yeo K.-T. Nagy J.A. et al.Keratinocyte-derived vascular permeability factor (vascular endothelial growth factor) is a potent mitogen for dermal microvascular endothelial cells.J Invest Dermatol. 1995; 105: 44-50Crossref PubMed Scopus (219) Google Scholar). VEGF also enhanced endothelial cell migration through upregulation of the αvβ3 integrin (Senger et al., 1996Senger D.R. Ledbetter S.R. Claffey K.P. Papadopoulos-sergiou A. Perruzzi C.A. Detmar M. Stimulation of endothelial cell migration by vascular permeability factor/vascular endothelial growth factor through cooperative mechanisms involving the αvβ3 integrin, osteopontin, and thrombin.Am J Pathol. 1996; 149: 293-305PubMed Google Scholar) that has been found upregulated on skin microvessels in psoriasis (Creamer and Barker, 1995Creamer J.D. Barker J.N. Vascular proliferation and angiogenic factors in psoriasis.Clin Exp Dermatol. 1995; 20: 6-9Crossref PubMed Scopus (44) Google Scholar), and of the α1β1, and α2β1 integrins (Senger et al., 1997Senger D.R. Claffey K.P. Benes J.E. Perruzzi C.A. Sergiou A.P. Detmar M. Angiogenesis promoted by vascular endothelial growth factor: Regulation through α1β1, and α2β1integrins.Proc Natl Acad Sci USA. 1997; 94: 13612-13617Crossref PubMed Scopus (444) Google Scholar). These findings suggested a major role of VEGF in the mediation of the vascular remodelling characteristic of chronic skin inflammation; however, the precise biologic importance of VEGF for this process is still unknown because VEGF deficiency is lethal during early embryonic development (Carmeliet et al., 1996Carmeliet P. Ferreira V. Breier G. et al.Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele.Nature. 1996; 380: 435-439Crossref PubMed Scopus (3323) Google Scholar;Ferrara et al., 1996Ferrara N. Carver M.K. Chen H. et al.Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene.Nature. 1996; 380: 439-442Crossref PubMed Scopus (2928) Google Scholar), preventing evaluation of skin angiogenesis in VEGF deficient mice. Moreover, whereas injection of VEGF into the skin leads to acutely increased microvascular permeability to plasma macromolecules (Senger et al., 1990Senger D.R. Connolly D.T. Van De Water L. Feder J. Dvorak H.F. Purification and NH2-terminal amino acid sequence of guinea pig tumor- secreted vascular permeability factor.Cancer Res. 1990; 50: 1774-1778PubMed Google Scholar), chronic topical delivery of VEGF to the skin could not be achieved. To study the specific biologic consequences of chronic VEGF overexpression in the skin, we generated transgenic mice, using a transgene vector in which the coding sequence of murine VEGF164 was cloned into a human keratin 14 promoter expression cassette. The human keratin 14 expression cassette has previously been shown to selectively target transgene expression to basal keratinocytes of the skin (Vassar et al., 1989Vassar R. Rosenberg M. Ross S. Tyner A. Fuchs E. Tissue-specific and differentiation-specific expression of a human K14 keratin gene in transgenic mice.Proc Natl Acad Sci USA. 1989; 86: 1563-1567Crossref PubMed Scopus (291) Google Scholar;Vassar and Fuchs, 1991Vassar R. Fuchs E. Transgenic mice provide new insights into the role of TGF-alpha during epidermal development and differentiation.Genes Dev. 1991; 5: 714-727Crossref PubMed Scopus (361) Google Scholar;Turksen et al., 1992Turksen K. Kupper T. Degenstein L. Williams I. Fuchs E. Interleukin 6: insights to its function in skin by overexpression in transgenic mice.Proc Natl Acad Sci USA. 1992; 89: 5068-5072Crossref PubMed Scopus (185) Google Scholar;Guo et al., 1993Guo L. Yu Q.C. Fuchs E. Targeting expression of keratinocyte growth factor to keratinocytes elicits striking changes in epithelial differentiation in transgenic mice.Embo J. 1993; 12: 973-986Crossref PubMed Scopus (230) Google Scholar), providing thereby an orthotopical model to assess the effects of epidermis-derived VEGF on skin angiogenesis in vivo. In this study, we found that VEGF transgenic mice were characterized by an increased density of tortuous cutaneous blood capillaries with elevated expression levels of VEGFR-1 and VEGFR-2, most prominently during the neonatal period, whereas