Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis

VPF/VEGF is a multifunctional cytokine that contributes to angiogenesis by both direct and indirect mechanisms. On the one hand, VPF/VEGF stimulates the ECs lining nearby microvessels to proliferate, to migrate, and to alter their pattern of gene expression. On the other hand, VPF/VEGF renders these same microvascular ECs hyperpermeable so that they spill plasma proteins into the extravascular space, leading to the clotting of extravasated fibrinogen with deposition of a fibrin gel. Extravascular fibrin serves as a provisional matrix that favors and supports the ingrowth of new blood vessels and other mesenchymal cells that generate mature, vascularized stroma. These same principles apply in tumors, in several examples of non-neoplastic pathology, and in physiological processes that involve angiogenesis and new stroma generation. In all of these examples, microvascular hyperpermeability and the introduction of a provisional, plasma-derived matrix precede and accompany the onset of EC division and new blood vessel formation. It would seem, therefore, that tumors have "borrowed" fundamental mechanisms that developed in multicellular organisms for purposes of tissue defense, renewal, and repair. VPF/VEGF, therefore has taught us something new about angiogenesis; namely, that vascular hyperpermeability and consequent plasma protein extravasation are important, perhaps essential, elements in its generation. However, this finding raises a paradox. While VPF/VEGF induces vascular hyperpermeability, other potent angiogenic factors apparently do not, at least in subtoxic concentrations that are more than sufficient to induce angiogenesis. Nonetheless, wherever angiogenesis has been studied, the newly generated vessels have been found to be hyperpermeable. How, therefore, do angiogenic factors other than VPF/VEGF lead to the formation of new and leaky blood vessels? We do not as yet have a complete answer to this question. One possibility is that at least some angiogenic factors mediate their effect by inducing or stimulating the expression of VPF/VEGF. In fact, there is already one clear example of this. TGF-alpha is a potent angiogenic factor but does not itself increase microvascular permeability. However, TGF-alpha strikingly upregulates VPF/VEGF expression in cultured keratinocytes and is thought to be responsible, at least in part, for the overexpression of VPF/VEGF in psoriasis. Moreover, overexpression of TGF-alpha, along with that of the EGF receptor with which it interacts, is characteristic of many malignant tumors, raising the possibility that TGF-alpha acts to stimulate VPF/VEGF expression in other types of epithelial cells and in this manner induces angiogenesis.(ABSTRACT TRUNCATED AT 400 WORDS)

Vascular permeability factor/vascular endothelial growth factor: an important mediator of angiogenesis in malignancy and inflammation

Vascular permeability factor (VPF), also known as vascular endothelial growth factor (VEGF), is a multifunctional cytokine that is overexpressed in many transplantable animal and autochtonous human cancers, in healing wounds, and in chronic inflammatory disorders such as psoriasis and rheumatoid arthritis. All of these entities are characterized by angiogenesis, altered extracellular matrix, and variable degrees of hypoxia. In addition, two VPF/VEGF receptors, flt-1 and kdr, are overexpressed by endothelial cells that line the microvessels that supply these tumors/inflammatory reactions. On the basis of these and other data, we have proposed a model of angiogenesis in which VPF/VEGF plays a central role: this model is applicable to tumors and also to the angiogenesis that occurs in non-neoplastic processes.

Increased expression of vascular permeability factor (vascular endothelial growth factor) in bullous pemphigoid, dermatitis herpetiformis, and erythema multiforme

Vascular permeability factor (VPF), also known as vascular endothelial growth factor (VEGF), plays an important role in the increased vascular permeability and angiogenesis associated with many malignant tumors. In addition, VPF/VEGF is strongly expressed by epidermal keratinocytes in wound healing and psoriasis, disorders that are also characterized by increased microvascular permeability and angiogenesis. In this study, we investigated the expression of VPF/VEGF in three bullous diseases with subepidermal blister formation that are characterized by hyperpermeable dermal microvessels and pronounced papillary dermal edema. The expression of VPF/VEGF mRNA was strongly up-regulated in the lesional epidermis of bullous pemphigoid (n = 3), erythema multiforme (n = 3), and dermatitis herpetiformis (n = 4) as detected by in situ hybridization. Epidermal labeling was particularly intense over blisters, but strong expression was also noted in areas of the epidermis adjacent to dermal inflammatory infiltrates at a distance from blisters. Moreover, the VPF/VEGF receptors, flt-1 and KDR, were up-regulated in endothelial cells in superficial dermal microvessels. High levels of VPF/VEGF (138-238 pM) were detected in blister fluids obtained from five patients with bullous pemphigoid. Addition of blister fluid to human dermal microvascular endothelial cells exerted a dose-dependent mitogenic effect that was suppressed after depletion of VPF/VEGF by immunoadsorption. These findings strongly suggest that VPF/VEGF plays an important role in the induction of increased microvascular permeability in bullous diseases, leading to papillary edema and fibrin deposition and contributing to the bulla formation characteristic of these disorders.

Ultrastructural localization of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) to the abluminal plasma membrane and vesiculovacuolar organelles of tumor microvascular endothelium

Vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) is a cytokine secreted by many animal and human tumors, activated macrophages, keratinocytes, rheumatoid synovial cells, embryonic tissues, and by cultured epithelial and mesenchymal cell lines. It acts selectively on vascular endothelial cells to increase their permeability to circulating macromolecules and to stimulate their replication. Although not detectably expressed by vascular cells in the human and animal tumors we have studied, VPF/VEGF accumulates in the microvessels supplying tumors and certain inflammatory reactions in which VPF/VEGF is also overexpressed. Light microscopic immunohistochemistry lacked the resolution necessary to localize VPF/VEGF precisely in such vessels. Therefore, we used a pre-embedding immunocytochemical method to localize VPF/VEGF at the ultrastructural level in the new blood vessels that are elicited in the peritoneal walls of mice bearing a transplantable mouse ascites tumor of ovarian origin. Intense immunostaining for VPF/VEGF was observed on the abluminal plasma membrane of tumor-associated microvascular endothelial cells and in vesiculovacuolar organelles (VVOs) present in these same endothelial cells. (VVOs are recently described cytoplasmic organelles present in tumor vascular endothelium that provide an important pathway for extravasation of circulating macromolecules.) In contrast to labeling of the abluminal plasma membrane and VVO vesicles and vacuoles, endothelial cytoplasmic organelles, such as multivesicular bodies and Weibel-Palade bodies, and the underlying basal lamina, did not stain with antibodies to VPF/VEGF. The distribution of VPF/VEGF here described corresponds to that anticipated for high-affinity VFP/VEGF receptors, although binding of VPF/VEGF to other endothelial cell surface structures, such as plasma membrane proteoglycans, is also a possibility.

Overexpression of vascular permeability factor (VPF/VEGF) and its endothelial cell receptors in delayed hypersensitivity skin reactions

Delayed hypersensitivity (DH) is a T cell-mediated form of immune response characterized by a predominantly perivascular, mononuclear cell infiltrate. The venules in DH reactions are hyperpermeable to plasma proteins, leading to extravasation of plasma fibrinogen and its extravascular clotting to form a fibrin gel that promotes induration and angiogenesis. The mechanisms responsible for microvascular hyperpermeability in DH are unknown. Recently, a cytokine named vascular permeability factor (VPF, also known as vascular endothelial growth factor or VEGF) has been implicated in the chronic vascular hyperpermeability and angiogenesis of solid and ascites tumors, healing wounds, rheumatoid arthritis, and psoriasis. These findings suggested that VPF/VEGF might also have a role in the pathogenesis of DH. Two model systems were studied: allergic contact dermatitis to poison ivy in human volunteers and classical tuberculin hypersensitivity in rats. In both, in situ hybridization revealed that the mRNAs encoding VPF/VEGF were strikingly overexpressed in keratinocytes of the epidermis; scattered mononuclear cells infiltrating the dermis also overexpressed VPF/VEGF mRNA, to a greater extent in rat tuberculin than in human contact reactions. In contact reactions, mRNAs for two VPF/VEGF vascular endothelial cell receptors, flt-1 and KDR, were also strikingly overexpressed. Abundant fibrin deposition in both models confirmed that dermal microvessels were indeed hyperpermeable to plasma fibrinogen. These results implicate VPF/VEGF as a potentially important mediator in the pathogenesis of cell-mediated immunity and provide further evidence that products of epithelial cells may regulate the inflammatory response.

Overexpression of vascular permeability factor (VPF/VEGF) and its endothelial cell receptors in delayed hypersensitivity skin reactions

Delayed hypersensitivity (DH) is a T cell-mediated form of immune response characterized by a predominantly perivascular, mononuclear cell infiltrate. The venules in DH reactions are hyperpermeable to plasma proteins, leading to extravasation of plasma fibrinogen and its extravascular clotting to form a fibrin gel that promotes induration and angiogenesis. The mechanisms responsible for microvascular hyperpermeability in DH are unknown. Recently, a cytokine named vascular permeability factor (VPF, also known as vascular endothelial growth factor or VEGF) has been implicated in the chronic vascular hyperpermeability and angiogenesis of solid and ascites tumors, healing wounds, rheumatoid arthritis, and psoriasis. These findings suggested that VPF/VEGF might also have a role in the pathogenesis of DH. Two model systems were studied: allergic contact dermatitis to poison ivy in human volunteers and classical tuberculin hypersensitivity in rats. In both, in situ hybridization revealed that the mRNAs encoding VPF/VEGF were strikingly overexpressed in keratinocytes of the epidermis; scattered mononuclear cells infiltrating the dermis also overexpressed VPF/VEGF mRNA, to a greater extent in rat tuberculin than in human contact reactions. In contact reactions, mRNAs for two VPF/VEGF vascular endothelial cell receptors, flt-1 and KDR, were also strikingly overexpressed. Abundant fibrin deposition in both models confirmed that dermal microvessels were indeed hyperpermeable to plasma fibrinogen. These results implicate VPF/VEGF as a potentially important mediator in the pathogenesis of cell-mediated immunity and provide further evidence that products of epithelial cells may regulate the inflammatory response.

