The potentially important role of matrix-degrading enzymes in breast cancer has been stated for many years, in particular in relation to the activities of MMP-2 and MMP-9 [14
]. Indeed, both collagenolytic enzymes have been individually correlated with breast cancer progression [19
] and tumor vascularization, invasion and metastasis, and differentiation and proliferation [20
]. We initially aimed to verify the occurrence of gelatinase activity in 30 tissue samples of breast carcinomas vs. the adjacent non-affected tissues. The results demonstrated that both MMP-9 and MMP-2 were present, as pro- and active forms in the majority of the tumoral tissues, and absent or detectable at very low levels in the adjacent non-affected paired tissues.
This result encouraged us to continue investigating the expression of the lytic activities of the pro- and active forms of MMP-2 and MMP-9 on a larger number of cases, consisting of 80 surgical tissue fragments.
These two collagenolytic enzymes have been traditionally considered to be the most important promoters of tumor invasion, because of their ability to degrade the first tissue barrier, i.e., the basal lamina. This, as is well known, has the role of maintaining the right tissue architecture and to prevent cell mobility of stationary tissues, such as epithelial tissue.
The first question of our approach was to understand if the two enzymatic activities were interrelated. The results showed a significant Pearson correlation between the pro-enzymatic forms and their corresponding activated forms; while a moderate correlation was found between the two active enzymes. These observations suggest that the two enzymes can act in an independent manner, but in some way complementary.
In the belief that genes and proteins never operate alone, we wanted to search for potential interactors of the two enzymes. For this purpose, we used the STRING platform that builds functional protein association networks based on literature evidence.
An impressive result was the observation of a noticeably high number of hypothetical functional interactions at several levels. The first level concerns the primary interactions that each enzyme can establish with its putative partners. The second level concerns the secondary interactions, and beyond, and presumes the presence of intermediaries among the functional interactions.
4.1. Common Interactors for MMP-2 And MMP-9
The first-level interactions were 20 contacts for MMP-9, and 11 for MMP-2. A point of interest was the observation that six contacts are shared between the two enzymes, namely: GAPDH, ANXA5, LGAL3, S100A4, DECR1 and VIM.
Glyceraldehyde-3-phosphate dehydrogenase, G3P (GAPDH).
Among the shared interactors, the presence of the G3P, which is generally over-expressed in breast cancer, as already reported in our previous studies [21
], deserves particular attention. The G3P, besides being a key enzyme of glycolysis, under some circumstances, can translocate into the nucleus and act as a co-promoter for cell proliferation [22
]. Its putative relationship with the MMPs has never reported before.
Annexin A5, ANXA5 (ANX5A).
This protein is a member of a multigene family consisting of 13 members of Ca2+
and phospholipid binding proteins with peripheral membrane location. In our previous proteomics study, we found a ubiquitous and over-expressed presence of ANXA5, which is known to play an anti-apoptotic role. Similarly to other members of the family, it is thought to be involved at different levels in the tumor development, progression and invasivity [24
]. Moreover, Annexin A5 is known to exert an anticoagulant effect, acting as an indirect inhibitor of the thromboplastin-specific complex [27
Galectin, LEG3 (LGALS3)
. Galectin is a galactose-specific lectin associated with the cell membrane and involved in several processes of membrane trafficking by interacting with other surface proteins [28
]. Intracellularly, it participates in the cytoskeleton organization and cell motility by interacting with ACTB and other associated proteins (COF1, EZRI, MIF, and CAPG). Extracellularly, it interacts with a variety of cell surface glycoproteins, and in cancer, it seems to participate in the dynamics of cell migration and in escaping the T cell-mediated immune response [29
The S100 proteins represent a multigene family of calcium-binding proteins of the EF-hand type, encoded by 21 genes in humans [30
]. Members of the S100 family are differentially expressed in normal tissues and are frequently upregulated in cancer [31
]. They may perform a large variety of functions, either intracellularly or extracellularly, in a cytokine-like manner through the receptor for “advanced glycation end products” (RAGE) [32
]. In particular, the S100A4 protein is known to be secreted by tumor and/or stromal cells to support tumorigenesis by stimulating angiogenesis and promoting endothelial cell migration [33
In addition, it has been reported that S100A4 may support MMP-9 and MMP-13 gene expression [34
], and also that it may enhance the activity of some MMPs, causing higher cell dissociation and cancer metastasis [35
Vimentin, VIME (VIM).