no abnormalities of lymphatic vessels were detected, establishing VEGF as a blood vessel-specific skin angiogenesis factor. Using intravital fluorescence microscopy, we detected dramatically increased leukocyte rolling and adhesion in postcapillary skin venules that were both completely inhibited after injection of blocking antibodies against E- and P-selectin. In addition, an anti-vascular cell adhesion molecule-1 (VCAM-1)/VLA-4 antibody combination almost completely normalized the enhanced leukocyte adhesion in transgenic mice. Thus, chronic orthotopic overexpression of VEGF in the epidermis was sufficient to induce cardinal pathologic features of psoriasis, providing further evidence for the role of VEGF as a major skin angiogenesis factor in chronic inflammation and as a novel molecular link between angiogenesis and leukocyte recruitment to sites of inflammation. A 980 bp murine VEGF164 cDNA (GenBank accession number M95200) was ligated into the BamHI restriction site of the keratin 14 expression cassette (Vassar et al., 1989Vassar R. Rosenberg M. Ross S. Tyner A. Fuchs E. Tissue-specific and differentiation-specific expression of a human K14 keratin gene in transgenic mice.Proc Natl Acad Sci USA. 1989; 86: 1563-1567Crossref PubMed Scopus (291) Google Scholar; kindly provided by Dr. Elaine Fuchs, University of Chicago; Figure 1), and a KpnI-HindIII fragment was purified and injected into FVB/N mouse zygotes. The injected embryos were transplanted into the uterus of pseudo-pregnant C21 mice. Transgenic founders were detected by Southern blot analysis of BamHI digested genomic DNA using a32P-labeled 980 bp murine VEGF cDNA as the probe. Genomic tail DNA also was subjected to polymerase chain reaction using two primers specific for human growth hormone sequences contained in the expression vector; 5′-CTCACCTAGCTGCAATGG-3′ and 5′-AAGGCACTGCCCTCTTGAAGC-3′. Initial denaturation at 94°C for 4 min was followed by 35 cycles at 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s, followed by final extension at 72°C for 5 min. Transgenic lines were established on the FVB genetic background. In situ hybridization of paraffin and frozen sections, obtained from four wild-type and four transgenic mice for each time point, was performed as described earlier (Detmar et al., 1994Detmar M. Brown L.F. Claffey K.P. et al.Overexpression of vascular permeability factor/vascular endothelial growth factor and its receptors in psoriasis.J Exp Med. 1994; 180: 1141-1146Crossref PubMed Scopus (627) Google Scholar) using pGEM or pBluescript II plasmids containing mouse VEGF, VEGFR-1, or VEGFR-2 cDNA fragments. In situ hybridizations were performed on abdominal, dorsal, and tail skin with comparable results. VEGFR-1 and -2 clones were a kind gift from Clive Wood (Genetics Institute, Cambridge, MA). The flk-1 and flt-1 sequences were isolated by polymerase chain reaction from a mouse fetal thymus cDNA library (Finnerty et al., 1993Finnerty H. Kelleher K. Morris G. et al.Molecular cloning of murine FLT and FLT4.Oncogene. 1993; 8: 2293-2298PubMed Google Scholar). The murine flk-1 transcription template was a 392 bp fragment encompassing amino acids 1–130 (nucleotides 268–660 of the flk-1 sequence described previously;Matthews et al., 1991Matthews W. Jordan C.T. Gavin M. Jenkins N.A. Copeland N.G. Lemischka I.R. A receptor tyrosine kinase cDNA isolated from a population of enriched primitive hematopoietic cells and exhibiting close genetic linkage to c-kit..Proc Natl Acad Sci USA. 1991; 88: 9026-9030Crossref PubMed Scopus (440) Google Scholar), cloned into pGEM-T (Promega, Madison, WI). The sequence for murine flt-1 was obtained by degenerate polymerase chain reaction cloning of kinase domains that resulted in a 640 bp cDNA fragment encoding the insert region from amino acid 832–1045 of a sequence described previously (Finnerty et al., 1993Finnerty H. Kelleher K. Morris G. et al.Molecular cloning of murine FLT and FLT4.Oncogene. 1993; 8: 2293-2298PubMed Google Scholar). The VEGFR-3 probe was kindly provided by Dr. Kari Alitalo and has been described elsewhere (Jeltsch et al., 1997Jeltsch M. Kaipainen A. Joukov V. Kukk E. Lymbousssaki A.X.M. Lakso M. Alitalo K. Hyperplasia of lymphatic vessels in VEGF-C transgenic mice.Science. 