Pathogenesis of ascites tumor growth: vascular permeability factor, vascular hyperpermeability, and ascites fluid accumulation

Previous studies have shown that accumulation of tumor ascites fluid results in large part from increased permeability of peritoneal lining vessels (Nagy et al., Cancer Res., 49: 5449-5458, 1989; Nagy et al., Cancer Res., 53: 2631-2643, 1993). However, the specific microvessels rendered hyperpermeable have not been identified nor has the basis of peritoneal vascular hyperpermeability been established. To address these questions, TA3/St and MOT carcinomas, well-characterized transplantable murine tumors that grow in both solid and ascites form, were studied as model systems. Ascites tumor cells of either type were injected i.p. into syngeneic A/Jax and C3Heb/FeJ mice, and ascites fluid and plasma were collected at intervals thereafter up to 8 and 28 days, respectively. Beginning several days after tumor cell injection, small blood vessels located in tissues lining the peritoneal cavity (mesentery, peritoneal wall, and diaphragm) became hyperpermeable to several macromolecular tracers (125I-human serum albumin, FITC-dextran, colloidal carbon, and Monastral Blue B). Increased microvascular permeability correlated with the appearance in ascites fluid of vascular permeability factor (VPF), a tumor cell-secreted mediator that potently enhances vascular permeability to circulating macromolecules. VPF was measured in peritoneal fluid by both a functional bioassay and a sensitive immunofluorometric assay. The VPF concentration, total peritoneal VPF, ascites fluid volume, tumor cell number, and hyperpermeability of peritoneal lining microvessels were found to increase in parallel over time. The close correlation of peritoneal fluid VPF concentration with the development of hyperpermeable peritoneal microvessels in these two well-defined ascites tumors suggests that VPF secretion by tumor cells is responsible, in whole or in part, for initiating and maintaining the ascites pattern of tumor growth.

Pathogenesis of ascites tumor growth: angiogenesis, vascular remodeling, and stroma formation in the peritoneal lining

In the accompanying papers, we demonstrated that two murine ascites tumors (MOT and TA3/St) induced peritoneal lining blood vessels to become hyperpermeable to plasma proteins, leading to extravasation of fibrinogen and its clotting to cross-linked fibrin in peritoneal lining tissues (peritoneal wall, mesentery, and diaphragm). In solid tumors, vascular hyperpermeability and fibrin deposition lead to the generation of vascularized connective tissue. In order to determine whether fibrin had similar consequences in ascites tumors, the vasculature and stroma of peritoneal lining tissues were analyzed at successive intervals after i.p. tumor cell injection. In both MOT and TA3/St ascites tumors, the size and number of peritoneal lining microvessels increased significantly by 5-8 days. Subsequently, peritoneal lining vessels increased in cross-sectional area by as much as 15-fold and peritoneal vascular frequency increased by up to 11-fold. Incorporation of [3H]thymidine by mesenteric blood vessels was negligible in control animals but came to involve 20 and 40% of endothelial cells lining mesenteric vessels in MOT and TA3/St ascites tumor-bearing mice, respectively. After an early dramatic increase in cross-sectional area, peritoneal lining microvessels subsequently underwent a novel form of remodeling to smaller average size as the result of transvascular bridging by endothelial cell cytoplasmic processes. Thus, both of the ascites tumors studied here induced angiogenesis and stroma similar to that elicited when these same tumors were grown in solid form. However, stroma developed more slowly in ascites than in solid tumors and was entirely confined to a compartment (peritoneal lining tissues) that was distinct from that (peritoneal cavity) containing the majority of tumor cells and ascites fluid. These findings are consistent with the hypothesis that vascular hyperpermeability, induced in both solid and ascites tumors by tumor cell-secreted vascular permeability factor, is a common early step in tumor angiogenesis, resulting in fibrinogen extravasation, fibrin deposition, and likely other alterations of the extracellular matrix that together stimulate new vessel and fibroblast ingrowth.

Pathogenesis of ascites tumor growth: fibrinogen influx and fibrin accumulation in tissues lining the peritoneal cavity

In the immediately preceding paper, we demonstrated that the microvasculature supplying peritoneal lining tissues of mice bearing either of two transplantable ascites carcinomas was hyperpermeable to circulating macromolecules. Solid tumors have been shown to exhibit similar levels of microvascular hyperpermeability, leading to extravasation of plasma proteins, including fibrinogen which clots on extravasation to form an extravascular fibrin gel. To determine whether similar extravasation and clotting of plasma fibrinogen occurred in ascites tumors, we used 125I-labeled fibrinogen (125I-F) as a tracer to measure inflow of fibrinogen into the peritoneal cavities, and influx and accumulation of fibrinogen/fibrin in the peritoneal lining tissues (peritoneal wall, mesentery, and diaphragm) of mice bearing syngeneic TA3/St or MOT ascites tumors. The percentage of circulating 125I-F that extravasated into the peritoneal cavity was increased from 10- to 50-fold in mice bearing either ascites tumor. Influx into the peritoneal walls of ascites tumor-bearing mice was 3-7 times that of control mice and became maximal on day 8 (TA3/St) and day 15 (MOT). Accumulation of 125I-F in ascites fluid and peritoneal lining tissues was also increased substantially in mice bearing these ascites tumors, reaching maximal values on days 7-8 (TA3/St) and 19-29 (MOT) at levels 2- to 3-fold (peritoneal wall) and 33- to 148-fold (ascites fluid) above control levels. Significant amounts of the 125I-F that accumulated in the peritoneal lining tissues of ascites tumor-bearing animals were insoluble in 3 M urea, consistent with clotting of 125I-F to cross-linked fibrin. Autoradiographs of SDS-PAGE gels performed on extracts of peritoneal lining tissues of both ascites tumors revealed the characteristic signature of cross-linked fibrin, i.e., gamma-gamma dimers and alpha-polymers. Fibrin was also identified in peritoneal lining tissues of both ascites tumors by immunohistochemistry. Taken together, these data indicate that fibrinogen, like other circulating macromolecules, extravasates into the peritoneal cavity and peritoneal lining tissues of ascites tumor-bearing mice and does so with kinetics similar to those of other macromolecular tracers we have studied. Moreover, a portion of the fibrinogen that extravasated into peritoneal lining tissues clotted to form a cross-linked fibrin meshwork which trapped tumor cells and favored their attachment to the peritoneal surface. By analogy with solid tumors, such fibrin deposits may also be expected to have a role in initiating angiogenesis and the generation of mature tumor stroma.

Expression of vascular permeability factor/vascular endothelial growth factor by human granulosa and theca lutein cells. Role in corpus luteum development

Vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) is a cytokine that is overexpressed in many tumors, in healing wounds, and in rheumatoid arthritis. VPF/VEGF is thought to induce angiogenesis and accompanying connective tissue stroma in two ways: 1), by increasing microvascular permeability, thereby modifying the extracellular matrix and 2), as an endothelial cell mitogen. VPF/VEGF has been reported in animal corpora lutea and we investigated the possibility that it might be present in human ovaries and have a role in corpus luteum formation. We here report that VPF/VEGF mRNA and protein are expressed by human ovarian granulosa and theca cells late in follicle development and, subsequent to ovulation, by granulosa and theca lutein cells. Therefore, VPF/VEGF is ideally positioned to provoke the increased permeability of thecal blood vessels that occurs shortly before ovulation. VPF/VEGF likely also contributes to the angiogenesis and connective tissue stroma generation that accompany corpus luteum/corpus albicans formation. Finally, VPF/VEGF was overexpressed in the hyperthecotic ovarian stroma of Stein-Leventhal syndrome in which it may also have a pathophysiological role.

Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in breast cancer

Solid tumors must induce a vascular stroma to grow beyond a minimal size, and the intensity of the angiogenic response has been correlated with prognosis in breast cancer patients. Vascular permeability factor (VPF), also known as vascular endothelial growth factor (VEGF), is a secreted protein that has been implicated in tumor-associated angiogenesis. Vascular permeability factor directly stimulates endothelial cell growth and also increases microvascular permeability, leading to the extravasation of plasma proteins, which alter the extracellular matrix in a manner that promotes angiogenesis. To determine whether VPF has a role in breast cancer, we used in situ hybridization to study VPF mRNA expression in normal breast tissue (13 specimens), comedo-type ductal carcinoma in situ (DCIS) (four specimens), infiltrating ductal carcinoma (12 specimens), infiltrating lobular carcinoma (two specimens), metastatic ductal carcinoma (three specimens) and metastatic lobular carcinoma (one specimen). Vascular permeability factor mRNA was expressed at a low level by normal duct epithelium but was expressed at high levels in tumor cells in all cases of comedo-type DCIS, infiltrating ductal carcinoma, and metastatic ductal carcinoma. In contrast, VPF mRNA was not expressed at high levels in infiltrating lobular carcinoma. We also used in situ hybridization to study the expression of two recently described endothelial cell surface VPF receptors, flt-1 and kdr. Vascular permeability factor receptor mRNA was strongly expressed in endothelial cells of small vessels adjacent to malignant tumor cells in DCIS, infiltrating ductal carcinoma, and metastatic ductal carcinoma. In contrast, no definite labeling for receptor mRNA was detected in infiltrating lobular carcinoma or nonmalignant breast tissue. The intense expression of VPF mRNA by breast carcinoma cells and of VPF receptor mRNA by endothelial cells of adjacent small blood vessels provides strong evidence linking VPF expression to the angiogenesis associated with comedo-type DCIS, infiltrating ductal, and metastatic ductal breast carcinoma.

Vascular endothelial growth factor/vascular permeability factor is temporally and spatially correlated with ocular angiogenesis in a primate model

Ischemia often precedes neovascularization. In ocular neovascularization, such as occurs in diabetic retinopathy, a diffusible angiogenic factor has been postulated to be produced by ischemic retina and to lead to neovascularization of the retina, optic nerve, or iris. However, no angiogenic factor has been conclusively identified that satisfies this hypothesis. Vascular endothelial growth factor/vascular permeability factor, hereafter referred to as VEGF, is a likely candidate for an ocular angiogenic factor because it is a secreted mitogen, specific for endothelial cells, and is upregulated by hypoxia. We investigated the association of VEGF with the development of experimental iris neovascularization in the cynomolgus monkey. Following the production of retinal ischemia by laser occlusion of all branch retinal veins, VEGF was increased in the aqueous fluid, and the aqueous VEGF levels changed synchronously and proportionally with the severity of iris neovascularization. Northern analysis and in situ hybridization revealed that VEGF messenger RNA is upregulated in the ischemic retina. These observations support the hypothesis that ocular neovascularization is regulated by a diffusible factor and identify VEGF as a likely candidate for a retina-derived vascular permeability and angiogenesis factor in vivo.

Osteopontin expression and distribution in human carcinomas

Osteopontin (OPN), a secreted adhesive glycoprotein, is significantly overexpressed in a variety of experimental models of malignancy. Moreover, increased levels of OPN have been detected in the blood of patients with metastatic carcinoma. To investigate OPN expression and distribution in human carcinomas directly, we studied a wide variety of common tumors by Northern analysis, in situ hybridization, and immunohistochemistry. All 14 tumors studied by Northern analysis showed very substantial increases in OPN messenger (m)RNA when compared to corresponding normal tissues. Moreover, intense labeling for OPN mRNA was detected in 71 of 76 carcinomas studied by in situ hybridization. In most of the carcinomas studied (colon, stomach, duodenum, pancreas, breast, lung, bladder, prostate, ovary, thyroid, and melanoma), tumor cells did not label detectably for OPN mRNA; however, macrophages intimately associated with tumor cells labeled strongly for the OPN transcript. In carcinomas of the kidney and endometrium, both tumor cells and host macrophages labeled strongly for OPN mRNA. The presence of OPN mRNA in macrophages was particularly pronounced at the edge of tumors (ie, the tumor/stroma interface) and in areas of tumor necrosis. Although in most cases tumor cells did not label detectably for OPN mRNA, both tumor cells and macrophages stained for OPN protein, suggesting that OPN secreted by macrophages may bind to tumor cells, possibly through the glycine-arginine-glycine-aspartate-serine cell binding domain in OPN. Collectively, these data suggest that OPN functions in adhesive interactions at the tumor/host interface and thereby may influence processes such as invasion and metastasis.

Overexpression of vascular permeability factor/vascular endothelial growth factor and its receptors in psoriasis

Psoriatic skin is characterized by microvascular hyperpermeability and angioproliferation, but the mechanisms responsible are unknown. We report here that the hyperplastic epidermis of psoriatic skin expresses strikingly increased amounts of vascular permeability factor (VPF; vascular endothelial growth factor), a selective endothelial cell mitogen that enhances microvascular permeability. Moreover, two VPF receptors, kdr and flt-1, are overexpressed by papillary dermal microvascular endothelial cells. Transforming growth factor alpha (TGF-alpha), a cytokine that is also overexpressed in psoriatic epidermis, induced VPF gene expression by cultured epidermal keratinocytes. VPF secreted by TGF-alpha-stimulated keratinocytes was bioactive, as demonstrated by its mitogenic effect on dermal microvascular endothelial cells in vitro. Together, these findings suggest that TGF-alpha regulates VPF expression in psoriasis by an autocrine mechanism, leading to vascular hyperpermeability and angiogenesis. Similar mechanisms may operate in tumors and in healing skin wounds which also commonly express both VPF and TGF-alpha.

Vascular permeability factor/endothelial growth factor (VPF/VEGF): accumulation and expression in human synovial fluids and rheumatoid synovial tissue

Vascular permeability factor (VPF, also known as vascular endothelial growth factor or VEGF), is a potent microvascular permeability enhancing cytokine and a selective mitogen for endothelial cells. It has been implicated in tumor angiogenesis and ascites fluid accumulation. Since development of the destructive synovial pannus in rheumatoid arthritis (RA) is associated with changes in vascular permeability (synovial fluid accumulation), synovial cell hyperplasia, and angiogenesis, we examined synovial fluids (SFs) and joint tissue for the expression and local accumulation of VPF/VEGF. VPF/VEGF was detected in all of 21 synovial fluids examined and when measured by an immunofluorimetric assay, ranged from 6.9 to 180.5 pM. These levels are biologically significant, since < 1 pM VPF/VEGF can elicit responses from its target cells, endothelial cells. Levels of VPF/VEGF were highest in rheumatoid arthritis fluids (n = 10), with a mean value (+/- SEM) of 59.1 +/- 18.0 pM, vs. 21.4 +/- 2.3 pM for 11 SFs from patients with other forms of arthritis (p = 0.042). In situ hybridization studies that were performed on joint tissues from patients with active RA revealed that synovial lining macrophages strongly expressed VPF/VEGF mRNA, and that microvascular endothelial cells of nearby blood vessels strongly expressed mRNA for the VPF/VEGF receptors, flt-1 and KDR. Immunohistochemistry performed on inflamed rheumatoid synovial tissue revealed that the VPF/VEGF peptide was localized to macrophages within inflamed synovium, as well as to microvascular endothelium, its putative target in the tissue. Together, these findings indicate that VPF/VEGF may have an important role in the pathogenesis of RA.

Compartmental distribution of tumor-specific monoclonal antibodies in human melanoma xenografts

Monoclonal antibodies (MAb) are attractive for tumor therapy because of their exquisite specificity. Although a majority of tumor cells in small (< or = 20 mg) solid tumors can be labeled following systemic administration of antitumor cell MAbs, little quantitative information is available as to the distribution of these MAbs within the several compartments that comprise solid tumors. Our goal was to provide such data in a well-characterized melanoma xenograft system. In accord with earlier work, i.v.-injected, melanoma-specific MAbs 436 and IND1, directed, respectively, against the 125 kD and HMW-melanoma-associated antigens, accumulated in M21 and SK-MEL-2 tumor xenografts in amounts of approximately 20% of injected dose/g. However, only 20-24% of the MAbs present in tumor xenografts was bound to tumor cells; the great majority (76-80%) was in the tumor extracellular fluid (ECF) and collagenous residue fractions. These results could not be accounted for by MAb degradation or release of MAbs from tumor cells during xenograft dissociation. Rather, they reflected in large part interactions of MAbs with antigens which tumors had shed into the ECF. Thus, 48 h after i.v. injection of 20 micrograms of melanoma-specific, biotin-tagged MAb, 46-66% of that present in the tumor ECF was complexed with melanoma-associated antigens. Overall, 61-73% of the MAbs recovered from tumor xenografts were bound to tumor antigens (either to tumor cells themselves or to tumor-shed antigens). In contrast, only approximately 4% of a melanoma-nonspecific MAb (B72.3) accumulated per g tumor after i.v. injection and nearly all of this was free in the ECF. Consistent with these data, fluorescence microscopy revealed that i.v.-injected, fluorescein-tagged MAbs achieved highest concentrations in tumor stroma, particularly at the tumor-host interface. Flow cytometry of dissociated solid tumors revealed that both the fraction of MAb-labeled tumor cells and the amount of MAb/tumor cell could be increased by increasing the administered i.v. dose of melanoma-specific MAb. Nonetheless, even at the highest i.v. injected dose (300 micrograms), 15-37% of tumor cells lacked detectable MAb labeling. Taken together, the data indicate that delivery of tumor cell-specific MAbs to solid tumors cannot be equated with their delivery to tumor cells. This distinction is important for immunotherapeutic approaches that require MAb contact with tumor cells.

