Stem Cells, Vol. 15, No. 3, 180-189,
May 1997
© 1997 AlphaMed Press
The Regulation of Neovascularization by Matrix Metalloproteinases and Their Inhibitors
Marsha A. Moses
The Children's Hospital and Harvard Medical School, Boston, Massachusetts, USA
Key Words. MMPs • TIMPs • Angiogenesis inhibitors • Proteolytic enzymes
Dr. Marsha A. Moses, Laboratory of Surgical Research, Children's Hospital, 320 Longwood Avenue, Boston, MA 02115, USA.
 |
Abstract
|
|---|
The process of new capillary formation from pre-existing vessels, angiogenesis, is a complex physiological event which is strictly controlled, occurring only very rarely under normal conditions. In contrast, there are a number of serious diseases, among them solid tumor growth, rheumatoid arthritis and several eye diseases, which are characterized by unrestricted new capillary growth and which are described as "angiogenic diseases." One of the key events required for successful angiogenesis is extracellular proteolysis. Increased attention has been focused on the matrix metalloproteinase (MMP) family of enzymes whose activity is a rate-limiting step in extracellular matrix remodeling. This review will present the accumulating body of evidence, from a number of laboratories, which documents the important role of MMP activity in the regulation of angiogenesis. Taken together, these data suggest that one strategy for controlling the deregulated angiogenesis characteristic of these serious angiogenic diseases may be one which is operative at the level of the control of MMP activity.
|
Introduction
|
|---|
Angiogenesis, also referred to as neovascularization, is a tightly regulated event which occurs very rarely under normal physiological conditions, two of which are wound healing and corpus luteum formation. With few other exceptions, vascular turnover in the adult is extremely low, so low that, in the adult, the capillary endothelial cells (EC) are said to be nearly quiescent. This feature was established by the demonstration that the potential doubling time of capillary endothelium from normal tissues can range from 47-20,000 days, in contrast to 2-13 days for tumor capillary endothelium [1]. The striking infrequency of occurrence of neovascularization under normal versus pathological conditions implies a strictly controlled system. Deregulation of this system and the consequent uncontrolled neovascularization has been correlated with the occurrence of a number of serious pathologies termed "angiogenic diseases," all of which are characterized by the aberrant and unrestricted growth of new microvessels [2]. Extracellular proteolysis is now being appreciated as an important biochemical and cellular regulator of neovascularization, vascular morphogenesis and vascular invasion [3-7]. The process of extracellular matrix (ECM) remodeling by capillary EC has been shown to be permissive for the migration, proliferation and capillary tube formation required for neovascularization [8]. In fact, in a recently proposed model, capillary formation is presented as the product of "alternate cycles of activation and inhibition of extracellular proteolysis" [9]. Plasminogen activator/plasmin and the matrix metalloproteinases and their endogenous inhibitors have been the most studied enzyme systems with respect to their role in the control of angiogenesis. The major goal of this concise review is to present the accumulating body of data which supports a role for matrix metalloproteinases (MMPs) and their inhibitors in the regulation of angiogenesis. The reader is referred to a number of recent reviews for an analysis of the role of the plasminogen activator/plasmin systems in angiogenesis [9, 10].
|
Angiogenesis: The Process
|
|---|
Angiogenesis is the process of new capillary formation from pre-existing vessels. This mechanism of vascular development is distinguished from vasculogenesis, which has been defined as the development of new blood vessels from in situ angioblasts which are themselves undergoing differentiation [11, 12]. Although both angiogenesis and vasculogenesis occur during embryonic development, angiogenesis or neovascularization can occur throughout one's adult life span. In the simplest of descriptions, new capillaries are formed during angiogenesis when capillary EC comprising the parent venule are stimulated by an angiogenic stimulus, such as fibroblast growth factor (FGF) and/or vascular endothelial cell growth factor (VEGF)/vascular permeability factor (VPF), to produce proteolytic enzymes such as MMPs and plasminogen activator, protease families which facilitate EC release via the degradation of the venule's basement membrane. Proteolytic activity is also required for the migration of EC into the perivascular stroma. These events are followed by sprout extension and subsequent lumen formation [13].
The breaching of the pericapillary membrane by EC as they "escape" from the parent venule, EC migration through the ECM, capillary sprout elongation, lumen formation, and the ECM remodeling which accompanies these events represent processes which require proteolytic activity [13, 14] and which are ultimately dependent on a shift in the proteolytic balance in favor of enzymatic activity (Fig. 1). Each of these distinct angiogenic processes have also been viewed as potential sites of therapeutic intervention using anti-angiogenic molecules [15]. A variety of in vitro and in vivo assays have been developed to study the distinct events comprising the process of neovascularization and to provide information as to the mechanisms by which a molecule might be exerting its anti-angiogenic effect in vivo. Among the most extensively used in vitro assays include those which measure capillary EC growth (proliferation, DNA synthesis), EC migration (chemotaxis and chemokinesis), invasion, capillary tube and sprout formation. These bioassays have been reviewed in detail elsewhere [15, 16]. In vivo, the chick chorioallantoic membrane assays [17] and its modifications [18], and the corneal pocket assay [19-21] and its modifications [22] are the most commonly used.
View larger version (29K):
[in this window]
[in a new window]
|
Figure 1. Simplified schematic of angiogenic processes that represent potential therapeutic intervention sites. Reprinted with permission [15].
|
|
|
Matrix Metalloproteinases and Tissue Inhibitors of Metalloproteinase (TIMPs)
|
|---|
An accumulating body of evidence suggests that the remodeling of ECM that occurs during normal growth, wound repair and angiogenesis, as well as during the development and progression of pathologic conditions including tumor growth and metastasis, is accomplished largely through the action of MMPs, whose activity is one of the rate-limiting steps of ECM degradation [23-25].