1997; 276: 1423-1425Crossref PubMed Scopus (1063) Google Scholar). Transcription reactions were carried out using a Riboprobe Gemini II kit (Promega) in the presence of (α-35S) UTP. Anti-sense (and control sense) probes were evaluated on alternate sections. For immunohistochemistry, 6 μm cryostat sections of ear, abdominal, or dorsal skin obtained from four wild-type and four transgenic mice were stained with anti-mouse platelet-endothelial cell adhesion molecule-1 (PECAM-1; CD31), anti-mouse intercellular adhesion molecule-1 (ICAM-1; CD54), anti-mouse E-selectin, P-selectin, L-selectin, anti-mouse CD3, CD4, CD8, CD 11a, CD11b, CD45, CD 49d (all from Pharmingen, San Diego, CA), anti-mouse collagen type IV, and anti-desmoplakin I/II (Biodesign International, Kenneburk, ME), and affinity purified rabbit anti-mouse collagen XVIII (Rehn and Pihlajaniemi, 1994Rehn M. Pihlajaniemi T. Alpha 1 (XVIII), a collagen chain with frequent interruptions in the collagenous sequence, a distinct tissue distribution, and homology with type XV collagen.Proc Natl Acad Sci USA. 1994; 91: 4234-4238Crossref PubMed Scopus (144) Google Scholar; kindly provided by Dr. Marko Rehn), using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA). Rat IgG1 (R59–40), rat IgG2a (R35–95), or hamster IgG (Ha4/8; all purchased from Pharmingen) were used as isotype matched negative controls in monoclonal antibody staining, and normal rabbit serum was used as control for polyclonal rabbit sera. For quantitation of microvascular densities in the skin, abdominal skin samples were obtained after CO2 euthanasia from six transgenic and six wild-type control mice each at 3 d, 11 d, 3 wk, and 6 wk after birth. Five micrometer paraffin sections were stained for CD31, and the number of vascular profiles per high-power field (×100 objective) was determined in the areas immediately below the epidermal–dermal basement membrane zone. At least five random fields per section were evaluated, and data were evaluated using the paired Student's t test. Three micrometer plastic sections were stained for chloroacetate-esterase reactivity (n = 6) using naphthol-AS-D-chloroacetate as substrate and 1% methyl green as counterstain. Mast cell staining was confirmed by methylene blue and Giemsa staining. Twenty-four day old wild-type or transgenic mice (n = 6 for each group) were injected intravenously with 50 μl fluoroscein isothiocyanate-labeled dextran (MW, 2000000; Sigma; 10 mg per ml), and video recording of ear skin perfusion was performed as described previously (Fukumura et al., 1995Fukumura D. Salehi H.A. Witwer B. Tuma R.F. Melder R.J. Jain R.K. Tumor necrosis factor alpha-induced leukocyte adhesion in normal and tumor vessels: effect of tumor type, transplantation site, and host strain.Cancer Res. 1995; 55: 4824-4829PubMed Google Scholar). Mice were also injected with 20 μl Rhodamin-6G (Molecular Probes, Eugene, OR; 50 mg per ml) to visualize peripheral blood leukocytes (Fukumura et al., 1995Fukumura D. Salehi H.A. Witwer B. Tuma R.F. Melder R.J. Jain R.K. Tumor necrosis factor alpha-induced leukocyte adhesion in normal and tumor vessels: effect of tumor type, transplantation site, and host strain.Cancer Res. 1995; 55: 4824-4829PubMed Google Scholar). In additional experiments, 24 d old mice (n = 3 for each group) were injected intraperitoneously with the following antibodies (150 μg each), either alone or in different combinations (Figure 6): blocking antibodies against E-selectin (clone 10E9.6), P-selectin (clone RB.40.34), ICAM-1 (clone 3E2), VCAM-1 (clone 429), L-selectin (clone MEL-14), lymphocyte function-associated antigen-1 (CD11a; clone M17/4), or VLA-4 (CD49d; clones R1–1 and 9C10; all purchased from Pharmingen) 3 h prior to examination. For fluorescence microlymphangiography, 5–10 μl of fluoroscein isothiocyanate-dextran (25 mg per ml) were injected intradermally into the tail skin (n = 3 for each group), and fluorescence micrographs were recorded 40–60 min after injection (Leu et al., 1994Leu A.J. Berk D.A. Yuan F. Jain R.K. Flow velocity in the superficial lymphatic network of the mouse tail.Am J Physiol. 