Ultrastructural immunogold localization of osteopontin in human gallbladder epithelial cells

We used a post-embedding ultrastructural immunogold method to localize osteopontin in human gallbladder epithelial cells. This glycoprotein, originally described in bone but recently found to have a much wider distribution in many epithelia and in some mesenchymal cells, was present in the filamentous glycocalyx, small apical cytoplasmic smooth membrane-bound vesicles, large membrane-bound cytoplasmic granules, and in portions of the Golgi complex in gallbladder columnar epithelial cells. These findings suggest that newly synthesized osteopontin is packaged in Golgi-derived granules that release their contents by classical exocytosis from the cell surface. At least a portion of secreted osteopontin remains on the cell surface, where it becomes integrated into the filamentous glycocalyx coating the luminal surface of gallbladder epithelial cells.

Increased expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in kidney and bladder carcinomas

Vascular permeability factor (VPF), also known as vascular endothelial growth factor, is a secreted protein implicated in tumor-associated microvascular hyperpermeability and angiogenesis. Tumor cells in 11 of 12 renal cell carcinomas expressed high levels of VPF messenger RNA (mRNA) by in situ hybridization, the only exception being a case of the relatively avascular papillary variant. Expression was further accentuated adjacent to areas of necrosis. Both tumor cells and endothelial cells in small vessels adjacent to tumor stained strongly for VPF protein by immunohistochemistry. Endothelial cells did not express detectable VPF mRNA, but did express high levels of mRNA for the VPF receptors flt-1 and KDR indicating that the endothelial cell staining likely reflects binding of VPF secreted by adjacent tumor cells. Three transitional cell carcinomas also labeled strongly for VPF mRNA. These data suggest an important role for VPF in the vascular biology of these two common human malignancies.

Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in adenocarcinomas of the gastrointestinal tract

Vascular permeability factor (VPF) is one of the most potent known inducers of microvascular hyperpermeability; in addition, it is a selective endothelial cell growth factor, hence its alternate name, vascular endothelial growth factor. VPF exerts its actions on the microvasculature by interacting with specific endothelial cell receptors. VPF is expressed by many transplantable animal tumors, by tumor cell lines in culture, and by certain normal cells in situ. The purpose of the present investigation was to determine whether and with what consistency VPF and its endothelial cell receptors are expressed in primary autochthonous human tumor of gastrointestinal tract origin, as determined by in situ hybridization and immunohistochemistry. Twenty-one primary adenocarcinomas (17 colon, 2 stomach, 1 small bowel, and 1 pancreas) were studied. The malignant epithelial cells expressed VPF mRNA strongly, in contrast to normal epithelium, hyperplastic polyps, and adenomas, which expressed little or no VPF mRNA. VPF expression was further increased in tumor cells immediately adjacent to zones of tumor necrosis; in such areas, occasional stromal cells also expressed VPF mRNA. In the ten colon carcinomas studied, tumor cells stained for VPF protein by immunohistochemistry. The endothelial cells of nearby stromal blood vessels also stained for VPF by immunohistochemistry and in addition expressed mRNAs encoding the VPF receptors flt-1 and kdr as determined by in situ hybridization. Endothelial cells away from the tumor did not stain for VPF and no definite mRNA expression for flt-1 or kdr was detected by in situ hybridization. The ganglion cells of the myenteric plexus of normal bowel expressed VPF mRNA and protein. These data indicate that primary autochthonous human tumors of gastrointestinal origin regularly express both VPF mRNA and VPF protein and that adjacent stromal vessels express mRNAs for both known VPF receptors. VPF is likely to contribute to tumor growth by promoting angiogenesis and stroma formation, both directly, through its action as an endothelial cell growth factor, and indirectly, by increasing vascular permeability, thereby leading to plasma protein extravasation, fibrin deposition, and the eventual replacement of the resulting matrix with vascularized stroma.

Vascular permeability factor (VPF, VEGF) in tumor biology

Vascular permeability factor (VPF), also known as vascular endothelial growth factor (VEGF), is a multifunctional cytokine expressed and secreted at high levels by many tumor cells of animal and human origin. As secreted by tumor cells, VPF/VEGF is a 34-42 kDa heparin-binding, dimeric, disulfide-bonded glycoprotein that acts directly on endothelial cells (EC) by way of specific receptors to activate phospholipase C and induce [Ca2+]i transients. Two high affinity VPF/VEGF receptors, both tyrosine kinases, have thus far been described. VPF/VEGF is likely to have a number of important roles in tumor biology related, but not limited to, the process of tumor angiogenesis. As a potent permeability factor, VPF/VEGF promotes extravasation of plasma fibrinogen, leading to fibrin deposition which alters the tumor extracellular matrix. This matrix promotes the ingrowth of macrophages, fibroblasts, and endothelial cells. Moreover, VPF/VEGF is a selective endothelial cell (EC) growth factor in vitro, and it presumably stimulates EC proliferation in vivo. Furthermore, VPF/VEGF has been found in animal and human tumor effusions by immunoassay and by functional assays and very likely accounts for the induction of malignant ascites. In addition to its role in tumors, VPF/VEGF has recently been found to have a role in wound healing and its expression by activated macrophages suggests that it probably also participates in certain types of chronic inflammation. VPF/VEGF is expressed in normal development and in certain normal adult organs, notably kidney, heart, adrenal gland and lung. Its functions in normal adult tissues are under investigation.

Vascular permeability factor (vascular endothelial growth factor) in guinea pig and human tumor and inflammatory effusions

Vascular permeability factor (VPF), also known as vascular endothelial growth factor, is a dimeric M(r) 34,000-42,000 glycoprotein that possesses potent vascular permeability-enhancing and endothelial cell-specific mitogenic activities. It is synthesized by many rodent and human tumor cells and also by some normal cells. Recently we developed a sensitive and specific time-resolved immunofluorometric assay for quantifying VPF in biological fluids. We here report findings with this assay in guinea pigs and patients with both malignant and nonmalignant effusions. Line 1 and line 10 tumor cells were injected into the peritoneal cavities of syngeneic strain 2 guinea pigs, and ascitic fluid, plasma, and urine were collected at various intervals. Within 2 to 4 days, we observed a time-dependent, parallel increase in VPF, ascitic fluid volume, and tumor cell numbers in animals bearing either tumor line; in contrast, VPF was not detected in plasma or urine, even in animals with extensive tumor burdens. However, low levels of VPF were detected in the inflammatory ascites induced by i.p. oil injection. In human studies, high levels of VPF (> 10 pM) were measured in 21 of 32 effusions with cytology-documented malignant cells and in only seven of 35 effusions without cytological evidence of malignancy. Thus, VPF levels in human effusions provided a diagnostic test for malignancy with a sensitivity of 66% and a specificity of 80% (perhaps as high as 97% in that six of the seven cytology-negative patients with VPF levels > 10 pM had cancer as determined by other criteria). As in the animal tumor models, VPF was not detected in serum or urine obtained from patients with or without malignant ascites. Many nonmalignant effusions contained measurable VPF but, on average, in significantly smaller amounts than were found in malignant effusions. VPF levels in such fluids correlated strongly (p = 0.59, P < 0.001) with monocyte and macrophage content. Taken together, these data relate ascitic fluid accumulation to VPF concentration in a well-defined animal tumor system and demonstrate, for the first time, the presence of VPF in human malignant effusions. It is likely that VPF expression by tumor and mononuclear cells contributes to the plasma exudation and fluid accumulation associated with malignant and certain inflammatory effusions. The VPF assay may prove useful for cancer diagnosis as a supplement to cytology, especially in tumors that grow in the pleural lining but not as a suspension in the effusions that they induce.

Pathogenesis of malignant ascites formation: initiating events that lead to fluid accumulation

Initiating events leading to the accumulation of malignant ascites in the peritoneal cavity were investigated in two syngeneic transplantable murine ascites-producing tumors, MOT mouse ovarian tumor and the TA3/St mammary carcinoma. The transport of two tracers, 125I-labeled human serum albumin (125I-HSA) and 51Cr-labeled red blood cells (51Cr-RBC), into and out of the peritoneal cavity was studied at early times after i.p. tumor cell injection, prior to abundant fluid accumulation, and at intervals of 5 to 360 min after i.v. or i.p. tracer injection. Tracer influx and efflux rates were estimated from the mass of tracer passing into or out of the peritoneal cavity following a bolus injection of tracer into either the blood or the peritoneal cavity. Efflux of 125I-HSA from the peritoneal cavity was markedly reduced (3- to 5-fold) within 1 day of i.p. injection of either type of tumor cell. Significantly reduced efflux preceded any increase in tumor cell number and by itself did not induce peritoneal fluid accumulation. 125I-HSA tracer influx from plasma to peritoneal fluid did not increase detectably until 5 to 7 days after tumor cell injection, when the tumor cell number had increased by 10- to 100-fold. Only at relatively late stages of ascites tumor growth, when the flow rate into the peritoneal cavity had increased relative to the flow rate out of the peritoneum, was there net peritoneal fluid accumulation. Thus, increased influx, in addition to impaired efflux, were required for malignant ascites accumulation. Following i.p. injection, the efflux rates of 125I-HSA always exceeded those of 51Cr-RBC, even in ascites tumor-bearing animals. Furthermore, 125I-HSA tracer disappeared from the peritoneal cavity more rapidly than it appeared in the plasma, suggesting that 125I-HSA moves more rapidly through the channels by which 51Cr-RBC egress from the peritoneum (primarily diaphragmatic lymphatics) and/or has access to additional pathways not open to 51Cr-RBC. Finally, flow rates into and out of the blood and peritoneum were used to obtain kinetic parameters that characterized tracer transport: k1, the rate constant for tracer transport from the blood to the peritoneum; k2, the rate constant for tracer transport from the peritoneal cavity to the blood; and k6, the rate constant for tracer transport from the peritoneal cavity to surrounding interstitial tissue.(ABSTRACT TRUNCATED AT 400 WORDS)