The MMPs are members of a multigene family of metal-dependent enzymes. These proteases have been classified into four broad categories originally based on substrate specificity. These specific enzymes are the collagenases (MMP-1/EC3.4.24.7; MMP-8/EC3.4.24.34; MMP-13), the gelatinases A (MMP-2/EC3.4.24.24) and B (MMP-9/EC3.4.24.35), the stromelysins (MMP-3/EC3.4.24.17; MMP-10/EC3.4.24.22; MMP-11/EC3.4.24.7) including a metalloelastase (MMP-12) and the membrane-type MMPs (MT-MMPs) [23-25].
The regulation of MMP activity occurs at several levels including gene transcriptional control (see below for positive regulation by angiogenic factors) and proenzyme activation and inhibition of activated MMPs by endogenous inhibitors. Like many other enzyme families, the MMPs are components of a system of "balanced proteolysis" wherein a finely tuned equilibrium exists between the amount of active enzyme and its proteinase inhibitor(s) [7]. These native metalloproteinase inhibitors comprise a family of proteins generally referred to by the acronym TIMPs (tissue inhibitor of metalloproteinase). Three of these inhibitors have been cloned and expressed to date: TIMP-1, TIMP-2 and TIMP-3 [26-31]. There are also a number of less well-characterized, lower molecular weight metalloproteinase inhibitors which await complete purification and identification. The amino acid sequence similarity between these TIMP family members is relatively high at about 45% (25% identity) and includes total conservation of 12 Cys residues known to form the characteristic six disulfide bonds [32].
The TIMPs are specific inhibitors of MMPs. They bind to form a tight (Ki < 1 nM), 1:1, noncovalent complex [33-35]. These endogenous inhibitors show little difference in their specificity for the MMPs with each TIMP able to inhibit all members of the MMP family [36]. TIMP-2 has been shown to have a four- to sevenfold preference for MMP-2 over TIMP-1 while TIMP-1 is approximately twofold more effective against MMP-1 [37]. The major structural and functional differences known to date between TIMPs-1, -2 and -3 lie in their glycosylation and their specificity of binding the latent proenzyme form of gelatinase A and B. TIMP-2 complexes only with progelatinase A, while TIMP-1 binds preferentially to progelatinase B [38, 39].
Most probably by virtue of their ability to inhibit MMP activity, TIMPs-1 and -2 have been shown to be capable of altering the metastatic phenotype of certain cancer cells [40-42]. They also have the ability to inhibit the invasion and metastasis of tumor cells in vivo [43]. Treatment of noninvasive, nonmetastatic Swiss 3T3 cells with antisense TIMP-1 conferred oncogenicity and metastatic potential to these cells [40] and, as a corollary, rat 4R cancer cells lost their invasive, metastatic phenotype when transfected with TIMP-2 [43].
In addition to their ability to inhibit MMP activity, new and important functions are now being attributed to the TIMPs. In addition to their ability to inhibit MMP activity, some of these proteins also possess erythroid potentiating activity [44, 45], mitogenic activity for a wide number of cell types [46, 47] and anti-angiogenic activity (see below). It remains to be determined whether these growth-modulating activities are related to MMP inhibitory activity [48].
|
A Role for Extracellular Proteolysis in Angiogenesis
|
|---|
It has long been appreciated that capillaries in vivo are surrounded and supported by a collagen-rich ECM. Capillary EC monolayers grown between two monolayers of collagen could be induced to form three-dimensional, branching and anastomosing capillary tubes with patent lumen [49].
The importance of the role of proteolysis during the angiogenic process was highlighted in one of the earliest studies of the angiogenic process in vivo. Ausprunk and Folkman [13] reported a study describing the sequence of events occurring during the formation of capillary vessels under the influence of an angiogenic stimulus, in this case a tumor implant. They noted that during neovascularization, the migration of capillary EC preceded cell proliferation and that one of the earliest events in the process of new capillary formation was that the EC from within the parent venule began to migrate out of the wall of the parent vessel forming a "capillary bud" which eventually elongated into a new capillary. After formation of the new capillary, DNA synthesis was found to be limited to the distal tips of the growing microvessel [13].
The requirement for EC migration out through the pericapillary membrane surrounding the parent venule and the subsequent fragmentation of the basal lamina which precedes this event were some of the earliest pieces of evidence supporting an important role for proteolytic activity during angiogenesis. The subsequent migration of capillary EC through the ECM of the tissue which is being vascularized towards the angiogenic stimulus also implicated proteolysis in successful neovascularization. As mentioned earlier, the plasminogen activator/plasmin and MMP enzyme families have been the most commonly studied proteases with respect to angiogenesis regulation.
|
Expression, Production and Differential Regulation of MMPs and TIMPs by EC
|
|---|
An increasing body of data has documented the differential expression and production of MMPs and TIMPs by EC. In some of the earliest reported studies, both human umbilical vein endothelial cells (HUVECs) and bovine capillary EC produced a phorbol myristate acetate (PMA)-inducible latent MMP-1 [50, 51]. Interestingly, the latter study demonstrated that, not only did capillary EC produce approximately 2x as much collagenase activity as did the large vessel bovine aortic endothelial cells, but also that there was no induction of this enzyme activity by tissue plasminogen activator (TPA) in the aortic EC [51]. This difference in the TPA response was suggested to represent an important and unique property of capillary EC versus their larger vessel counterparts.
Rabbit brain capillary EC were shown to secrete procollagenase and prostromelysin as well as a number of endogenous MMP inhibitors, including TIMP-1, in response to TPA treatment. Removal of the inhibitors permitted the activation of the proenzymes, demonstrating that the expression of collagenase and stromelysin by these microvascular EC was regulated by their endogenous inhibitors [3, 52, 53].
