201. Wild-type versus mutant MMP-8 in melanoma: ‘When you come to a fork in the road, take it’
- Author
-
Jessica S. Blackburn and Constance E. Brinckerhoff
- Subjects
Tumor microenvironment ,Tumor suppressor gene ,Angiogenesis ,Melanoma ,Mutant ,Wild type ,Dermatology ,Biology ,medicine.disease ,General Biochemistry, Genetics and Molecular Biology ,Immune system ,Oncology ,Mutant protein ,Cancer research ,medicine - Abstract
Matrix metalloproteinases (MMPs) are a family of at least 20 enzymes that are traditionally noted for their ability to degrade all components of the extracellular matrix, thereby, assigning them critical roles in connective tissue remodeling in normal physiology and in disease pathology (Reviewed in Brinckerhoff and Matrisian, 2002; Vincenti and Brinckerhoff, 2007). While levels of MMPs are usually low during homeostasis, conventional wisdom holds that their overexpression and dysregulation in many diseases, including cancer, account for their prominent role in disease progression. In contrast, the current paper by Pallavalli et al. (2009) postulates the exact opposite that over-expression of a specific MMP is actually protective against aggressive melanoma. The authors carried out mutational analyses of eight MMP genes in melanomas from 32 patients, and identified 28 somatic mutations. They found that MMP-8, a collagenase, was one of the most commonly mutated genes, and identified five distinct mutations. Importantly, these mutated forms of MMP-8 exhibited reduced collagenolytic activity, but cells expressing mutant MMP-8 showed increases in tumor growth on soft agar in vitro and in tumor formation in vivo, compared to cells with wild type MMP-8. Thus, mutant MMP-8 is associated with aggressive melanoma, leading the authors to conclude that wild-type MMP-8 is protective against invasive melanoma, and may function as a tumor suppressor gene. The notion that MMPs have important functions unrelated to matrix degradation is not new (Brinckerhoff and Matrisian, 2002; Vincenti and Brinckerhoff, 2007). They contribute to crucial events in disease pathology, such as cell proliferation, differentiation, apoptosis, and angiogenesis, thereby illustrating their ability to modulate a multitude of cell behaviors. In addition, it is well recognized that some level of MMP activity is essential for normal physiology, and knocking out a particular MMP in mice has demonstrated that the complete absence of a certain MMP can be associated with an unfavorable phenotype (Vincenti and Brinckerhoff, 2007). In keeping with the present study, knock out of MMP-8 in mice was associated with increased skin tumor susceptibility and with enhanced metastasis of melanoma (Gutierrez-Fernandez et al., 2008; Vincenti and Brinckerhoff, 2007). In the skin tumor model, the MMP-8 null mice displayed decreased inflammation suggesting that MMPs may have a protective role vis-a-vis the immune ⁄ inflammatory responses. In the melanoma model, tumor cells expressing wild-type MMP-8 showed increased adhesion to matrix proteins, which was absent in cells expressing mutant MMP-8. From this, the authors conclude that MMP-8 potentiates cell adhesion, which may explain how this proteinase reduces cell invasion (Gutierrez-Fernandez et al., 2008). Interestingly, the authors also hint at a possible role for MMP-8 in an intact immune system, claiming that secretion of MMP-8 by tumor-infiltrating neutrophils in wild type (versus MMP-8 null) mice could contribute to reduced tumor dissemination (Gutierrez-Fernandez et al., 2008). Thus, the findings that expression of a mutant, non-collagenolytic protein in human melanomas is associated with increased tumor growth in vitro and increased tumor formation in vivo confirm the prior experimental results in mice. They are, therefore, not totally surprising. However, their observations take several important steps forward by extending the experimental studies to actual patient samples, and by documenting numerous mutations in the coding region of human MMP-8 isolated from melanomas. Their data are extensive and compelling, clearly showing by over expressing mutant protein and knocking down wild-type, that wild-type protein is protective and that mutant is linked to melanoma progression. The concept of proteinases functioning as tumor suppressors has been suggested previously (Lopez-Otin and Matrisian, 2007). Nonetheless, Palavalli et al. now formally propose that MMP-8 functions as a tumor suppressor gene. Their claim is supported by two salient observations: (1) in seven tumors, both alleles of MMP-8 were affected, while other mutations in the gene were accompanied by loss of heterozygosity, both of which are characteristics associated with tumor suppressor genes, and (2) there was a higher than expected ratio of non-synonymous to synonymous somatic mutations suggesting that they are ‘driver mutations,’ which are dominant. Thus, ‘tumor suppressor’ may be an important function that can be added to the repertoire of activities associated with at least one MMP family member. The mutations in MMP-8 were found in the proteglycan-binding and catalytic domains, and both abrogated collagenolytic activity. Furthermore, as cells expressing mutant MMP-8 were more aggressive, it follows that this aggressive behavior is unrelated to the ability of MMP-8 to degrade collagen. However, the mechanism(s) by which MMP-8 exerts its putative tumor suppressive effects is unclear. The suggestion that MMP-8 null mice have compromised immune ⁄ inflammatory responses raises the possibility that this mechanism is operational here. The fact that severely immuno-compromised NOD ⁄SCID mice were used in the in vivo experiments with human melanomas may preclude assessment of effects of the immune system. This contrasts with studies in the MMP-8 null mice, which harbor an intact immune system, with the exception of no MMP-8 expression by the host cells (Gutierrez-Fernandez et al., 2008). Thus, in keeping with previous studies that implicate a protective effect of MMPs on the immune ⁄ inflammatory responses, this remains an intriguing, and reasonably likely, possibility for MMP-8 in human melanoma. A corollary to this concept is the importance of tumor–host cell interactions, and their influence on the tumor microenvironment. This corollary has raised the possibility that wild-type MMP-8 has novel function(s) in suppressing melanoma. It has *Yogi Berra Coverage on: Palavalli, L.H., Prickett, T.D., Wunderlich, J.R., Wei, X., et al. (2009). Analysis of the matrix metalloproteinase family reveals that MMP8 is often mutated in melanoma. Nat. Genet. Advanced onlinepublication,doi:10.1038/ng.340.
- Published
- 2009