Myocardial fibrin deposition in experimental viral myocarditis that progresses to dilated cardiomyopathy

Myocardial fibrosis is a characteristic late feature in cases of viral myocarditis that progress to dilated cardiomyopathy. However, the pathogenesis of the myocardial fibrosis in such cases is unknown. Prior studies have shown that in healing wounds and tumor stroma generation, interstitial fibrin deposition precedes the development of fibrosis. Therefore, interstitial fibrin deposition in the myocardium was investigated in a murine model of myocarditis in which dilated cardiomyopathy develops. Inbred male C3H/He mice inoculated with coxsackievirus B3 were killed 0, 3, 7, 14, 21, 30, and 60 days after infection. Paraffin sections of hearts were stained with hematoxylin-eosin, Masson's trichrome stain, and antibodies to fibrinogen/fibrin by use of an immunoperoxidase technique. Pretreatment of all mice with anticoagulants and antifibrinolytics 5 minutes before death was used to prevent artifactual fibrin deposition and fibrinolysis during tissue manipulation. Tissue fixation in formalin supplemented with acetic acid served to extract non-cross-linked fibrin, fibrinogen, and fibrinogen and fibrin degradation products, thus ensuring that clotted and cross-linked fibrin was the major immunoreactant. Myocardial fibrin deposition and fibrosis were each quantitated by computer-assisted image analysis. Myocardial fibrin deposition first appeared on day 3, was maximal on day 14, and disappeared by day 30. Conversely, myocardial fibrosis was not detectable until day 14 and was maximal at day 60. Thus, as in healing wounds and developing tumor stroma, fibrin deposition preceded fibrosis in this murine model of myocarditis that progresses to dilated cardiomyopathy.

Macrophages and fibroblasts express embryonic fibronectins during cutaneous wound healing

Fibronectins (FN) comprise a family of adhesive glycoproteins that are prominent components of wound healing. These proteins arise by alternative splicing of a single gene transcript at three sites, termed EIIIA, EIIIB, and V. Extravasated plasma FN, which lacks the EIIIA and EIIIB domains, along with fibrin, comprise the "provisional" matrix that forms within minutes of tissue injury. By 2 days after cutaneous excisional wounding in rats, total FN messenger RNA (mRNA) expression is increased locally and dramatically within the surrounding dermis, in the subjacent muscle (panniculus carnosus) and, notably, at the wound margins. Moreover, in contrast to normal skin, 2-day wounds express EIIIA- and EIIIB-containing "embryonic" FN mRNAs. To identify the cells responsible for synthesizing the various FN isoforms, we performed in situ hybridization with probes for the various FN mRNAs. Collagen and lysozyme probes were employed to distinguish fibroblasts from macrophages. At early intervals (2 days) after wounding, macrophages were the principal cells that expressed FN mRNA. Moreover, many of these cells expressed embryonic FN mRNAs. At 7 to 10 days, when the wound defect was maturing, fibroblasts were the major cells synthesizing these embryonic FNs. It is widely accepted that wound macrophages phagocytose debris and provide degradative enzymes and cytokines essential for early stages of tissue repair. Our findings suggest an additional function for wound macrophages--synthesis of embryonic FNs providing an extracellular matrix that facilitates wound repair, perhaps by promoting cell migration.

Inhibition of vascular permeability factor (vascular endothelial growth factor) with antipeptide antibodies

Vascular permeability factor (VPF), also known as vascular endothelial cell growth factor (VEGF), is a 34- to 43-kDa dimeric protein synthesized and secreted by a variety of tumor and normal cells. At nanomolar concentrations, VPF causes an increase in microvascular permeability and is thought to be responsible for enhanced permeability of tumor blood vessels and for the fluid accumulation associated with solid and ascites tumors. In addition, VPF/VEGF is a mitogen for endothelial cells and may play an important role in maintaining vascular endothelium and in promoting tumor angiogenesis. Antibodies were raised against a series of synthetic peptides derived from the predicted human VPF amino acid sequence. The antibodies were assayed for their ability to bind native and denatured/reduced VPF. Antibodies to peptides from the N- and C-termini bound both denatured/reduced and native VPF; antibodies directed to internal segments (e.g., amino acids 27-48 and 85-101) strongly bound denatured/reduced VPF but were substantially less effective at binding native VPF. These results suggest that the N- and C-termini are exposed regions of the protein in solution. Individually, antibodies to the N- and C-termini each partially blocked VPF permeability activity, and, in combination, blocked nearly 100% of this activity. Also, the N- and C-terminal antibodies blocked the VPF-mediated stimulation of both endothelial cell growth and increase in free cytosolic calcium.

Fibroblast migration in fibrin gel matrices

In healing wounds and many solid tumors, locally increased microvascular permeability results in extravasation of fibrinogen and its extravascular coagulation to form a fibrin gel, with concomitant covalent cross-linking of fibrin by factor XIIIa. Subsequently, inflammatory cells, fibroblasts, and endothelial cells migrate into the gel and organize it into granulation tissue and later into mature collagenous connective tissue. To gain insight into some of the cell migration events associated with these processes, we developed a quantitative in vitro assay that permits the study of fibroblast migration in fibrin gels. Early passage human or rat fibroblasts were allowed to attach to tissue culture dishes and then were overlaid with a thin layer of fibrinogen that was clotted with thrombin. Fibroblasts began to migrate upwards into the fibrin within 24 hours and their numbers and the distance migrated were quantified over several days. The extent of fibroblast migration was affected importantly by the nature of the fibrin clot. Fibroblasts migrated optimally into gels prepared from fibrinogen at concentrations of -3 mg/ml; ie, near normal plasma fibrinogen levels. Migration was greatly enhanced by extensive cross-linking of the fibrin alpha-chains by factor XIIIa, as occurs when clotting takes place in vivo. When fibrinogen was clotted in Dulbecco's modified Eagle's medium, gamma-chains were cross-linked, but alpha-chain cross-linking was strikingly inhibited, and fibroblasts migrated poorly. Gels prepared from factor XIII-depleted fibrinogen exhibited neither alpha-nor gamma-chain cross-linking and did not support fibroblast migration. Further purification of fibrinogen by anion exchange high pressure liquid chromatography depleted fibrinogen of fibronectin, plasminogen, and other impurities; this purified fibrinogen clotted to form fibrin gels that supported reproducible fibroblast migration.

Ultrastructural localization of prostaglandin endoperoxide synthase (cyclooxygenase) to isolated, purified fractions of guinea pig peritoneal macrophage and line 10 hepatocarcinoma cell lipid bodies

Subcellular fractions of purified cytoplasmic, nonmembrane-bound lipid bodies were prepared from [3H]-arachidonic acid-labeled guinea pig peritoneal macrophages and line 10 hepatocarcinoma cells. These fractions, which contained [3H]-arachidonyl lipids, were shown to be devoid of contaminating cellular membranes by electron microscopy, and to contain prostaglandin endoperoxide (PGH) synthase by postembedding immunogold electron microscopy. These findings support a proposed role for these lipid-rich organelles in the generation of eicosanoids by oxidative metabolism of arachidonate in the cyclooxygenase pathway of inflammatory and neoplastic cells.

Vascular permeability factor mRNA and protein expression in human kidney

Vascular permeability factor (VPF), also known as vascular endothelial growth factor (VEGF), is a potent microvascular permeability-enhancing mediator as well as a selective mitogen for vascular endothelium. In this study, in situ hybridization and immunohistochemistry co-localized VPF mRNA and protein to glomerular visceral epithelial cells in human kidneys. Northern analysis confirmed the presence of VPF mRNA of expected size. The finding of VPF in renal glomerular epithelium identifies a potent mediator of permeability and endothelial proliferation whose role in renal physiology and pathology requires investigation.

Pathways of macromolecular tracer transport across venules and small veins. Structural basis for the hyperpermeability of tumor blood vessels

BACKGROUND

Blood vessels supplying tumors are hyperpermeable to macromolecules, but the mechanisms responsible are poorly understood.

EXPERIMENTAL DESIGN

To investigate the structural basis for the leakiness of tumor blood vessels, we performed a transmission electron microscopic study of three syngeneic transplantable carcinomas (mouse ovarian carcinoma and the line 1 and line 10 bile duct guinea pig carcinomas) at early intervals after intravenous injection of several macromolecular tracers. Tracers with widely differing physical properties were studied: horseradish peroxidase, ferritin, 150 kilodalton fluorescein isothiocyanate-dextran and gold-bovine serum albumin.