The perivascular ECM is composed predominantly of Type I collagen, and two specific MMPs, interstitial collagenase (MMP-1) and neutrophil collagenase (MMP-8), are capable of degrading Type I collagen [9]. In a recent study, the activity of MMP-1 was shown to be an absolute requirement for angiogenesis in vitro [54]. Using HUVECs as well as human dermal microvascular EC (HDMVECs) in a quantitative collagen gel invasion assay, it was shown that MMP-1, -2 and -9 were produced by phorbol ester-stimulated HUVECs and HDMVECs during EC invasion and endothelial tubule formation in this system. This collagen degradation could be significantly inhibited by the MMP inhibitors BB-94 (batimastat), a wide spectrum synthetic MMP inhibitor [55] and TIMP-1, evidence that the suppression of MMP activity was sufficient to inhibit capillary tube formation in vitro. Importantly, it was demonstrated that the inhibitors suppressed the generation of tropocollagen fragments from the native collagen substrate thereby specifically implicating MMP-1 as a key requirement for angiogenesis in vitro in this model [54].
A number of physiological modulators of angiogenesis have been shown to differentially regulate the expression and activity of MMPs by EC. Reports that the stimulation of HUVEC by VEGF/VPF, the potent secreted angiogenic factor, induced the expression of MMP-1 at the exclusion of other MMPs [56] and that the angiogenic mitogen, bFGF, can stimulate production of MMPs by EC, including an unidentified gelatinase and interstitial collagenase but not stromelysin [57], further support the importance of MMP-1 in the angiogenic process.
MMP-3 but not MMP-1 or -2, is induced by the inflammatory mediator TNF-alpha in human vein and microvascular EC as is the case for chondrocytes and fibroblasts. This expression was enhanced further, in the case of MMP-1 and MMP-3, by the addition of PMA. Expression of MMP-9 and TIMP-1 is inducible by PMA, and MMP-9 expression could be further enhanced by TNF-alpha or interleukin-alpha [58]. Importantly, the expression of neither MMP-2 or TIMP-2 was affected by treatment of these cell types with either PMA and/or TNF-alpha. The authors suggested that the differential regulation of MMP expression by these inflammatory mediators and by PMA may serve to locally regulate the proteolytic activity of these enzymes in order to preserve tissue integrity [58].
Retinoids, in particular the retinol and retinoic acid derivatives of vitamin A, when added at physiological doses (1 x 10–6 M), are inhibitors of capillary EC proliferation in vitro [59] and angiogenesis in vivo [60]. Studies have shown that at least one mechanism by which these retinoids are regulating neovascularization is through their modulation of EC production of MMPs and TIMPs [61]. Treatment of capillary EC with either retinol or retinoic acid leads to increases in EC production of the gelatinases MMP-2 and MMP-9 as well as in the levels of active MMP-1. In all three cases, retinol was a more effective stimulator of MMP production than was retinoic acid. Interestingly, biosynthetic labeling studies revealed that retinol treatment stimulated the production of TIMP-1 by EC in contrast to retinoic acid treatment which stimulated production of TIMP-2. These findings are consistent with the previously described retinoid-induced alterations of the ECM of capillary EC which were implicated in the retinoid-induced suppression of EC growth [62].
The role of thrombin as a regulator of EC function and as a potential physiologic regulator of angiogenesis has received recent attention. Of particular interest is the report that proteolytically active thrombin can induce the activation of MMP-2 in HUVECs by a process probably involving the endothelial plasma membrane [63]. This finding represents a novel activation mechanism for MMPs by EC and smooth muscle cells. The authors of this study concluded that the remodeling of the capillary endothelium required during angiogenesis can be induced by the thrombin which is produced during hemostatic processes [63]. An independent report that same year also demonstrated a TPA-sensitive, plasma membrane-dependent activation of MMP-2 in HUVECs [64]. Furthermore, this activation appeared to be dependent on MT-MMP activity, suggesting an important role for this MMP in the regulation of basement membrane degradation critical for angiogenesis.
In a study designed to analyze the role of the Type IV collagenases, MMP-2 and MMP-9 in the process of tube formation on Matrigel (without the addition of stimulators of either MMP production or of angiogenesis), it was determined that MMP-2 was produced by the EC (HUVEC) or microvascular EC, as they differentiated into capillary tube-like structures, and that this enzymatic activity could be blocked by the addition of TIMP-1 or TIMP-2 to the cultures resulting in a concomitant decrease in tube formation [65]. The addition of MMP-2 alone at low doses (~2 x 10–9 M) enhanced capillary tube formation in this model.
|
Matrix Metalloproteinase Inhibitors as Suppressors of Neovascularization
|
|---|
Some of the most convincing data documenting the requirement of MMP activity during the process of neovascularization have come from studies in which the blocking of MMP activity in both in vitro and in vivo models resulted in the inhibition of angiogenesis. Some of the earliest reports were the outgrowth of studies designed to isolate and characterize inhibitors of angiogenesis from avascular tissues, the obvious logic being that the lack of neovascularization in these tissues was most likely a function of the presence of the anti-angiogenic factors present in them. Native rabbit cartilage explants were shown to inhibit angiogenesis in vivo in the rabbit corneal pocket assay, a classic in vivo angiogenesis model system [66]. Consistent with this report was the evidence that cartilage extracts could inhibit EC proliferation in culture [67] and that these extracts contained proteolytic inhibitors [68]. Soon thereafter, partially purified cartilage extracts containing collagenase inhibitory activity were shown to inhibit neovascularization in vivo using two different animal models [69, 70]. Taken together, these studies provided the foundation for the use of MMP inhibitors as potential inhibitors of neovascularization.
