In vivo models of angiogenesis.

The process of building new blood vessels (angiogenesis) and controlling the propagation of blood vessels (anti-angiogenesis) are fundamental to human health, as they play key roles in wound healing and tissue growth. More than 500 million people may stand to benefit from anti- or pro-angiogenic treatments in the coming decades [National Cancer Institute (USA), Cancer Bulletin, volume 3, no. 9, 2006]. The use of animal models to assay angiogenesis is crucial to the search for therapeutic agents that inhibit angiogenesis in the clinical setting. Examples of persons that would benefit from these therapies are cancer patients, as cancer growth and spread is angiogenesis-dependent, and patients with aberrant angiogenesis in the eye, which may lead to blindness or defective sight. Recently, anti-angiogenesis therapies have been introduced successfully in the clinic, representing a turning point in tumor therapy and the treatment of macular degeneration and heralding a new era for the treatment of several commonly occurring angiogenesis-related diseases. On the other hand, pro-angiogenic therapies that promote compensatory angiogenesis in hypoxic tissues, such as those subjected to ischemia in myocardial or cerebral hypoxia due to occluding lesions in the coronary or cerebral arteries, respectively, and in cases of poor wound healing, are also being developed. In this review, the current major and newly introduced preclinical angiogenesis assays are described and discussed in terms of their specific advantages and disadvantages from the biological, technical, economical and ethical perspectives. These assays include the corneal micropocket, chick chorioallantoic membrane, rodent mesentery, subcutaneous (s.c.) sponge/matrix/alginate microbead, s.c. Matrigel plug, s.c. disc, and s.c. directed in vivo angiogenesis assays, as well as, the zebrafish system and several additional assays. A note on quantitative techniques for assessing angiogenesis in patients is also included. The currently utilized preclinical assays are not equivalent in terms of efficacy or relevance to human disease. Some of these assays have significance for screening, while others are used primarily in studies of dosage-effects, molecular structure activities, and the combined effects of two or more agents on angiogenesis. When invited to write this review, I was asked to describe in some detail the rodent mesenteric-window angiogenesis assay, which has not received extensive coverage in previous reviews.

Intratumoral microvascular density in malignant lymphomas of B-cell origin.

Sections of surgical lymph-node biopsies of four types of malignant non-Hodgkin's lymphoma of B-cell origin (B-NHL) classified according to the R.E.A.L. terminology or lymphadenitis were immunostained in order to demonstrate endothelial CD34 (QBEnd 10) and to determine the microvascular density and vessel-size distribution using an interactive image-analysis technique. Only microvessels displaying a cross-sectional area corresponding to a diameter of between 3.2 and 34.6 microm were included. The intratumoral microvascular density (iMVD) was found to be significantly higher in chronic lymphatic leukaemia (CLL, n = 13) compared with the clinically more aggressive mantle cell lymphoma (MCL, n = 9) and diffuse large B-cell lymphoma (DLBCL, n = 14). iMVD in CLL was also higher than in the follicular neoplastic parts (FL FOLL) of follicular lymphoma (FL, n = 16). In FL FOLL the microvessel density was, moreover, significantly lower than in the surrounding non-neoplastic FL tissue. In lymphadenitis (LA, n = 10) the iMVD was higher than in DLBCL, FL FOLL and MCL. The data suggest that future studies focusing on the relationship between iMVD and the clinical outcome within each particular NHL group should be carried out in order to verify whether iMVD is a prognostic factor in NHL, as it is in carcinomas.

Orally administered bovine lactoferrin systemically inhibits VEGF(165)-mediated angiogenesis in the rat.

Lactoferrin (Lf) systemically suppresses tumor growth and metastasis by unknown mechanisms. We have studied the effect of orally administered iron-unsaturated bovine Lf on angiogenesis induced by VEGF(165) and IL-1-alpha in adult rats using the mesenteric-window angiogenesis assay. VEGF(165) is a major angiogenic factor in most, if not all, tumors and other angiogenesis diseases of clinical relevance. A number of objective angiogenesis variables were analyzed using microscopic morphometry and image analysis. Lf treatment significantly inhibited the VEGF(165)-mediated response in terms of microvessel spatial extension, overall vascularity and incidence of crossover. The response to IL-1-alpha decreased significantly only in terms of microvessel crossover. In vitro, Lf exerted an antiproliferative effect on endothelial cells. To our knowledge, Lf is the first endogenous protein that has been shown to be antiangiogenic following oral administration. The oral administration of Lf thus appears to be of potential interest as an antiangiogenesis treatment modality in the clinical setting. Since tumor growth is angiogenesis dependent, the extensive therapeutic potential warrants further study to elucidate the mechanisms responsible for the angiostatic effect of Lf.