RESULTS

All tracers leaked primarily from venules and small veins at the tumor-host interface, for the most part vessels lined by a continuous endothelium. The predominant pathway by which all four tracers exited venules in all three tumors was by way of a system of smooth membrane-bound, interconnecting vesicles and vacuoles; these tended to cluster together at irregular intervals in the endothelial cell cytoplasm to form organelle-like structures, vesiculo-vacuolar organelles (VVO). In favorable sections, VVO interfaced with both the luminal and abluminal surfaces of endothelial cells. HRP alone crossed venules and small veins through apposed inter-endothelial cell junctions. Tracers also exited vessels by way of endothelial fenestrae where these occurred (rarely) in mouse ovarian tumor-associated venules. VVO occurred with similar frequency and complexity in the continuous endothelium-lined venules and small veins that supplied the normal subcutis of either tumor-bearing or control animals. As in tumor-associated vessels, VVO provided the predominant pathway by which all four tracers exited normal vessels, but VVO labeling and extravasation were both much greater in tumor than in control vessels (p < 0.001 for ferritin).

CONCLUSIONS

VVO are prominent structures in both tumor-supplying and control vessel endothelial cells and provide the primary pathway for macromolecular extravasation. The large increase in permeability characteristic of tumor vessels is likely attributable to upregulation of VVO function.

Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing

Persistent microvascular hyperpermeability to plasma proteins even after the cessation of injury is a characteristic but poorly understood feature of normal wound healing. It results in extravasation of fibrinogen that clots to form fibrin, which serves as a provisional matrix and promotes angiogenesis and scar formation. We present evidence indicating that vascular permeability factor (VPF; also known as vascular endothelial growth factor) may be responsible for the hyperpermeable state, as well as the angiogenesis, that are characteristic of healing wounds. Hyperpermeable blood vessels were identified in healing split-thickness guinea pig and rat punch biopsy skin wounds by their capacity to extravasate circulating macromolecular tracers (colloidal carbon, fluoresceinated dextran). Vascular permeability was maximal at 2-3 d, but persisted as late as 7 d after wounding. Leaky vessels were found initially at the wound edges and later in the subepidermal granulation tissue as keratinocytes migrated to cover the denuded wound surface. Angiogenesis was also prominent within this 7-d interval. In situ hybridization revealed that greatly increased amounts of VPF mRNA were expressed by keratinocytes, initially those at the wound edge, and, at later intervals, keratinocytes that migrated to cover the wound surface; occasional mononuclear cells also expressed VPF mRNA. Secreted VPF was detected by immunofluoroassay of medium from cultured human keratinocytes. These data identify keratinocytes as an important source of VPF gene transcript and protein, correlate VPF expression with persistent vascular hyperpermeability and angiogenesis, and suggest that VPF is an important cytokine in wound healing.

Expression and distribution of osteopontin in human tissues: widespread association with luminal epithelial surfaces

Osteopontin, a glycoprotein with a glycine-arginine-glycine-aspartate-serine (GRGDS) cell-binding domain, has been described in bone and is also known to be expressed in other organs, particularly kidney. The goal of the present work was to define the distribution of osteopontin synthesis and deposition in a wide variety of normal adult human tissues using a multifaceted approach that included immunohistochemistry, in situ hybridization, and Northern analysis. Immunohistochemical studies have revealed the unexpected finding that osteopontin is deposited as a prominent layer at the luminal surfaces of specific populations of epithelial cells of the gastrointestinal tract, gall bladder, pancreas, urinary and reproductive tracts, lung, breast, salivary glands, and sweat glands. Northern analyses identified gallbladder as a major site of osteopontin gene transcription comparable in magnitude with that of kidney, and immunoblotting identified osteopontin in bile. In situ hybridization localized osteopontin gene transcripts predominantly to the epithelium of a variety of organs as well as to ganglion cells of bowel wall. Osteopontin of epithelial cell origin, like bone-derived osteopontin, promoted GRGDS-dependent cell spreading in attachment assays. We postulate that osteopontin secreted by epithelium binds integrins on luminal surfaces. Collectively, these findings suggest an important role for osteopontin on many luminal epithelial surfaces communicating with the external environment.

Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages, and tumors

Vascular permeability factor (VPF), also known as vascular endothelial growth factor (VEGF), increases microvascular permeability and is a specific mitogen for endothelial cells. Expression of VPF/VEGF previously was demonstrated in a variety of tumor cells, in cultures of pituitary-derived cells, and in corpus luteum. Here we present evidence, by Northern analysis and in situ hybridization, that the VPF/VEGF gene is expressed in many adult organs, including lung, kidney, adrenal gland, heart, liver, and stomach mucosa, as well as in elicited peritoneal macrophages. The highest levels of VPF/VEGF transcripts were found in epithelial cells of lung alveoli, renal glomeruli and adrenal cortex, and in cardiac myocytes. The prominence of VPF/VEGF mRNA in these tissues suggests a possible role for VPF/VEGF in regulating baseline microvascular permeability, which is essential for tissue nutrition and waste removal. We also demonstrate particularly high VPF/VEGF mRNA levels in several human tumors, where it may be involved in promoting tumor angiogenesis and stroma generation, both as an endothelial cell mitogen and indirectly by its permeability enhancing effect that leads to the deposition of a provisional fibrin gel matrix.

Spatial distribution of tumor-specific monoclonal antibodies in human melanoma xenografts

The time-dependent (1-72-h) spatial distribution of three biotinylated anti-melanoma monoclonal antibodies (MAbs), a control MAb, and several macromolecular tracers was studied in two small (4-12-mg), well-characterized human melanoma xenografts (SK-MEL-2, M21) growing in the s.c. space of athymic nude mice. The specific MAbs (436, IND1, and 9.2.27) recognize two different melanoma cell surface antigens (Mr 125,000 glycoprotein melanoma-associated antigen and high molecular weight melanoma-associated antigen) and have equilibrium association constants differing by two orders of magnitude (10(8)-10(10) M-1). SK-MEL-2 tumors were poorly vascularized and were composed of one or several collections of tumor cells with few intratumor blood vessels. In contrast, M21 tumors induced a strong angiogenic response and were organized into multiple small tumor cell nests separated from each other by fine blood vessels. Neither tumor developed extensive connective tissue stroma. In both tumors, hyperpermeable blood vessels were concentrated at the tumor-host interface but some intratumor vessels in M21 tumors were also leaky. Macromolecular tracers extravasated extensively from leaky vessels into tumor stroma but penetrated poorly into tumor parenchyma. All three tumor-specific MAbs stained tumor cell surfaces in a time-dependent fashion such that one-half or more of all tumor cells were stained by 24-48 h. Tumor cell staining was favored by increased density of tumor cell antigens but, at the doses studied, was little affected by differences in affinity among tumor-specific antibodies. The distribution of MAb staining was nonuniform in two respects: (a) peripherally situated tumor cells were more likely to be stained than centrally placed cells, and only in the smallest tumors did MAb reach centrally placed tumor cells; and (b) staining was nonuniform in different parts of the same tumor. The inhomogeneity of tumor cell staining by tumor-specific MAb was attributable to several factors, including: tumor blood vessel number, distribution, perfusion and permeability; distribution of tumor connective tissue stroma; small volume of the parenchymal interstitial space and relatively impaired diffusion of macromolecules in that space (low effective diffusivity of MAb); and interactions between specific MAbs and tumor cells. Of these factors, those associated with the parenchymal compartment apparently were rate limiting, and strategies that enhance parenchymal penetration are likely to improve solid tumor therapy with MAbs.

A quantitative analysis of tumor specific monoclonal antibody uptake by human melanoma xenografts: effects of antibody immunological properties and tumor antigen expression levels

The time-dependent (5 min-72 h) localization of 3 radiolabeled anti-melanoma monoclonal antibodies (MAbs 436, IND1, and 9.2.27) was studied in paired label experiments in small (4-12 mg) s.c. human melanoma xenografts (SK-MEL-2 and M21) in athymic nude mice. MAb 436 recognizes a Mr 125,000 cell surface melanoma-associated glycoprotein antigen (125 kDa-MAA); MAbs IND1 and 9.2.27 recognize a high molecular weight melanoma-associated antigen, but with equilibrium association constants differing by 2 orders of magnitude (10(8)-10(10) M-1). The two tumors were found to differ in their antigen expression levels and in both interstitial and vascular volumes. Accumulation of MAbs in both tumors was determined primarily by antigen expression levels and also by physiological factors such as vascular permeability and vascular volume; at the dose administered (20 micrograms/mouse), differences in MAb affinity among specific MAbs had minimal effect on accumulation. Quantitative flow cytometry measurements showed that antigen expression in vivo differed from that of cultured tumor cells. In vivo, expression of the Mr 125,000 MAA decreased by a factor of about 2.5 in both tumors. In contrast, the in vivo expression of the high molecular weight MAA decreased in M21 tumors but increased by 2.0-3.5-fold in SK-MEL-2 tumors. Data were analyzed using a three-compartment pharmacokinetic model (C. Sung et al., Cancer Res., 52:377-384, 1992) to provide plasma-to-tissue transport constants (k), the interstitial fluid flow rate (L), and estimates of the in vivo interstitial MAb binding site concentration (B0). For all MAbs, the plasma-to-tissue transport constants were consistently greater for M21 tumors (0.44-0.85 microliter/min/g) than for SK-MEL-2 tumors (0.28-0.66 microliter/min/g), and values of k for both tumors were approximately 1 order of magnitude greater than those for skeletal muscle (0.06-0.08 microliter/min/g). The model-estimated binding site concentration of melanoma-specific antibodies was 15-70 times lower than that predicted by experimental measurements of tumor antigen concentrations. Factors that may contribute to this discrepancy include inaccessibility of tumor cell binding sites to MAb and MAb catabolism. In summary, these results indicate that, for the MAb dose used in this study, variables pertaining to the tumor target (i.e., antigen expression levels, vascular volume, and vascular permeability) are the most important for determining MAb accumulation in tumors.