During this period, in vitro angiogenesis studies were providing evidence that the suppression of MMP activity could negatively regulate EC functions. In a study designed to understand the role of different protease families in the formation of capillary-like tubular structures in vitro, a number of serine-, thiol-, carboxy-, and metalloproteinase inhibitors were tested for their effects on this system. Only the metalloproteinase inhibitor, 1,10-phenanthroline, prevented PMA-stimulated EC invasion into the collagen gels and the formation of the capillary tubes [71]. Later, in a different EC invasion assay, TIMP-1, 1,10-phenanthroline and anti-type IV and anti-interstitial collagenase antibodies were shown to inhibit capillary EC invasion through the human amniotic membrane [57].
The first report that a metalloproteinase inhibitor could inhibit angiogenesis in vivo came from a series of studies which tested the effect of a cartilage-derived TIMP. Tested in two different in vivo systems, this cartilage-derived inhibitor (CDI) was shown to be a potent inhibitor of embryonic neovascularization on the chick chorioallantoic membrane [72] as well as of tumor-induced neovascularization in the rabbit corneal pocket assay (Fig. 2) [15]. In studies designed to understand the mechanism(s) by which this TIMP might be exerting its anti-angiogenic effects, CDI was found to inhibit FGF-driven capillary EC proliferation, migration [72] and capillary tube and sprout formation (Fig. 3). In a later study, the single cell type in cartilage, the chondrocyte, was shown to secrete this anti-angiogenic TIMP which shared similar in vivo and in vitro angiogenesis inhibitory activity with its tissue-derived counterpart [73].
View larger version (288K):
[in this window]
[in a new window]
|
Figure 2. Inhibition of tumor-induced angiogenesis by a cartilage-derived TIMP. (A) Control corneas implanted with empty EVAc polymer pellet. (B) Test corneas implanted with EVAc polymer pellet impregnated with cartilage-derived TIMP. Test corneas showed significant inhibition of V2 carcinoma-induced new capillary growth. Reprinted with permission [15].
|
|
View larger version (338K):
[in this window]
[in a new window]
|
Figure 3. Inhibition of capillary endothelial cell sprout formation in vitro by a cartilage-derived TIMP (phase contrast). (A) Capillary endothelial cell aggregate displaying normal sprout formation. (B) Inhibition of sprout formation by treatment with cartilage-derived TIMP (unpublished observations, J. Sudhalter and M.A. Moses, 1991).
|
|
Subsequently, other groups have confirmed the anti-angiogenic activity of various MMP inhibitors. Minocycline, a tetracycline derivative which is an inhibitor of collagenase, was shown to inhibit neovascularization [74]. TIMP-2, but not TIMP-1, a neutralizing antibody to MMP-2 or batimastat (BB-94) was shown to inhibit basic FGF-driven capillary EC proliferation [75]. Interestingly, TIMP-1 has been shown to be a promoter of EC proliferation [76] and TIMP-3 does not inhibit capillary EC proliferation (unpublished observation, C. Fernandes, W. Wiederschain, and M.A. Moses, 1995, and personal communication, B. Anand-Apte, 1995). Alpha-2 macroglobulin, another collagenase inhibitor, is also incapable of inhibiting EC proliferation [72]. The observation that not all MMP inhibitors are capable of inhibiting microvascular EC proliferation despite their common ability to inhibit MMP activity suggests a mechanism of action distinct from that of MMP inhibition such as a receptor-mediated event or one in which EC mitogenic pathways are disrupted [75]. The recent demonstrations that plasma membrane-associated MT-MMP is expressed and synthesized by EC (HUVECs) [63] and that FGF receptor 1 may serve as a specific target for MMP-2 on the cell surface [77] may offer potential mechanisms and are currently under study.
In contrast, each of the endogenous metalloproteinase inhibitors which has been tested for its ability to inhibit EC migration as analyzed in a modified Boyden chamber assay can do so, including TIMP-1 [78], cartilage-derived TIMP [72, 73], TIMP-2 [75], and TIMP-3 (personal communication, B. Anand-Apte, 1995). Batimastat, although an inhibitor of EC invasion of a layer of Matrigel, does not inhibit EC migration [55]. That MMP activity is required for EC invasion through collagen has been documented by studies demonstrating that TIMPs-1, -2 and -3 inhibit collagen gel invasion [54].
In vivo, polyamine-induced chick yolk sac vessels were shown to be inhibited by treatment with either TIMP-1 or TIMP-2 [76]. Recombinant TIMP-1 was shown to be an inhibitor of FGF-stimulated neovascularization when tested in a rat corneal pocket assay [78]. Batimastat could inhibit the neovascularization induced by an endothelioma cell supernatant impregnated in a Matrigel pellet which was injected s.c. into mice [55].
The mechanisms by which these MMP inhibitors, both endogenous and synthetic, are regulating the process of angiogenesis in vivo are currently under intensive study. With respect to the TIMPs, the issue is a complex one, given the fact that these inhibitors are proving to be multifunctional. As mentioned above, TIMP-1 and TIMP-2 are known to have erythroid potentiating activity and are capable of stimulating the growth of human and murine erythroid precursors including burst forming units-erythroid, and more mature precursors (colony forming units-erythroid) as well as human erythroleukemia cells K562 [26, 44]. TIMP-1 has been shown to bind specifically to erythroid precursors [45] suggesting the existence of a TIMP-1 receptor-mediated growth signal. In addition, TIMP-1 also stimulates the proliferation of human keratinocytes and specifically binds to these cells with a KD of 8.7 nM [47]. Additionally, TIMP-2 has recently been shown to directly bind to the surface of a number of cells including HT-1080 and MCF-7 cell lines [79].