2.5 kDa and 5.0 kDa heparin fragments specifically inhibit microvessel sprouting and network formation in VEGF165-mediated mammalian angiogenesis.

Tumour growth is angiogenesis dependent. Thrombosis and thromboembolism are very common in cancer patients. These patients are often treated with heparin as an anti-coagulant. Many tumour angiogens, including VEGF165, and endogenous anti-angiogenesis factors bind heparin tightly. Using the non-surgical mesenteric-window angiogenesis assay, we studied in detail the systemic effect of heparin fractions with a mean MW of 2.5, 5.0 and 16.4 kDa on the microvessel sprouting and network formation in angiogenesis mediated by VEGF165 in rats. The microvessel network was assessed objectively in terms of the number and lengths of segments (the distance between two successive branching points), the number of branching points, the degree of tortuosity, the index of interconnecting loop formation, the index of intersection, as well as the number and lengths of sprouts. Compared with the saline control, the 2.5 kDa fraction significantly shortened the microvessel sprouts and the microvessel segments but increased the microvessel tortuosity in statistical terms; the 5.0 kDa fraction statistically significantly shortened the sprouts, decreased the number of segments and the number of microvessel branching points; whereas the 16.4 kDa fraction statistically significantly elongated the longest segments. Moreover, statistically significant differences were found between the three heparin fractions in terms of microvessel tortuosity (2.5 vs. 16.4 kDa), index of loop formation (5.0 vs. 2.5 + 16.4 kDa) and index of intersection (5.0 vs. 16.4 kDa). These findings demonstrate that heparin fragments size-specifically inhibit microvessel sprouting and network formation in VEGF165-mediated angiogenesis. As VEGF165 is a potent angiogen in human tumours, we suggest that heparin enriched in 2.5 kDa species and 5.0 kDa species especially should be exploited as a combined anti-coagulant and specific adjuvant anti-angiogenic agent in cancer patients who require anti-coagulant therapy.

NO and de novo mammalian angiogenesis: further evidence that NO inhibits bFGF-induced angiogenesis while not influencing VEGF165-induced angiogenesis.

Using the non-surgical rat mesenteric window angiogenesis assay, we studied the systemic effect of (i) the nitric oxide (NO)-releasing vasodilator isosorbide-5-mononitrate (ISMN) and (ii) the NO-synthase inhibitor L-NAME on angiogenesis induced by the intraperitoneal injection of bFGF and VEGF165. The response was assessed objectively and quantitatively by microscopic morphometry and image analysis in terms of the vascularized area (VA; a measurement of microvessel spatial extension), the microvascular length (MVL; a composite measurement of microvessel density), the total microvascular length (TMVL=VA x MVL), the number of microvessel segments per unit tissue volume (No. MS), the length of the microvessel segments (Le. MS) and the degree of microvessel tortuosity (MVT). Additional architectural features of the network were assessed in terms of variables introduced here: the number of microvessel branching points per unit tissue volume (No. BP), the index of interconnecting microvessel loop formation (In. LF), the index of microvessel intersection (In. IS), the number of microvessel sprouts per unit tissue volume (No. SP) and their length (Le. SP). In bFGF-mediated angiogenesis, L-NAME significantly, augmented angiogenesis, whereas ISMN significantly inhibited angiogenesis. By contrast, neither L-NAME nor ISMN affected the angiogenic response to VEGF165.

Oral administration of a nitric oxide synthase inhibitor enhances de novo mammalian angiogenesis mediated by TNF-alpha, saline and mast-cell secretion.