1994; 267: H1507-13PubMed Google Scholar). The capillary tortuosity index was calculated by dividing 100 by the average length (μm) of capillaries without branching; a minimum of 500 branches was evaluated in each animal. The rolling count was calculated as 100 × number of rolling leukocytes/total leukocyte flux and adhesion density was calculated as the number of adherent leukocytes per mm2 vascular surface of postcapillary venules as described (Fukumura et al., 1995Fukumura D. Salehi H.A. Witwer B. Tuma R.F. Melder R.J. Jain R.K. Tumor necrosis factor alpha-induced leukocyte adhesion in normal and tumor vessels: effect of tumor type, transplantation site, and host strain.Cancer Res. 1995; 55: 4824-4829PubMed Google Scholar). The aim of this study was to characterize the biologic consequences of chronic VEGF overexpression in the epidermis of transgenic mice, using a keratin 14 promoter expression cassette containing the mouse VEGF164 gene to target VEGF expression selectively to basal epidermal keratinocytes. Southern blot analysis of genomic DNA revealed transgene incorporation in nine of 30 mice with copy numbers between three and ≈20 (data not shown). Whereas one of these transgenic mice died and four did not transmit the transgene, four founders with 8–10 transgene copies in their genome transmitted the transgene to their offspring with a Mendelian inheritance pattern. Targeted expression of the K14-VEGF transgene was confirmed by in situ hybridization, demonstrating selectively increased VEGF expression in the basal keratinocyte layer of the epidermis and in the outer root sheath keratinocytes of hair follicles that also express keratin 14 (Figure 2a–f). Within the upper dermis and surrounding hair follicles, microvessels expressed highly increased levels of VEGFR-2 mRNA (Figure 2g–j) and moderately increased levels of VEGFR-1 mRNA (Figure 2k, l). No VEGFR-1 mRNA expression was detectable in wild-type skin (not shown). In contrast, expression of VEGFR-3 (Flt-4) appeared unchanged (data not shown). These findings suggest a positive feedback mechanism, resulting in selective induction of the two high-affinity, endothelial cell VEGF receptors by keratinocyte-derived VEGF in vivo. VEGF transgenic mice were characterized by visibly increased skin vascularization (Figure 3a). Immunohistochemical staining for PECAM-1, an endothelial junction molecule (Figure 3b, c) (Dejana et al., 1995Dejana E. Corada M. Lampugnani M.G. Endothelial cell-to-cell junctions.Faseb J. 1995; 9: 910-918Crossref PubMed Scopus (383) Google Scholar), and for collagen types IV and XVIII revealed an increased number of dermal microvessels within transgenic skin (Figure 3f, g and data not shown). Morphometric analysis of microvascular profiles per area unit, using paraffin sections stained for CD31, showed that microvascular density in wild-type controls was highest during the neonatal period (Figure 4), and was substantially lower after 3 and 6 wk. These findings corresponded to the degree of visible skin vascularization. VEGF transgenic mice were characterized by highly increased numbers of microvessels per high power microscope field in the abdominal skin of 3 d old mice (+58.3% versus control; p < 0.01; n = 6) (Figure 4). In 11 d old, 3 wk old, and 6 wk old mice, the increase was less dramatic (+30.2–31.3% versus control; p < 0.05; n = 6), due to the diminished expression of the transgene construct in the epidermis of older animals (Vassar et al., 1989Vassar R. Rosenberg M. Ross S. Tyner A. Fuchs E. Tissue-specific and differentiation-specific expression of a human K14 keratin gene in transgenic mice.Proc Natl Acad Sci USA. 1989; 86: 1563-1567Crossref PubMed Scopus (291) Google Scholar; data not shown). Comparable results were obtained in dorsal and ear skin. The induced dermal microvessels in VEGF transgenic mice were not stained by monoclonal antibodies against desmoplakin I and II (data not shown) expressed by lymphatic, but not by blood vascular endothelial cells (Schmelz et al., 1994Schmelz M. Moll R. Kuhn C. Franke W.W. Complexus adhaerentes, a new group of desmoplakin-containing junctions in endothelial cells: II. Different types of lymphatic vessels.Differentiation. 