Understanding the structural basis of the differences in the ability of TIMPs to modulate angiogenic processes could have valuable clinical ramifications. For example, given their shared ability to inhibit MMP activity the potential therapeutic use of TIMPs and synthetic MMP inhibitors against neoplastic and other diseases of deregulated ECM remodeling has been predicated on their anti-invasive features. However, the impact of the use of these inhibitors on the angiogenic process merits consideration as well. For example, TIMP-2, as an inhibitor of FGF-stimulated EC proliferation, might have the additional feature of inhibiting angiogenesis-dependent tumor growth and dissemination via a direct antiproliferative effect on capillary EC in addition to its anti-MMP activity. On the basis of this important bioactivity, it has been suggested that TIMP-2 may be an even more effective therapeutic than TIMP-1, which does not share this antiproliferative effect on EC [75].
|
Processing of Angiogenic Modulators by MMPs
|
|---|
Historically, MMPs have been distinguished from each other on the basis of their substrate specificity. These traditional substrates include such ECM components as collagens, fibronectin and laminin among others. In a series of exciting and very recent reports, however, this paradigm of limited specificity is coming into question. For example, MMPs have now been shown to process a number of precursor or parent proteins into their bioactive form. These MMP-processed proteins include such angiogenesis-relevant ones as FGF-receptor 1 (FGFR1) [77], TNF-alpha [80-82] and the TNF-alpha receptor ectodomain [83].
Recently, MMP-2 but not MMP-9 was shown to release the active soluble ectodomain of FGFR1 which, in turn, can then modulate the angiogenic activity of FGF [77]. These studies also suggest that FGFR1 may be a specific target for MMP on the cell surface. In another series of studies, the processing of TNF-alpha, an inhibitor of EC proliferation in vitro [84], from its inactive precursor protein into its active form [80-82] and the processing of its receptor [83] have been shown to be mediated by a system sensitive to MMP inhibitors. These data suggest that MMP inhibitors might also be therapeutically useful in the treatment of diseases mediated by TNF-alpha.
Since these first reports, evidence has been found to suggest that candidate substrates of MMPs may encompass an even greater group of biologically important factors. Arribas and coworkers have recently demonstrated that diverse cell surface protein ectodomains are released by a system sensitive to metalloprotease inhibitors [85]. They showed that two structurally distinct agents, TAPI-2 and 1,10-phenanthroline, specific MMP inhibitors, blocked the PMA-activated shedding of proTGF-alpha, the cell adhesion receptor L-selectin, interleukin 6 receptor alpha subunit, and B-amyloid precursor protein.
Finally, it has been shown recently that at least one MMP, a macrophage metalloelastase, is capable of processing angiostatin, a 38 kD internal fragment of plasminogen which has been shown to have potent anti-angiogenic and antimetastatic activity [86], from its inactive parent molecule to its active, anti-angiogenic form [87]. Taken together, these studies suggest an additional mechanism by which MMPs may be playing an important role in the regulation of neovascularization.
|
Future Considerations
|
|---|
The process of angiogenesis, like that of tumorigenesis and metastasis, has been described as being particularly sensitive to three biological events: a loss of growth control, an imbalanced regulation of proteolysis and an imbalanced regulation of cellular motility [7]. The data presented above clearly document the fact that the deregulation of one particular family of proteinases, the MMPs, can have a profound effect on the formation of new capillaries. Historically, the classic model of the role of MMPs during angiogenesis suggested that a shift in the endogenous proteolytic balance in favor of MMP activity was a "pro-angiogenic" event whereas a shift in favor of MMP inhibitors was "anti-angiogenic." However, this simplest of paradigms warrants reconsideration in light of some recent studies which challenge this model.
For example, given the ability of a Lewis Lung carcinoma-derived macrophage metalloelastase to process angiostatin from plasminogen [87], the possibility exists that MMPs may actually be exerting an unexpected anti-angiogenic effect via their ability to process anti-angiogenic factors. This is of particular interest when one considers the fact that several angiogenesis inhibitors are themselves fragments of larger molecules [88]. Applied to the clinical arena, one might imagine a scenario whereby administration of inhibitors of proteolysis and the subsequent decrease in proteolytic activity could result in the decreased production of angiogenesis inhibitors which, in turn, might cause a stimulation of tumor growth and metastasis—the opposite of the desired effect. With respect to the role of TIMPs in regulating angiogenesis, it is important to consider the fact that the mechanisms by which these inhibitors may be modulating new capillary growth are not the same. For example, TIMP-2 is an inhibitor of FGF-driven capillary EC proliferation whereas TIMP-1 is a modest stimulator of EC growth [75], and TIMP-3 (unpublished observation, C. Fernandes, W. Wiederschain, and M.A. Moses, 1995, and personal communication, B. Anand-Apte, 1995) and BB-94 [55, 75] appear to show no effect on EC growth. In order to prudently and specifically design the most appropriate settings for the clinical use of MMP inhibitors, it will be essential to understand the mechanisms by which these inhibitors are exerting their regulatory effect on EC functions in vitro and angiogenesis in vivo.
|
References
|
|---|
-
Hobson B, Denekamp J. Endothelial proliferation in tumors and normal tissues: continuous labelling studies. Br J Cancer 1984;49:405-413.[Medline]
-
Folkman J, Klagsbrun M. Angiogenic factors. Science 1987;235:442-447.[Abstract/Free Full Text]
-
Herron GS, Banda MJ, Clark EJ et al. Secretion of metalloproteinases by stimulated capillary endothelial cells. II. Expression of collagenase and stromelysin activities is regulated by endogenous inhibitors. J Biol Chem 1986;261:2814-2818.[Abstract/Free Full Text]
-
Pepper MS, Montesano R. Proteolytic balance and capillary morphogenesis. Cell Differ Dev 1990;32:319-328.[Medline]
-
Montesano R, Pepper MS, Mohle-Steinlein U et al. Increased proteolytic activity is responsible for the aberrant morphogenic behavior of endothelial cells expressing the middle T oncogene. Cell 1990:62:435-445.[Medline]
-
Moses MA, Langer R. A metalloproteinase inhibitor as an inhibitor of neovascularization. J Cell Biochem 1991;47:1-6.[Medline]
-
Liotta LA, Steeg PS, Stetler-Stevenson W. Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation. Cell 1991;64:327-336.[Medline]
-
Ingber DE, Madri JA, Folkman J. A possible mechanism for inhibition of angiogenesis by angiostatic steroids: induction of capillary basement membrane dissolution. Endocrinology 1986;119:1768-1775.[Abstract]
-
Mignatti P, Rifkin DB. Plasminogen activators and matrix metalloproteinases in angiogenesis. Enz Prot 1996;49:117-137.