In healthy adult rats, the test tissue that is used in the mesenteric-window angiogenesis assay is natively vascularized, lacks physiological angiogenesis, and is unperturbed by surgical intervention. Using the rat MWAA oral treatment with the nitric oxide (NO) synthase inhibitor Nw-nitro-L-arginine methyl ester (L-NAME) enhanced the angiogenic response (compared with controls receiving the inactive enantiomer D-NAME) following i.p. injections of (i) TNF-alpha at an approximate physiological dose, (ii) Compound 48/80, which is a highly selective secretagogue causing mast-cell secretion in situ and a very strong angiogenic response, and (iii) saline of a grade not made for infusion, causing a weak angiogenic response. Angiogenesis was assessed quantitatively using microscopic morphometry and image analysis in terms of objective variables recording the microvascular spatial extension, microvascular density, number and length of microvessel segments (extending between two successive branching points) and the degree of microvessel tortuosity. The data strongly suggest that endogenous NO inhibits all three mammalian angiogenesis reactions, although to a markedly different extent. Notably, the present data are virtually the opposite of those that have been reported from other mammalian angiogenesis models, the test tissues of which display deranged homeostasis, such as surgical intervention and/or ischemia.

Constitutively synthesized nitric oxide is a physiological negative regulator of mammalian angiogenesis mediated by basic fibroblast growth factor.

We recently reported that the systemically administered nitric oxide synthase (NOS) inhibitor Nw-nitro-L-arginine methyl ester, L-NAME, administered before, during and after the angiogenic treatment stimulated angiogenesis induced by basic fibroblast growth factor, bFGF, in the rat. This suggests that suppression of constitutively expressed NOS, cNOS, plus inducible NOS, iNOS, and thus reduced production of nitric oxide, NO, was the stimulating factor. In those studies, the rat mesenteric-window angiogenesis assay was used. Moreover, the systemic administration of a NO releaser inhibited bFGF-mediated angiogenesis. Using the same experimental system, we have now studied whether the inhibition of cNOS alone in adult animals under physiological conditions, i.e. prior to the administration of the angiogenic stimulation with bFGF, affected the subsequent angiogenic response. cNOS constitute endothelial cell NOS (ecNOS) and neuronal NOS (nNOS). L-NAME or its inactive enantiomer Nw-nitro-D-arginine methyl ester, D-NAME, were given continuously in the drinking water (1.0 g/L) during 14 days prior to the start of the treatment with bFGF. The treatment with L-NAME significantly enhanced the subsequent angiogenic response. NO synthesized under physiological conditions by ecNOS in endothelial cells and platelets or nNOS in platelets may thus act as a first constitutional angiostatic factor in bFGF-mediated mammalian angiogenesis.

Microvascular density in terms of number and length of microvessel segments per unit tissue volume in mammalian angiogenesis.

The number and length of all the microvessel segments present per unit of tissue volume are the two determinants of microvascular density. In the present study, a microscopic interactive image-analysis method was introduced, using the length of individual microvessel segments extending between two successive points of bifurcation (Le. MS) and their number (No. MS) per unit of tissue volume in large populations of microvessel segments, which were measured using the nonsurgical rat mesenteric-window angiogenesis assay. In addition, microvessel tortuosity was quantitatively assessed. The method presented here was applied to the angiogenic response over 3 weeks to the directly acting angiogen VEGF165, which was administered i.p. at three dose levels, i.e., 4.8, 48, and 480 pM. Following the VEGF165-treatment, statistically significant changes were found not only in the No. MS but also in the distribution of Le. MS: shortening of vessels being an indicator of vessel branching and increased vessel length being an indicator of angiogenesis through elongation. The shape of the overall distribution of Le. MS was, however, basically the same on any observation occasion regardless of whether significant angiogenesis was present or not. VEGF165 induced a roughly dose-dependent angiogenic response in terms of No. MS and the length of the shortest (0-10% percentile) and/or the longest (90-100% percentile) microvessel segments. Moreover, the onset of the early stimulating effect of VEGF165 on microvessel branching was also dose-dependent. The variables which were introduced here were shown to display a very high degree of sensitivity and resolution and are apparently unrivalled when it comes to the study of density in populations of microvessels.

Nitric oxide suppresses bFGF- and IL-1-alpha-mediated but not VEGF165-mediated angiogenesis in natively vascularized mammalian tissue.