Predicted and observed effects of antibody affinity and antigen density on monoclonal antibody uptake in solid tumors

The uptake and binding of monoclonal antibodies (MAbs) in solid tumors after a bolus i.v. injection are described using a compartmental pharmacokinetic model. The model assumes that MAb permeates into tumor unidirectionally from plasma across capillaries and clears from tumor by interstitial fluid flow and that interstitial antibody-antigen interactions are characterized by the Langmuir isotherm for reversible, saturable binding. Typical values for plasma clearance and tumor capillary permeability of a MAb and for interstitial fluid flow and interstitial volume fraction of a solid tumor were used to simulate the uptake of MAbs at various values of the binding affinity or antigen density for a range of MAb doses. The model indicates that at low doses, an increase in binding affinity may lead to an increase in MAb uptake. On the other hand, at doses approaching saturation of antigen or when uptake is permeation limited, an increase in the binding affinity from moderate to high affinity will have only a small effect on increasing MAb uptake. The model also predicts that an increase in antigen density will greatly increase MAb uptake when uptake is not permeation limited. Our experiments on MAb uptake in melanoma tumors in athymic mice after injection of 20 micrograms MAb (initial plasma concentration, about 120 nM) are consistent with these model-based conclusions. Two MAbs differing in affinity by more than 2 orders of magnitude (3.8 x 10(8) M-1 and 5 x 10(10) M-1) but with similar in vivo antigen densities in M21 melanoma attained similar concentrations in the tumor. Two MAbs of similar affinity but having a 3-fold difference in in vivo antigen density in SK-MEL-2 melanoma showed that the MAb targeted to the more highly expressed antigen attained a higher MAb concentration. We also discuss the model predictions in relation to other experiments reported in the literature. The theoretical and experimental findings suggest that, for high dose applications, efforts to increase MAb uptake in a tumor should emphasize the identification of an abundantly expressed antigen on tumor cells more than the selection of a very high affinity MAb.

Distribution of vascular permeability factor (vascular endothelial growth factor) in tumors: concentration in tumor blood vessels

Vascular permeability factor (VPF) is a highly conserved 34-42-kD protein secreted by many tumor cells. Among the most potent vascular permeability-enhancing factors known, VPF is also a selective vascular endothelial cell mitogen, and therefore has been called vascular endothelial cell growth factor (VEGF). Our goal was to define the cellular sites of VPF (VEGF) synthesis and accumulation in tumors in vivo. Immunohistochemical studies were performed on solid and ascites guinea pig line 1 and line 10 bile duct carcinomas using antibodies directed against peptides synthesized to represent the NH2-terminal and internal sequences of VPF. These antibodies stained tumor cells and, uniformly and most intensely, the endothelium of immediately adjacent blood vessels, both preexisting and those newly induced by tumor angiogenesis. A similar pattern of VPF staining was observed in autochthonous human lymphoma. In situ hybridization demonstrated VPF mRNA in nearly all line 10 tumor cells but not in tumor blood vessels, indicating that immunohistochemical labeling of tumor vessels with antibodies to VPF peptides reflects uptake of VPF, not endogenous synthesis. VPF protein staining was evident in adjacent preexisting venules and small veins as early as 5 h after tumor transplant and plateaued at maximally intense levels in newly induced tumor vessels by approximately 5 d. VPF-stained vessels were also hyperpermeable to macromolecules as judged by their capacity to accumulate circulating colloidal carbon. In contrast, vessels more than approximately 0.5 mm distant from tumors were not hyperpermeable and did not exhibit immunohistochemical staining for VPF. Vessel staining disappeared within 24-48 h of tumor rejection. These studies indicate that VPF is synthesized by tumor cells in vivo and accumulates in nearby blood vessels, its target of action. Because leaky tumor vessels initiate a cascade of events, which include plasma extravasation and which lead ultimately to angiogenesis and tumor stroma formation, VPF may have a pivotal role in promoting tumor growth. Also, VPF immunostaining provides a new marker for tumor blood vessels that may be exploitable for tumor imaging or therapy.

Glycosylation is essential for efficient secretion but not for permeability-enhancing activity of vascular permeability factor (vascular endothelial growth factor)

The hyperpermeability of the microvasculature supplying solid tumors is largely attributable to a heterodimeric Mr 34,000-43,000 tumor-secreted protein, vascular permeability factor. Upon reduction, the vascular permeability factor secreted by line 10 tumor cells is resolved by SDS-PAGE into 3 discrete bands of Mr 24,000, 19,500, and 15,000. We demonstrate here that line 10 vascular permeability factor is an N-linked glycoprotein. Nonglycosylated vascular permeability factor migrates on reduced SDS-PAGE as two bands of Mr 20,000 and 15,000. Pulse-chase studies demonstrated that all three chains of native vascular permeability factor were secreted rapidly following synthesis and at equal rates, with a cellular half-retention time of approximately 37 min. When glycosylation was prevented by tunicamycin, individual bands of nonglycosylated vascular permeability factor were also secreted at equivalent rates, but much more slowly (approximately 60 min) than native glycoprotein. Both glycosylated and nonglycosylated forms of vascular permeability factor were equally potent at increasing dermal vessel permeability.

Alterations in proteoglycan synthesis common to healing wounds and tumors

Wound healing and tumor stroma generation share several important properties, including hyperpermeable blood vessels, extravasation of fibrinogen, and extravascular clotting. In both, the deposits of fibrin gel serve initially as provisional stroma and later are replaced by granulation tissue. Proteoglycans (PG) are also important constituents of the extracellular matrix, but their composition and role in healing wounds and tumor stroma generation are poorly understood. The authors used immunohistochemical and biochemical methods to investigate the dermatan sulfate proteoglycan (DSPG) and chondroitin sulfate proteoglycan (CSPG) composition of healing skin wounds and solid tumors. By immunohistochemistry, the great majority of normal guinea pig and human dermis stained weakly for CSPG and strongly for decorin. In contrast, the granulation tissue of healing skin wounds and scars stained intensely for CSPG and weakly or not at all for decorin; however decorin staining was restored to normal intensity after digestion with chondroitin ABC lyase, suggesting that decorin antigenic sites had been masked by glycosaminoglycan (GAG) chains. Like wounds, the stroma of several carcinomas (line 1 guinea pig, human breast, colon, basal cell, and squamous) stained strongly for CSPG and weakly or not at all for decorin, but decorin staining developed after chondroitin ABC lyase digestion. Thus healing wounds and tumor stroma express a common pattern of altered PG staining, adding another to the properties these pathologic entities share. Proteoglycans extracted from healing wounds after in situ labelling with [35S] Na sulfate contained more CSPG than normal dermis with significantly longer GAG chains. Granulation tissue also synthesized more DSPG than normal skin, with greater heterogeneity and longer GAG chains. These alterations in PG synthesis correlate with the cell proliferation, migration, and collagen synthesis that accompany wound healing and may provide clues to the mechanisms responsible for both wound healing and tumor stroma generation.

Structure of solid tumors and their vasculature: implications for therapy with monoclonal antibodies

Delivery of monoclonal antibodies to solid tumors is a vexing problem that must be solved if these antibodies are to realize their promise in therapy. Such success as has been achieved with monoclonal antibodies is attributable to the local hyperpermeability of the tumor vasculature, a property that favors antibody extravasation at tumor sites and that is mediated by a tumor-secreted vascular permeability factor. However, leaky tumor blood vessels are generally some distance removed from target tumor cells, separated by stroma and by other tumor cells that together represent significant barriers to penetration by extravasated monoclonal antibodies. For this reason, alternative approaches may be attractive. These include the use of antibody-linked cytotoxins, which are able to kill tumor cells without immediate contact, and direction of antibodies against nontumor cell targets, for example, antigens unique to the tumor vascular endothelium or to tumor stroma.

Tumor-secreted vascular permeability factor increases cytosolic Ca2+ and von Willebrand factor release in human endothelial cells

Vascular permeability factor (VPF), a tumor-secreted heparin-binding protein (Mr approximately 38,000), is responsible for increased vessel permeability and fluid accumulation associated with tumor growth. Vascular permeability factor also promotes the growth of human umbilical vein endothelial cells (EC) and bovine pulmonary ECs in vitro. It is shown for the first time that guinea pig VPF (half-maximal and maximal dose approximately 0.4 and 22 pmol/l (picomolar), respectively), as well as human VPF, are potent stimuli for human ECs resulting in [Ca2+]i increases (maximal three- to fourfold) and inositol triphosphate (IP3) formation. Unlike the maximal responses to thrombin and histamine, the [Ca2+]i response to a maximal VPF dose was preceded by a characteristic 10- to 15-second delay. Guinea pig VPF also selectively increased [Ca2+]i in cultured aortic and pulmonary artery ECs, but not aortic smooth muscle cells, human fibroblasts, or neutrophils. Affinity-purified rabbit antibody (raised to a synthetic peptide representing VPF N-terminal amino acids 1 to 24) adsorbed all vessel permeability-increasing activity, EC growth-promoting activity, and specifically all activity responsible for increasing EC [Ca2+]i. Similar to other mediators that increase [Ca2+]i in cultured ECs, VPF also induced a 200% increase in von Willebrand factor release. Together these data indicate that VPF acts directly on ECs and that rapid cellular events in its in vivo/in vitro actions are likely to involve phospholipase C activation, [Ca2+]i increase, and von Willebrand factor release.