-
Pepper MS, Montesano R, Mandriota SJ et al. Angiogenesis: a paradigm for balanced extracellular proteolysis during cell migration and morphogenesis. Enz Prot 1996;49:138-162.
-
Risau W, Lemmon V. Changes in the vascular extracellular matrix during embryonic vasculogenesis and angiogenesis. Dev Biol 1988;125:441-450.[Medline]
-
Risau W. Embryonic angiogenesis factors. Pharmacol Ther 1991;51:371-376.[Medline]
-
Ausprunk DH, Folkman J. Migration and proliferation of endothelial cells in preformed and newly formed blood vessels during tumor angiogenesis. Microvasc Res 1977;14:53-65.[Medline]
-
Kalebic T, Garbisa S, Glaser B et al. Basement membrane collagen: degradation by migrating endothelial cells. Science 1983;221:281-283.[Abstract/Free Full Text]
-
Moses MA, Langer R. Angiogenesis inhibitors. Biotechnology 1991;9:630-639.[Medline]
-
Blood CH, Zetter BR. Tumor interactions with the vasculature: angiogenesis and tumor metastasis. Biochim Biophys Acta 1990;1032:89-118.[Medline]
-
Crum R, Szabo S, Folkman J. A new class of steroids inhibits angiogenesis in the presence of heparin or a heparin fragment. Science 1985;230:1375-1378.[Abstract/Free Full Text]
-
Nguyen M, Shing Y, Folkman J. Quantitation of angiogenesis and antiangiogenesis in the chick embryo chorioallantoic membrane. Microvasc Res 1994;47:31-40.[Medline]
-
Gimbrone M, Cotran R, Folkman J. Tumor growth neovascularization: an experimental model using the rabbit cornea. J Natl Cancer Inst 1974;52:413-427.
-
Mutan VR, Auerbach R. Angiogenesis in the mouse cornea. Science 1979;205:1418-1419.[Abstract/Free Full Text]
-
Fournier GA, Lutty GA, Watt S et al. A corneal micropocket assay for angiogenesis in the rat eye. Invest Ophthalmol Vis Sci 1981;21:351-354.[Abstract/Free Full Text]
-
Kenyon BM, Voest EE, Chen CC et al. A model of angiogenesis in the mouse cornea. Invest Ophthalmol Vis Sci 1996;37:1625-1632.[Abstract/Free Full Text]
-
Matrisian L. Metalloproteinases and their inhibitors in matrix remodeling. Trends Genet 1990;6:121-125.[Medline]
-
Woessner JF. The family of matrix metalloproteinases. Ann NY Acad Sci 1994;732:11-21.[Medline]
-
Birkedal-Hansen H. Proteolytic remodeling of extracellular matrix. Curr Opin Cell Biol 1995;7:728-735.[Medline]
-
Docherty AJP, Lyons A, Smith BJ et al. Sequence of human tissue inhibitors of metalloproteinases and its identity to erythroid potentiating activity. Nature 1985;318:66-69.[Medline]
-
Carmichael DF, Sommer A, Thompson RC et al. Primary structure and cDNA cloning of human fibroblast collagenase inhibitor. Proc Natl Acad Sci USA 1986;83:2407-2411.[Abstract/Free Full Text]
-
Murray JB, Allison K, Sudhalter J et al. Purification and partial amino acid sequence of a bovine cartilage-derived collagenase inhibitor. J Biol Chem 1986;261:4154-4159.[Abstract/Free Full Text]
-
Stetler-Stevenson WG, Krutzch HC, Liotta LA. Tissue inhibitor of metalloproteinases (TIMP-2). A new member of the metalloproteinase inhibitor family. J Biol Chem 1989;264:17374-17378.[Abstract/Free Full Text]
-
DeClerck TA, Yean TD, Ratzkin BJ et al. Purification and characterization of two related but distinct metalloproteinase inhibitors secreted by bovine aortic endothelial cells. J Biol Chem 1989;264:17445-17453.[Abstract/Free Full Text]
-
Pavloff N, Staskus PW, Kishnani NS et al. A new inhibitor of metalloproteinases from chicken: ChIMP-3. J Biol Chem 1992;67:17321-17326.
-
Williamson RA, Martson FAO, Angal S et al. Disulfide bond assignment in human tissue inhibitor of metalloproteinases (TIMP). Biochem J 1990;268:267-274.[Medline]
-
Cawston TE, Murphy G, Mercer E et al. The interaction of purified rabbit bone collagenase with purified rabbit bone metalloproteinase inhibitor. Biochem J 1983;211:313-318.[Medline]
-
Welgus HG, Jeffery JJ, Eisen AZ et al. Human skin fibroblast collagenase: interaction with substrate and inhibitor. Coll Relat Res 1985;5:167-179.[Medline]
-
Murphy G, Koklitis P, Carne AF. Dissociation of tissue inhibitor of metalloproteinases (TIMP) from enzyme complexes yield fully active inhibitor. Biochem J 1989;261:1031-1034.[Medline]
-
Ward RV, Hembry RM, Reynolds JJ et al. The purification of tissue inhibitor of metalloproteinase-2 from its 72 kDa progelatinase complex. Demonstration of biochemical similarities. Biochem J 1991;278:179-187.