Using the rat mesenteric window angiogenesis assay, we studied the systemic effect of the nitric oxide synthase inhibitor Nw-nitro-L-arginine methyl ester, L-NAME, on angiogenesis induced by basic fibroblast growth factor (bFGF), interleukin-1-alpha (IL-1) or vascular endothelial growth factor (VEGF165). Using technically independent morphometric and image analysis methods, the angiogenic response was quantified in variables related to (i) microvascular spatial extension and (ii) microvascular density. The test tissue, which is natively vascularized and lacks significant angiogenesis physiologically, was unaffected by surgical intervention. Two daily intraperitoneal injections of bFGF (2.20 nmole), IL-1 (1.18 nmole) and VEGF165 (480 pmole) for 5 days elicited significant angiogenesis in the mesenteric windows. L-NAME (0.5 g/L in drinking water) caused further enhancement of the angiogenic response produced by human recombinant bFGF (p<0.001), bovine purified bFGF (p<0.05) or murine recombinant IL-1 (p<0.05). In contrast, the L-NAME treatment did not affect the angiogenic response produced by human and murine recombinant VEGF165. These data suggest that nitric oxide can act as an endogenous suppressor of mammalian de novo angiogenesis, which is a new finding, and, moreover, that angiogenesis induced by VEGF165 on the one hand and by bFGF and IL-1 on the other in the rat mesenteric window depends on different pathways.

Interleukin-1-alpha and de novo mammalian angiogenesis.

In the literature, the role of interleukin-1 (IL-1) as an angiogen is controversial. The ability of IL-1-alpha to induce de novo angiogenesis in adult rats was studied using the mesenteric window angiogenesis assay (MWAA). Murine recombinant IL-1-alpha was injected intraperitoneally twice daily on Days 0 to 4 at 11.8 pM, 118 pM, and 1.18 nM and groups of animals were sacrificed on Days 7, 14, 21 and 28; controls received the vehicle. Angiogenesis was quantified in terms of microvascular spatial extension and density using technically independent microscopic techniques and image analysis. Compared with the vehicle control, the treatment with IL-1-alpha at doses of 118 pM and 1.18 nM induced statistically significant angiogenesis throughout the study period, whereas IL-1-alpha at 11.8 pM did not induce significant angiogenesis in statistical terms until Days 21 and 28. Compared with the previously reported angiogenic response to VEGF165, bFGF, IL-8, and TNF-alpha using the rat MWAA and the same standardized experimental protocol, the IL-1-alpha treatment displayed a higher degree of efficacy and potency than that of bFGF, IL-8, and TNF-alpha. Moreover, the duration of the significant response to IL-1-alpha exceeded that of bFGF, IL-8, and TNF-alpha. The present data indicate that IL-1-alpha at near-physiologic doses is a very effective angiogenic factor in the system used here. The response may well be multifactorially mediated, as is discussed, and the molecular mechanisms which are involved remain to be clarified.

Angiogenesis: new aspects relating to its initiation and control.

Angiogenesis, the formation of new microvessels from parent microvessels, involves remodeling the basement membrane and interstitial extracellular matrix (ECM) using degrading proteases produced by the endothelial cells (ECs) and other adjacent cells, and the synthesis of ECM molecules by these cells. Degraded ECM releases previously bound heparin-binding cytokines (and growth factors) which are able to act as ligands to high-affinity receptors on various target cells, including ECs. The EC carries receptors for a number of cytokines which are produced by neighboring cells or released from the ECM and which can either induce or suppress the angiogenic phenotype of the EC. ECs are able to synthesize and secrete cytokines with auto- and paracrine effects. Angiogenesis, which virtually never occurs physiologically in adult tissues (except in the ovary, the endometrium and the placenta), is essential in wound healing and inflammation. Angiogenesis is, in fact, strictly controlled by a redundancy of pro- and anti-angiogenic paracrine peptide molecules, some of which have recently been described. The expression and synthesis of two distinct anti-angiogenic factors is, for example, controlled by the p53 tumor suppressor gene. In certain hypoxic conditions, chronic inflammatory diseases and syndromes, angiogenesis is of pathogenic and prognostic significance. Angiogenesis is, moreover, essential for the growth and metastatic spread of solid tumors. This indicates the potential for developing new therapeutic strategies not only for tumors but also in diseases such as rheumatoid arthritis, psoriasis, liver cirrhosis and diabetic retinopathy. Moreover, the therapeutic induction of angiogenesis in ischemic tissues using recombinant cytokines is also promising for clinical application. In fact, the first successful human gene therapy for stimulating angiogenesis has recently been reported.

A 5.0-kD heparin fraction systemically suppresses VEGF165-mediated angiogenesis.