Another interesting partner of the MMPs appears to be VIM. Vimentins are class-III intermediate filaments found in various non-epithelial cells, especially mesenchymal cells. Its high expression in breast cancer testifies to the occurrence of the epithelial-mesenchyme transition, already postulated at the proteomic level [21
]. Its partnership with MMP-2 and -9 is suggestive for joined mechanisms of matrix degradation and cell migration.
2,4-Dienoyl-CoA reductase, DECR (DECR1). Somewhat ambiguous is the involvement of DECR1 2,4-dienoyl-CoA reductase, a mitochondrial auxiliary enzyme of beta-oxidation, for which no reports have been produced in relation to cancerogenesis, or with the MMPs in particular.
4.2. MMP-9 Direct Interactors
Fourteen out of the 20 putative interactors were exclusive for MMP-9. They were the following:
Transgelin, TAGL (TAGLN)
is an actin-binding protein involved in calcium interactions and contractile properties of the cell. Using an expression cloning strategy, Nair et al. identified transgelin as a novel suppressor of MMP-9 expression [37
]. This finding attributes the unusual role of tumor-suppressor to the MMP-9, which deserves further attention.
Macrophage Migration Inhibitory Factor, MIF (MIF)
is an interesting small protein (approximately 12.5 kDa) involved in several biological activities, including the stimulation of the production of cytokines, chemokines, growth factors and angiogenic factors that may favor tumor growth and metastatic spreading. The overexpression of MIF in breast cancer cells, and its reported interaction with HSP90 and CXCR-4, is known to induce resistance to apoptosis and stimulation of proliferation via the AKT pathway. This opens new scenery regarding the possible correlations between matrix degradation and cell proliferation. Moreover, MIF is involved in the innate immune response and in regulating the function of macrophages in host defense [38
Heat Shock 70 kDa Protein 4, HSP74 (HSPA4)
and Heat Shock 27kDa Protein, (HSP27) (HSPB1)
are two significant members of the multigenic heat shock protein family [41
]. The HSPA4 (heat shock 70kDa protein 4) is a major component of the HSP chaperone family involved in the folding of nascent proteins and in the degradation of misfolded polypeptides [42
The other subset of chaperones consists of HSPs of a molecular weight of less than 30 kDa (sHSPs). Among some of their functions, sHPSs participate in cell survival, cytoskeletal motility, and disruption of protein aggregation [43
]. Moreover, it has been reported that the HSPs display elevated expression levels in cancer, where they may perform anti-apoptotic activities, both spontaneous and generated by therapy [44
]. In particular, the high expression of HSPB1 (HSP27) has been associated with poor prognosis in several carcinomas and osteosarcomas [45
]. Their putative correlation with MMP-9 suggests a synergy between the two key mechanisms of cancer progression: matrix degradation and anti-apoptotic effects.
Peptidylprolyl isomerase A (cyclophilin A), PPIA (PPIA)
accelerates the folding of proteins and catalyzes the cis-trans isomerization of proline imidic peptide bonds in oligopeptides. Recently, a key role of PPIA in tumor biology has been proposed [46
]. The postulated interaction with MMP-9 remains to be clarified.
Other presumed interactors are listed below.
Catalase, CATA (CAT)
occurs in almost all aerobic organisms and serves to protect cells from the toxic effects of hydrogen peroxide (H2
). Moreover, it can reduce the activity of MMPs and promote cellular growth of many cell types, including T-cells, B-cells, myeloid leukemia cells, transformed fibroblast cells and others [47
Enoyl-CoA hydratase, ECHM (ECHS1)
is a key enzyme involved in the oxidation of fatty acids and branched-chain amino acids. Moreover, it has been reported to be associated with the progression of a variety of tumors [49
], including gastric cancer [50
]. However, no correlation has been described so far with the matrix metalloproteases.