-
Howard EW, Bullen EC, Banda MJ. Preferential inhibition of 72- and 92-kDa gelatinases by tissue inhibitors of metalloproteinase-2. J Biol Chem 1991;266:13064-13069.[Abstract/Free Full Text]
-
Goldberg GI, Marmer BL, Grant GA et al. Human 72-kDa type IV collagenase forms a complex with a tissue inhibitor of metalloproteinase designated TIMP-2. Proc Natl Acad Sci USA 1989;86:8207-8211.[Abstract/Free Full Text]
-
Wilhelm SM, Collier IE, Marmer BL et al. SV40-transformed human fibroblasts secrete a 92-kDa type IV collagenase which is identical to that secreted by normal human macrophages. J Biol Chem 1989;264:17213-17221.[Abstract/Free Full Text]
-
Khokha R, Waterhouse P, Yagel S et al. Antisense RNA-induced reduction of murine TIMP levels confers oncogenicity on Swiss 3T3 cells. Science 1989;243:947-950.[Abstract/Free Full Text]
-
Hicks NJ, Ward RV, Reynolds JJ. A fibrosarcoma model derived from mouse embryo cells: growth properties and secretion of collagenase and a metalloproteinase inhibitor (TIMP) by tumor cell lines. Int J Cancer 1984;33:835-844.[Medline]
-
Schultz RM, Silberman S, Percky B. Inhibition of human recombinant tissue inhibitor of metalloproteinases of human amnion invasion by lung colonization by murine B16-F10 melanoma cells. Cancer Res 1988;48:5539-5545.[Abstract/Free Full Text]
-
DeClerck YA, Perez N, Shimada H et al. Inhibition of invasion and metastasis in cells transfected with an inhibitor of metalloproteinases. Cancer Res 1992;52:701-708.[Abstract/Free Full Text]
-
Gasson JC, Golde DW, Kaufman SB et al. Molecular characterization and expression of the gene encoding human erythroid-potentiating activity. Nature 1985;315:768-771.[Medline]
-
Alvalos BR, Kaufman SE, Tomonaga M et al. K562 cells produce and respond to human erythroid-potentiating activity. Blood 1988;71:1720-1725.[Abstract/Free Full Text]
-
Hayakawa T, Yamashita K, Tanzawa K et al. Growth-promoting activity of tissue inhibitor of metalloproteinase-1 (TIMP-1) for a wide range of cells. A possible new growth factor in serum. FEBS Lett 1992;298:29-32.[Medline]
-
Bertaux B, Hornebeck W, Eisen AZ et al. Growth stimulation of human keratinocytes by tissue inhibitor of metalloproteinases. J Invest Dermatol 1991;97:679-685.[Medline]
-
Bodden MK, Harber GJ, Birkedal-Hansen B et al. Functional domains of human TIMP-1 (tissue inhibitor of metalloproteinases). J Biol Chem 1994;269:18943-18952.[Abstract/Free Full Text]
-
Montesano R, Orci L, Vassalli P. In vitro rapid organization of endothelial cells into capillary-like networks is promoted by collagen matrices. J Cell Biol 1983;97:1648-1652.[Abstract/Free Full Text]
-
Moscatelli D, Presta M, Rifkin DB. Tetradecanoyl phorbol acetate stimulates latent collagenase production by cultured human endothelial cells. Cell 1980;20:343-351.[Medline]
-
Gross JL, Moscatelli D, Jaffe EA et al. Plasminogen activator and collagenase production by cultured capillary endothelial cells. J Cell Biol 1982;95:974-981.[Abstract/Free Full Text]
-
Herron GS, Werb Z, Dwyer D et al. Secretion of metalloproteinases by stimulated capillary endothelial cells. I. Production of procollagenase and prostromelysin exceeds expression of proteolytic activity. J Biol Chem 1986;261:2810-2813.[Abstract/Free Full Text]
-
Unemori EN, Bouhana KS, Werb Z. Vectorial secretion of extracellular matrix proteins, matrix degrading proteinases and tissue inhibitor of metalloproteinases by endothelial cells. J Biol Chem 1990;265:445-451.[Abstract/Free Full Text]
-
Fisher C, Gilbertson-Beadling S, Powers EA et al. Interstitial collagenase is required for angiogenesis in vitro. Dev Biol 1994;162:499-510.[Medline]
-
Taraboletti G, Garafola A, Belotti D et al. Inhibition of angiogenesis and murine hemangioma growth by batimastat, a synthetic inhibitor of matrix metalloproteinases. J Natl Cancer Inst 1995;87:293-298.[Abstract/Free Full Text]
-
Unemori EN, Ferrara N, Baurer EA et al. Vascular endothelial growth factor induces interstitial collagenase expression in human endothelial cells. J Cell Physiol 1991;153:557-562.
-
Mignatti P, Tsuboi R, Robbins E et al. In vitro angiogenesis on the human amniotic membrane: requirement for basic fibroblast growth factor-induced proteinases. J Cell Biol 1989;108:671-678.[Abstract/Free Full Text]
-
Hanemaaijer R, Koolwijk P, Le Clercq L et al. Regulation of matrix metalloproteinase expression in human vein microvascular endothelial cells. Biochem J 1993;296:803-809.