The systemic effect of heparin fractions with mean molecular masses of 2.5, 5.0 and 16.4 kD on angiogenesis induced by vascular endothelial growth factor isoform 165 was studied using the truly quantitative rat mesenteric-window angiogenesis assay. The angiogenic treatment with 5 ml of VEGF165 at 480 pM was given intraperitoneally on days 0-4 and heparin fractions were given subcutaneously on days 0-13; animals were sacrificed on day 14. As the overlaps between the molecular mass distributions of the three fractions were relatively small, they essentially represent three different populations of heparin molecules. The doses of the heparins given were equal in terms of weight, but different in terms of the number of molecules and biologic activity. Angiogenesis was assessed in terms of vascularized area (VA), a measurement of microvascular spatial extension, and microvascular length (MVL), a measurement of microvascular density, using technically independent variables and image analysis. The total microvascular length was computed from VA x MVL. Treatment with the 5.0-kD fraction suppressed angiogenesis significantly in statistical terms compared with treatment with 2.5- and 16.4-kD heparins and the saline in controls. Interestingly, the 2.5-kD heparin fraction which was used here has previously been shown statistically significantly to suppress angiogenesis mediated by basic fibroblast growth factor in the same experimental system. Our data thus suggest that the systemic angiosuppressive effect of heparin in different mammalian angiogenic reactions is distinctly related to structural features such as molecular size.

Relaxin, a potent microcirculatory effector, is not angiogenic.

The ability of relaxin (RLX), which is a potent microcirculatory effector in many species including the rat, to induce de novo angiogenesis in vascularized mammalian tissue was tested using the rat mesenteric-window angiogenesis assay. RLX was administered intraperitoneally on days 1-5 at doses of 0.33, 3.3 and 33 nM. Controls received the vehicle by the same route. Groups of animals were sacrificed at the end of the 1st, 2nd and 3rd weeks. Using computer-aided microscopic morphometry including image analysis, the response was quantified by sensitive, technically independent, highly reproducible methods in terms of the vascularized area (VA), a measure of microvascular spatial extension, and the microvascular length (MVL), a measure of microvascular density. The total MVL was computed from VA x MVL. The results obtained show that RLX did not cause significant changes in any of the variables tested, regardless of dose and observation time. These findings indicate that RLX is apparently unable to mediate significant de novo angiogenesis in the system used in contrast to previously tested angiogens such as basic fibroblast growth factor, vascular endothelial growth factor, isoform 165, and tumor necrosis factor-alpha. In previous studies, RLX has been shown to exert antitumor activity on breast cancer cells in vitro. In the search for a possible role for RLX as an anticancer agent in vivo, it is important to know that this peptide is not angiogenic, since de novo angiogenesis is known to be a prerequisite for tumor growth and metastatic spread.

Interleukin-8 and de novo mammalian angiogenesis.

In the rat mesenteric-window angiogenesis assay (MWAA), the test tissue is natively vascularized, lacks significant physiological angiogenesis and its homeostasis is unperturbed by surgical intervention. Using the rat MWAA, it is shown here that interleukin-8 (IL-8), administered at approximately physiological doses, is able to induce de novo angiogenesis. Human recombinant IL-8 was administered intraperitoneally at two daily doses of 25 pM, 250 pM and 2.5 nM for 5 consecutive days (days 0-4). Using microscopic, computer-aided techniques including image analysis, the de novo angiogenic response was quantified in groups of animals on days 7, 14 and 21 in terms of the relative vascularized area (VA), a measure of the microvascular spatial extension, and the microvascular length (MVL), a measure of microvascular density or length. The total microvascular length (TMVL) was computed from VA x MVL. A statistically significant angiogenic effect was found in terms of MVL on day 7 and in terms of VA and TMVL on day 14 following the treatment with 2.5 nM, whereas MVL was significantly increased in statistical terms on day 14 following the treatment with IL-8 at the low dose of 25 pM. Notably, IL-8 at the intermediate dose of 250 pM did not induce a statistically significant angiogenic effect in terms of VA, MVL or TMVL on any observation occasion, thereby suggesting that the dose-related angiogenic effect of IL-8 may be nonlinear. This appears to be the first paper showing that IL-8 is able to induce de novo angiogenesis in normally vascularized mammalian tissue.

Vascular endothelial growth factor and de novo mammalian angiogenesis.