1994; 57: 97-117Crossref PubMed Scopus (99) Google Scholar). This indicated that the induced dermal microvasculature was of blood vessel rather than lymphatic origin.Figure 4Increased density of microvessels underlying the epidermal–dermal basement membrane in the abdominal skin of VEGF transgenic mice (black bars), as compared with wild-type controls (white bars). Microvascular density was measured by the average number of microvascular profiles per high-power field. Mean ± SD (n = 6 for each group). *p < 0.05;*p < 0.01.View Large Image Figure ViewerDownload (PPT) Angiogenesis and increased mast cell localization to sites of newly forming microvasculature are often associated (Meininger and Zetter, 1992Meininger C.J. Zetter B.R. Mast cells and angiogenesis.Cancer Biol. 1992; 3: 73-77PubMed Google Scholar). When Epon sections of skin of VEGF transgenic mice were stained for chloroacetate esterase activity (Figure 3d, e) or were Giemsa stained, a significantly increased density of dermal mast cells was detected (+35.2% versus wild-type littermates, p < 0.05, n = 6). These results suggest a role of VEGF in mast cell recruitment to sites of angiogenesis in vivo. In contrast, skin-infiltrating leukocytes were not increased in VEGF transgenic mice, as assessed by immunohistochemical staining for CD3, CD4, CD8, and CD45 expressed by lymphocytes and CD11b (αM-integrin) expressed by granulocytes and macrophages. To functionally characterize the induced dermal blood vessels in VEGF transgenic mice, intravital microscopy of ear skin was performed after intravenous injection of fluorescently labeled dextran (Fukumura et al., 1995Fukumura D. Salehi H.A. Witwer B. Tuma R.F. Melder R.J. Jain R.K. Tumor necrosis factor alpha-induced leukocyte adhesion in normal and tumor vessels: effect of tumor type, transplantation site, and host strain.Cancer Res. 1995; 55: 4824-4829PubMed Google Scholar). Diameter, red blood cell velocity, and shear rate of postcapillary venules were similar in VEGF transgenic and wild-type mice (Figure 5g, h); however, in accordance with the data obtained by morphometric analysis of CD31 stained skin sections, the microvascular density was significantly increased in the skin of 3 wk old transgenic mice (+20.1% versus control) (Figure 5a, b, i). These capillaries had a highly increased tortuosity index (+69.0%), corresponding to a decreased average capillary length without branching (Figure 5j). In addition, capillaries in VEGF transgenic ear skin were hyperpermeable, as evidenced by early leakage of fluoroscein isothiocyanate-dextran into the perivascular space (Figure 5b and data not shown). In contrast, microlymphangiography demonstrated normal numbers and diameters of lymphatic vessels in transgenic skin (Figure 5e, f). Postcapillary venules in VEGF transgenic ear skin showed a significantly increased proportion of rolling leukocytes (47.9% versus 15.6% in wild-type skin, p < 0.005) (Figure 5k) and, even more prominent, an enhanced density of adherent leukocytes (251.9 versus 61.3 cells per mm2, p < 0.005) (Figure 5c, d, l). The enhanced leukocyte rolling and adhesion was not due to increased numbers of circulating leukocytes, as white blood cells counts were comparable in wild-type and transgenic mice. Blocking antibodies against E- and P-selectin brought the proportion of rolling and adherent leukocytes back down to levels seen in wild-type mice (Figure 6a). In contrast, an anti-ICAM-1/lymphocyte function-associated antigen-1 antibody combination had no effect on leukocyte adhesion; however, no pronounced alterations of endothelial cell expression levels of E- or P-selectin were detected within the skin of VEGF transgenic mice by immunohistochemical staining (data not shown). Leukocyte adhesion in transgenic mouse skin was potently inhibited by antibody combinations of anti-VCAM-1/VLA-4 and of anti-VCAM-1/ICAM-1 (Figure 6). This suggested that the increased rolling and adhesion of peripheral blood leukocytes in postcapillary venules of VEGF transgenic mice resulted from specific cell adhesion molecule interactions. Previously, we identified VEGF as