-
Braunhut SJ, Gudas LJ, Kurokawa T et al. Expression of FGF by F9 teratocarcinoma cells as a function of differentiation. J Cell Biol 1989;108:2467-2476.[Abstract/Free Full Text]
-
Ingber D, Folkman J. Inhibition of angiogenesis through modulation of collagen metabolism. Lab Invest 1988;59:44-51.[Medline]
-
Braunhut SJ, Moses MA. Retinoids modulate endothelial cell production of matrix-degrading proteases and tissue inhibitors of metalloproteinases (TIMP). J Biol Chem 1994;269:13472-13479.[Abstract/Free Full Text]
-
Paige K, Palomares M, D'Amore PA. Retinol-induced modification of the extracellular matrix of endothelial cells: its role in growth control. In Vitro Cell Dev Biol 1991;27:151-157.
-
Zucker S, Conner C, DiMassmo BI et al. Thrombin induces the activation of progelatinase A in vascular endothelial cells. J Biol Chem 1995;270:23730-23738.[Abstract/Free Full Text]
-
Lewalle JM, Munaut C, Pichot B et al. Plasma membrane-dependent activation of gelatinase A in human vascular endothelial cells. J Cell Physiol 1995;165:475-483.[Medline]
-
Schnaper HW, Grant DS, Stetler-Stevenson WG et al. Type IV collagenase(s) and TIMPs modulate endothelial cell morphogenesis in vitro. J Cell Physiol 1993;156:235-246.[Medline]
-
Brem H, Folkman J. Inhibition of tumor angiogenesis mediated by cartilage. J Exp Med 1975;41:427-439.
-
Eisenstein R, Kuettner KE, Neopolitan C et al. The resistance of certain tissues to invasion. III. Cartilage extracts inhibit the growth of fibroblasts and endothelial cells in culture. Am J Pathol 1975;81:337-347.[Abstract]
-
Sorgente N, Kuettner KE, Soble L et al. The resistance of certain tissues to invasion. II. Evidence for extractable factors in cartilage which inhibit invasion by vascularized mesenchyme. Lab Invest 1975;32:217-222.[Medline]
-
Langer R, Brem H, Falterman R et al. Isolation of a cartilage factor that inhibits tumor neovascularization. Science 1976;193:70-72.[Abstract/Free Full Text]
-
Langer R, Conn H, Vacanti J et al. Control of tumor growth by infusion of an angiogenesis inhibitor. Proc Natl Acad Sci USA 1980;77:4331-4335.[Abstract/Free Full Text]
-
Montesano R, Orci L. Tumor-promoting phorbol esters induce angiogenesis in vitro. Cell 1985;42:469-477.[Medline]
-
Moses MA, Sudhalter J, Langer R. Identification of an inhibitor of neovascularization from cartilage. Science 1990;2488:1408-1410.
-
Moses MA, Sudhalter J, Langer R. Isolation and characterization of an inhibitor of neovascularization from scapular chondrocytes. J Cell Biol 1992;119:474-481.
-
Tamargo RJ, Bok RA, Brem H. Angiogenesis inhibition by minocycline. Cancer Res 1991;51:672-675.[Abstract/Free Full Text]
-
Murphy AN, Unsworth EJ, Stetler-Stevenson WG. Tissue inhibitor of metalloproteinases-2 inhibits bFGF-induced human microvascular endothelial cell proliferation. J Cell Physiol 1993;157:351-358.[Medline]
-
Takigawa M, Nishida Y, Suzuki F et al. Induction of angiogenesis in chick yolk-sac membrane by polyamines and its inhibition by tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2). Biochem Biophys Res Commun 1990;171:1264-1271.[Medline]
-
Levi E, Fridman R, Miao H et al. Matrix metalloproteinase 2 releases active soluble ectodomain of fibroblast growth factor receptor 1. Proc Natl Acad Sci USA1996;93:7069-7074.[Abstract/Free Full Text]
-
Johnson MD, Kim HC, Chesler L et al. Inhibition of angiogenesis by tissue inhibitor of metalloproteinase. J Cell Physiol 1994;160:194-202.[Medline]
-
Emmert-Buck MR, Emonard HP, Corcoran ML et al. Cell surface binding of TIMP-2 and pro-MMP-2/TIMP-2 complex. FEBS Lett 1995;364:28-32.[Medline]
-
Gearing AJH, Beckett P, Christodoulou M et al. Processing of tumor necrosis factor-alpha precursor by metalloproteinases. Nature 1994;370:555-557.[Medline]
-
McGeehan GM, Becherer JD, Bast RC et al. Regulation of tumor necrosis-alpha processing by a metalloproteinase inhibitor. Nature 1994;370:558-561.[Medline]
-
Mohler KM, Sleath PR, Fitzner JN et al. Protection against a lethal dose of endotoxin by an inhibitor of tumour necrosis factor processing. Nature 1994;370:218-220.[Medline]
-
Crowe P, Walter BN, Mohler KM et al. A metalloproteinase inhibitor blocks shedding of the 80-kD TNF receptor and TNF processing in T lymphocytes. J Exp Med 1995;181:1205-1210.[Abstract/Free Full Text]
-
Frater-Schroder M, Risau W, Hallman R et al. Tumor necrosis factor type alpha, a potent inhibitor of endothelial cell growth in vitro. Proc Natl Acad Sci USA 1987;84:5277-5281.[Abstract/Free Full Text]
-
Arribas J, Coodly L, Vollmer P et al. Diverse cell surface protein ectodomains are shed by a system sensitive to metalloprotease inhibitors. J Biol Chem 1996;271:11376-11382.[Abstract/Free Full Text]
-
O'Reilly MS, Holmgren L, Shing Y et al. Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 1994;79:315-328.[Medline]
-
Dong Z, Kumar R, Fidler IJ. Generation of the angiogenesis inhibitor, angiostatin, by Lewis lung carcinoma is mediated by macrophage elastase. Proc Am Assoc Cancer Res 1996;37:403a.
-
Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 1996;86:353-364.[Medline]
accepted for publication December 27, 1996.
Top
Abstract
Introduction
