In the process of embryologic development precursor cells undergo differentiation to specific cellular fates through interaction with signaling molecules delivered by the extracellular milieu. These developmental cues are derived from both soluble components (growth factors and cytokines) as well as structural elements (collagens, fibronectins, laminins, etc.) of the extracellular matrix (ECM). The result is a highly differentiated organism that utilizes homeostatic mechanisms to assure that order and function are maintained in its tissues. Disruption of tissue homeostasis results in chronic disease states, such as cancer. It is now well recognized that the genetic changes critical for tumor development, e.g. oncogene expression and loss of tumor suppressor gene expression, are not associated with development of tumor metastasis, but it is the tumor microenvironment that plays a critical role in the evolution of the metastatic phenotype [Bhowmick, 2004 #97; Bissell, 2005 #96; Gupta, 2006 #100; Witz, 2008 #99]. The critical role of the tumor microenvironment and the inefficiency of the metastatic process imply that organ specific homeostatic mechanisms, that maintain normal tissue integrity, are robust and inimical to invasive tumor cells. This concept was elegantly demonstrated by Mintz and Illmensee who showed that injection of malignant murine teratocarcinoma cells into a mouse blastocysts resulted in normal, non-malignant tissue formation in genetically mosaic mice [Mintz, 1975 #101]. More recently, Hendrix and colleagues have demonstrated that the human embryonic stem cell microenvironment suppresses the tumorigenic phenotype of malignant melanoma [Postovit, 2008 #152]. Thus the microenvironment of normal tissues posses conserved barriers that must be overcome during formation of the tumor microenvironment at the primary tumor site, during tumor cell invasion, as well as in the subsequent establishment of metastatic foci [Bissell, 2005 #96; Gupta, 2006 #100; Witz, 2008 #99] Formation of a tumor microenvironment that facilitates tumor progression is complex and requires the removal of intrinsic physical, biological and chemical barriers. This process is initiated by protoelytic remodeling of the ECM and in particular the basement membrane [Kalluri, 2003 #98]. These barriers have been described as selective pressures for the metastatic phenotype [Gupta, 2006 #100; Witz, 2008 #99]. It is now well recognized that both the malignant tumor cells as well as a variety of host responses, such as angiogenesis and inflammation, are responsible for the evolution of the tumor microenvironment. Furthermore, specific genes play dual roles in that they both antagonize and promote tumor progression [Bhowmick, 2004 #97; Gupta, 2006 #100; Witz, 2008 #99]. TGF-beta, secreted by stromal fibroblasts, is a prototypic example [Bhowmick, 2004 #97; Witz, 2008 #99]. While the changes in tumor cell responses to TGF-beta may involve loss-of-function mutations in receptors and downstream signaling elements, the mechanisms involved in these processes are still not fully understood and may be tumor specific. Given that TIMP-2, like TGF-beta, has multiple roles in the extracellular matrix, it is the goal of this laboratory to identify, and understand the mechanisms and inter-relationships of these multiple functions in tissue homeostasis as well as in the tumor microenvironment. TIMP-2 is by definition an inhibitor of matrix metalloproteinases [DeClerck, 1991 #160; Stetler-Stevenson, 1989 #21]. Formation of a pro-MMP-2/TIMP-2 complex promotes formation of a ternary complex with MT1-MMP that results in cell surface activation of the pro-MMP-2, and subsequent release of the both activated MMP-2 and the inhibitor [Baker, 2002 #103; Lambert, 2004 #12].It is well known that TIMP-2 is anti-tumorigenic. Forced expression of TIMP-2 inhibits tumor cell invasion in vitro and in vivo, and reduces metastasis formation [Albini, 1991 #114; DeClerck, 1991 #161; DeClerck, 1992 #115; Montgomery, 1994 #117]. More recent studies on human tumor samples reveal that TIMP-2 levels can be reduced by genetic mechanisms (e.g. the G418C promoter polymorphism resulting in loss of an Sp1 site and reduced TIMP-2 expression) as observed in gastric cancers, as well as squamous cell carcinoma of the head and neck [Kubben, 2006 #141; P, 2006 #155; Vairaktaris, 2007 #154; Yang, 2008 #153]. Alternatively, TIMP-2 levels may also be reduced by epigenetic mechanisms, i.e. promoter hypermethylation, as reported for human lymphomas, cervical and prostate cancers [Galm, 2005 #140; Ivanova, 2004 #156; Pulukuri, 2007 #139]. We have proposed that free TIMP-2 levels observed in normal tissues are also reduced in the tumor microenvironment through interactions with the active sites of MMPs that are highly expressed at the tumor stromal interface and are associated with tumor progression [Stetler-Stevenson, 2008 #159]. Restoration of TIMP-2 protein levels by a variety of methods, such as forced expression in tumor cells, gene therapy approaches using adenoviral expression, or the more recently reported TIMP-2 transgenic mouse, resulted in reduced tumor growth and angiogenesis, as well as a reduction in tumor growth and metastasis formation [Blavier, 2006 #157; Brand, 2000 #158]. In some cases tumor growth was reduced by as much as 10-fold [Valente, 1998 #123]. Finally, TIMP-2 binds to the alpha3beta1 integrin to suppress cell growth in a variety of normal and neoplastic cell types [Baker, 2002 #103; Hoegy, 2001 #48; Jaworski, 2005 #87; Jaworski, 2006 #88; Jaworski, 2006 #85; Lambert, 2004 #12; Oh, 2006 #81; Oh, 2004 #83; Perez-Martinez, 2005 #84; Seo, 2003 #49; Seo, 2006 #80]. Specific domains have been shown to mediate the first two biological activities, MMP inhibition and pro-MMp-2 activation. Like all members of the TIMP family, it is the amino group of the N-terminal cysteine residue coordinating with the zinc atom at the active site of MMP family members that is responsible for their inhibitory activity [Gomis-Rth, 1997 #9]. As we have shown using our mutant Ala+TIMP-2, blocking the interaction of the N-terminal cysteine with the zinc, effectively abrogates TIMP-2 MMP inhibitory activity [Hoegy, 2001 #48; Wingfield, 1999 #47]. The C-terminal domain of TIMP-2 binds to the hemopexin domain of pro-MMP-2 to mediate formation of the pro-enzyme/inhibitor complex [Baker, 2002 #103; Lambert, 2004 #12]. The free N-terminus of the TIMP-2 in this bimolecular complex interacts with the active site of the cell surface MT1-MMP to mediate formation of a trimolecular complex of pro-MMP-2/TIMP/MT1-MMP. Subsequently, recruitment of a second MT1-MMP mediates activation of pro-MMP-2, release of the active enzyme and subsequent release of the intact inhibitor TIMP-2. Finally, the C-terminal domain of TIMP-2, specifically disulfide-bonded loop 6, has anti-angiogenic activity, however, the mechanism of this effect has not been completely elucidated [Fernandez, 2003 #52]. These findings suggest that defining the domain(s) responsible for TIMP-2-binding to alpha3beta1 will be critical to further dissecting the multiple biological activities of this complex molecule, as well as defining the functional contributions of this activity to the microenvironment in both normal and malignant tissues.
Showing the most recent 10 out of 34 publications