The time course and potency of the de novo angiogenic response to vascular endothelial growth factor isoform 165 (VEGF165) at approximate physiologic doses was assessed using the nonsurgical mesenteric-window angiogenesis assay in adult rats. Daily i.p. injections of VEGF165 at 4.8, 48, and 480 pM were given for Days 0-4, controls receiving the vehicle. Groups of 10 animals per treatment group were sacrificed on Days 7, 14, and 21. Using microscopic morphometry and image analysis, the vascularized area, a measure of microvascular spatial extension, and the microvascular length, a measure of microvascular density, were measured. At 4.8 and 480 pM, VEGF165 induced significant angiogenesis as early as on Day 7, suggesting a direct angiogenic effect. This very rapid angiogenic response to VEGF165 is distinct from other angiogenesis reactions, including angiogenesis induced by basic fibroblast growth factor, studied using the same methodology. During the early angiogenic phase, the increase in microvascular spatial expansion dominated over the increase in microvascular density. The specific response to VEGF165 peaked on Day 7 for the 4.8 pM dose and on Day 14 for the 480 pM dose, which was the more potent. It is, moreover, noteworthy that no statistically significant response was induced by the intermediate dose of VEGF165 (48 pM) at any observation time. The data thus suggest that, although VEGF165 at near physiologic doses apparently acts as a direct angiogen, its dose-related effect in terms of angiogenesis is nonlinear.

Basic-fibroblast-growth-factor-mediated de novo angiogenesis is more effectively suppressed by low-molecular-weight than by high-molecular-weight heparin.

We recently reported that the subcutaneous (s.c.) administration of a low-molecular-weight heparin (LMWH) fraction significantly inhibited de novo angiogenesis in the mesentery induced by the intraperitoneal (i.p.) injection of saline to adult rats compared with unfractionated heparin and high-molecular-weight heparin (HMWH) fractions. The present study assesses the effect on basic fibroblast growth factor (bFGF)-mediated de novo angiogenesis in the mesentery of the systemic administration of a LMWH fraction (2.6 kD) and a series of four HMWH fractions (about 20 kD) with varying degrees of polydispersity, charge density and anticoagulant activity. bFGF, a prototypic heparin-binding angiogenic growth factor, was injected i.p. at 220 pM on days 0-4. The heparins were given s.c. on days 0-13 or 0-14 at doses which were approximately within the range used clinically. Angiogenesis was assessed by microscopic morphometry and image analysis in groups of animals killed on days 14 and 15. Compared with the saline control, the LMWH and three of the HMWHs significantly inhibited angiogenesis in terms of microvascular length (MVL), a measure of microvascular density. Interestingly, the vascularized area (VA), a measure of microvascular spatial extension, and the total microvascular length (VA x MVL) were significantly lower in the LMWH-treated animals than in the animals treated with one of the HMWHs. The total microvascular length was, moreover, significantly reduced in the LMWH-treated animals compared with the combined data of all the HMWH-treated animals. No significant effects were related to the degree of charge density and anticoagulant activity of the heparins. In view of the putative significant angiogenic role of bFGF in human angiogenesis diseases, the present findings may have implications for the choice of anticoagulant treatment modality for patients suffering from cancer and other angiogenesis diseases.

Evidence of a dual role of endogenous histamine in angiogenesis.

The specific activation of mast cells in situ causes vigorous local mast-cell mediated angiogenesis (MCMA). The mast cell is a major source of histamine and, as recently reported, specific histamine H1- and H2-membrane receptor antagonists are able individually to significantly suppress MCMA in rats, as assessed using the mesenteric window angiogenesis assay (MWAA). In addition to membrane receptors for histamine, a type of intracellular histamine receptors, designated Hic, has been described. It is now demonstrated that the potent Hic-receptor antagonist DPPE (N,N-diethyl-2-[4-(phenylmethyl)phenoxy]ethanamine HCl), administered parenterally, stimulates MCMA significantly in rats, as quantified by the MWAA. Although the target cell(s) are not known, there are several ways by which their Hic receptors could be activated: uptake of histamine released from mast cells, mobilization from preformed cytoplasmic and nuclear stores, and production of de novo histamine by histidine decarboxylase activity. The fact that the occupancy by histamine of H1- and H2-membrane receptors stimulates MCMA and the occupancy by histamine of Hic inhibits MCMA suggests that endogenous histamine is capable of regulating angiogenesis by a dual mode of action. This is apparently the first report ascribing a dual role of this type in angiogenesis to a single molecule.