Thioredoxin, THIO (TXN)
is a multifunctional cellular factor with thiol-mediated redox activity. It plays pivotal roles in the regulation of many cellular processes, including proliferation, apoptosis, and gene expression, both in normal and tumoral cells [44
]. Its frequent high expression has been reported in many cancers [51
Transferrin, TRFE (TF)
is a transport protein which can bind two Fe (3+) ions in association with the binding of an anion, usually bicarbonate. It is responsible for the transport of iron from sites of absorption and heme degradation to those of storage and utilization. Serum transferrin may also have a further role in stimulating cell proliferation [54
High mobility group protein B1, HMGB1 (HMGB1)
is a multifunctional redox-sensitive protein with various roles in different cellular compartments. In the nucleus, it is one of the major non-histone/chromatin-associated proteins and acts as a DNA chaperone involved in replication, transcription, DNA repair and genome stability. It has been reported that it promotes host inflammatory reactions to external signals and immune responses. In the cytoplasm, it functions as a damage-associated molecule inducing inflammatory mediator release [55
, such as the group of S100-A7,-A8 and -A9 calcium binding proteins, are known to exert several functions in normal and tumoral tissues, where they may be expressed at high levels (as has been reported by our group [56
] and other authors [57
]). Their correlation with MMP-9 has never been proposed before.
Actin, (ACTB) ACTB
is the primary component of the cytoskeleton in eukaryotic cells. The actin cytoskeleton is therefore responsible for the integrity and shape-maintenance of cells. The cytoskeleton reorganization is a physiological event during cell growth, differentiation, and senescence of stationary cells, and it is responsible for the cell motility of migratory cells [61
Under pathological conditions, like cancer, the actin cytoskeleton may undergo deregulated fragmentation. It has been recently reported that induced alteration of actin cytoskeletal integrity in human trabecular meshwork cells (HTMC) is associated with MMP-2 activation, presumably through the upregulation of its activator, MT1-MMP [62
4.3. MMP-2 Direct Interactors
The functional interactors attributed exclusively to MMP-2 are the following:
Bifunctional purine biosynthesis protein PURH, PUR9 (ATIC)
is a bifunctional enzyme that catalyzes two steps in purine biosynthesis. Moreover, it has been reported that it may promote auto-phosphorylation and internalization of the insulin receptor (INSR) [63
]. No information is currently found on potential functional interactions with MMP-2 and other MMPs
Cathepsin D, CATD (CTSD)
is an acid protease that is active in the breakdown of intracellular proteins, which has been reported to be involved in the pathogenesis of several diseases, such as breast cancer and possibly Alzheimer’s disease via non-proteolytic pathways [64
]. The postulated direct correlation between CTSD and MMP-2 is an interesting observation which needs further insight.
Fascin, FSCN1 (FSCN1)
is one of the organizers of the actin filaments into bundles. Therefore, it plays a role in the formation of microspikes, membrane ruffles, and stress fibers, as well as other cell projections such as filopodia, which are essential for cell motility and migration. Its frequent overexpression in cancer has been related to the promotion of an actin-independent cell migration [66
]. Its correlation with MMP-2 could be instrumental in the increased aptitude of cells to migrate during the invasive phase of the breast cancer,
Heat shock protein HSP 90-alpha, HS90A (HSP90AA1)
is a molecular chaperone that has been recognized as one of those responsible for the structural maintenance of cells, and also for cell cycle control and signal transduction pathways [67
]. It has been reported that HSP90, when translocated into the nucleus, may influence the activity of many transcription factors [68
14-3-3 Protein epsilon, 1433E (YWHAE)
is a member of a multigenic protein family. In mammals, it consists of seven members (β, ε, η, γ, τ, ζ and σ), which act as adapter proteins involved in the regulation of a large number of signaling pathways [69
] and in the maintenance of epithelial cell polarity [71
]. It is interesting to note that this protein is frequently overexpressed in breast cancer [21
]. However, no correlation with collagenolytic activities has been reported before.
In conclusion, the present investigation (based on a double line of approach, experimental and “in silico”) highlights for the first time the complexity of the interactive networks that MMP-2 and MMP-9, through a series of interactors (partly common and partly exclusive) may accomplish within the cell. These complex interactive molecular circuits, where the two collagenolytic enzymes appear to be included, suggest their potential involvement in other important cellular activities, besides that of remodeling the extracellular matrix. These interactors are deduced by the direct inquiry between MMP-2 and MMP-9 with the proteomic platform of the breast tissues, and are not inclusive of the MMP inhibitors (TIMP) and activator (MMP14). The scenery becomes even more complex when analyzing the members of the second level of interactors, or beyond. In our opinion, this is the first comprehensive description of potential activities where MMP-2 and MMP-9 can be involved in the complicated scenario in which the mechanisms of tumor progression are correlated with unfavorable prognosis.