Cytoplasmic lipid bodies of human eosinophils. Subcellular isolation and analysis of arachidonate incorporation

Lipid bodies are non-membrane-bound cytoplasmic inclusions that are prominent in leukocytes engaged in inflammatory responses. As demonstrated by electron microscopic autoradiography, lipid bodies can serve as intracellular sites of 3H-arachidonic acid localization in eosinophils and other cells. To evaluate the role of lipid bodies as stores of esterified arachidonate, subcellular fractionation of lipid-body-rich human eosinophils was used to isolate lipid bodies free of other organelles. In lipid bodies isolated from 3H-arachidonate-labeled eosinophils, 3H-arachidonate was esterified almost totally in glycerolipids, predominantly in classes of phospholipids, including phosphatidyl-inositol and phosphatidylcholine. Lipid bodies, especially in leukocytes participating in inflammation, could represent intracellular sources of esterified arachidonate available for eicosanoid formation.

Purification and NH2-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor

Rodent and human tumor cell lines secrete a potent vascular permeability factor (VPF) which causes a rapid and substantial increase in microvascular permeability to plasma proteins without causing mast cell degranulation, or endothelial cell damage or without exciting an inflammatory cell infiltrate [D. R. Senger, S. J. Galli, A. M. Dvorak, C. A. Perruzzi, V. S. Harvey, and H. F. Dvorak. Science (Wash. DC), 219: 983-985, 1983; D. R. Senger, C. A. Perruzzi, J. Feder, and H.F. Dvorak. Cancer Res., 46: 5629-5632, 1986]. VPF now has been purified to homogeneity from guinea pig tumor cell culture medium; it is a Mr 34,000-43,000 protein, and a NH2-terminal amino acid sequence has been derived. A synthetic peptide corresponding to amino acid residues 1-24 of the native protein was used to raise rabbit antibodies which bind all of the vessel permeability-increasing activity secreted by guinea pig tumor cells and which stain purified VPF on immunoblots. These findings establish that this NH2-terminal amino acid sequence was derived from the permeability factor. Homology searches found no identity or close similarity between VPF NH2-terminal sequence and database sequences, indicating that VPF is distinct from other proteins for which sequence data are available. In particular, no sequence similarity was found between tumor-secreted VPF and other mediators of increased vessel permeability including plasma and glandular kallikreins.

Alteration of aliphatic lipid proton NMR linewidths by malignant tumors in guinea pigs

Water-suppressed proton nuclear magnetic resonance spectroscopy was used to observe plasma lipoprotein lipid methyl and methylene resonances from guinea pigs which had been injected with viable or heat-killed line 1 or line 10 tumor cells or sterile oil. It was shown that the widths of these resonances became significantly sharper as the number of tumor cells grew. Plasma from tumor-free control animals showed no change in the NMR linewidths. It is concluded that the changes observed reflect a specific host response to viable tumor cells, and in these models there is a reciprocal relationship between the number of viable tumor cells and the linewidths of plasma lipoprotein methyl and methylene resonances.

Exchange of macromolecules between plasma and peritoneal cavity in ascites tumor-bearing, normal, and serotonin-injected mice

Fluorescein-labeled dextrans (FITC-D) from 3 to 5000 kDa (Stokes' radii from 1 to 40 nm) were used to study influx from the plasma into the peritoneum and efflux from the peritoneal cavity into the plasma in normal and ascites tumor-bearing mice and in mice whose peritoneal vessels had been rendered hyperpermeable by serotonin. Two syngeneic transplantable murine ascites tumors were studied: mouse ovarian tumor and the TA3/St breast adenocarcinoma. To control for effects of peritoneal fluid volume, influx and efflux were also analyzed in mice that had received 5 ml of 5% bovine serum albumin i.p. as "artificial ascites." Following i.v. or i.p. injection, levels of FITC-D in the plasma and peritoneal fluid were quantitated by fluorimetry at successive time intervals from 5 to 360 min posttracer injection. Influx and efflux data were analyzed with a model consisting of three compartments (plasma, peritoneal cavity, and the extravascular space of all other organs) to yield kinetic parameters that characterized macromolecular transport. Depending on the size of the FITC-D tracer, from 3- to 50-fold more FITC-D accumulated in mouse ovarian tumor or TA3/St tumor ascites fluid, and 3- to 10-fold more FITC-D accumulated in the peritoneum of serotonin-treated than normal mice, all of it intact by gel exclusion chromatography. Influx of the FITC-D from plasma into the peritoneum, as characterized by the rate constant k1, was 2- to 40-fold greater in ascites tumor-bearing animals and 2- to 10-fold greater in serotonin-treated animals than in controls. Control animals with artificial ascites showed at most a 4-fold increase in the value of k1. As judged by fluorescence microscopy, the permeability of peritoneal-lining vessels in ascites tumor-bearing animals was greatly increased to FITC-D of 70 to 5000 kDa. Efflux of FITC-D, characterized by the rate constant k2, was reduced from 5- to 50-fold in ascites tumor-bearing animals but was unchanged or actually somewhat enhanced following serotonin treatment. Efflux in animals that had received artificial ascites was reduced 2.5- to 12.5-fold, correlating increased peritoneal fluid volume with decreased efflux. We conclude that tracer accumulation in malignant ascites fluid results from both increased influx as well as impaired efflux. Influx, and to a lesser extent efflux, were significantly affected by tracer size. However, within the range of FITC-D tested, we found no absolute size barrier to macromolecular transport from plasma to the peritoneal cavity, or vice versa.(ABSTRACT TRUNCATED AT 400 WORDS)

Leaky vessels, fibrin deposition, and fibrosis: a sequence of events common to solid tumors and to many other types of disease

Solid tumors must induce new blood vessels if they are to grow beyond minimal size. As an initial step in this process, tumors secrete a vascular permeability factor that renders the local microvasculature hyperpermeable to fibrinogen and to other plasma proteins. Extravasated fibrinogen is rapidly clotted to crosslinked fibrin gel. Over time, this gel is invaded by macrophages, fibroblasts, and endothelial cells and undergoes "organization," such that it is replaced by vascularized granulation tissue and finally by mature connective tissue. This sequence of events is not unique to tumors but occurs in wound-healing and in a wide variety of other disease processes, including some that prominently affect the lung.

Reappearance of an embryonic pattern of fibronectin splicing during wound healing in the adult rat

The adhesive extracellular matrix glycoprotein fibronectin (FN) is thought to play an important role in the cell migration associated with wound healing. Immunolocalization studies show abundant FN in healing wounds; however, these studies cannot define the cellular site(s) of FN synthesis, nor do they distinguish the different and potentially functionally distinct forms of FN that can arise from alternative splicing of the primary gene transcript. To examine these questions of FN synthesis and splicing during wound healing, we have performed in situ hybridization with segment-specific probes on healing wounds in adult rat skin. We find that the FN gene is expressed at increased levels after wounding both in the cells at the base of the wound and in subjacent muscle and dermis lateral to the wound. Interestingly, however, the pattern of splicing of FN mRNA was different in these areas. In adjacent dermis and muscle, the splicing pattern remains identical with that seen in normal adult rat skin, with two of the three spliced segments (EIIIA and EIIIB) excluded from FN mRNA. In contrast, these two segments are included in the FN mRNA present in the cells at the base of the wound. As a result, the mRNA in this region is spliced in a pattern identical with that found during early embryogenesis. The finding that the pattern of FN splicing during wound healing resembles an embryonic pattern suggests that alternative splicing may be used during wound healing as a mechanism to generate forms of FN that may be functionally more appropriate for the cell migration and proliferation associated with tissue repair.

Microvascular injury in pathogenesis of interferon-induced necrosis of subcutaneous tumors in mice

DBA/2 mice were injected sc with cells from the highly malignant Friend erythroleukemia cell (FLC) 3Cl8 subline, which is resistant to mouse interferon alpha/beta, or with the ESb lymphoma. When interferon alpha/beta was injected intratumorally or peritumorally, tumor growth was markedly suppressed, and established vascularized tumor nodules became progressively necrotic. Tumor necrosis was of the coagulation type that usually results from deprivation of blood flow. Morphologic examination of approximately 1,000 blood vessel profiles and approximately 2,000 endothelial cells in 1-micron Epon sections of sc 3C18 FLC tumors showed that interferon treatment resulted in rapid and pronounced vascular endothelial cell damage that preceded tumor necrosis. No inflammatory cell infiltrate was observed. Our results suggest that interferon alpha/beta exerted an antitumor effect in these tumor models by damaging tumor blood vessels, causing disruption of tumor blood flow, which led to ischemic tumor necrosis.