MMP-1 is one of the prototypical MMPs and is expressed in a wide variety of tissues in the human body, including the liver, kidney, intestine, stomach, placenta, bladder, and pancreas, in pathological conditions, it increases the bioavailability of insulin-like growth factor-1 (IGF-1), promotes cell proliferation and migration, epithelial regeneration, and inflammation, and has a proteolytic activity that degrades physical barriers and promotes cancer progression (25). Previous research suggested that MMP-1 stimulates protease-activated receptor (PAR) 1 by cleaving the same Arg-Ser link that thrombins cleave, which encourages breast cancer cell proliferation and invasion (26).

The primary structural protein of bone is collagen I, and the onset of bone resorption depends on the degradation of this protein. Type I collagen is known to be degraded by MMP-1 at neutral pH levels. Results from an earlier study showed that MM patients have intrinsically high levels of MMP1, and co-culturing them with RPMI8226 cells makes this phenomenon more obvious (27). However, the exact mechanism by which MMP-1 contributes to bone deterioration is still unknown. Another study (28) established that osteoblasts induced myeloma cells to release MMP-1, urokinase plasminogen activator (uPA), and hepatocyte growth factor (HGF); conversely, contact with myeloma cells caused osteoblasts to produce MMP-1 in high amounts and the stimulation of osteoblastic MMP-1 expression by myeloma cells was driven by p38. Additionally, the contact between myeloma cells and osteoblasts can significantly increase the ability of myeloma cells to invade and migrate. This indicates that the interaction between MMP-1 and p38/MAPK plays a crucial role in modulating the interplay between myeloma cells and BMSCs. The p38/MAPK pathway is a significant signaling cascade that participates in various cellular activities, such as growth, proliferation, differentiation, migration, and apoptosis (29, 30). In particular, the activation of the p38/MAPK pathway in myeloma cells increases their invasive and migratory abilities when they come into contact with BMSCs. In summary, the relationship between MMP-1 and p38/MAPK is critical for the investigation of the mechanisms of MM progression.

MMP-2 and MMP-9 also referred to as gelatinase-A and gelatinase-B, are extensively expressed in almost all human tissues. Moreover, their functional architecture includes a fibronectin section within the catalytic domain (31). This extra fibronectin domain can be bound to and processed by denatured collagen or gelatine. These enzymes, according to research, not only degrade various ECM molecules, such as Collagen types I, IV, V, VII, X, IX, aggrecan, fibronectin, laminin, vitronectin, and elastin, but also break down non-ECM molecules like pro-TNF-α, pro-IL-β, pro-IL-8, TGF-β, and monocyte chemoattractant protein-3 (32). Gelatinases are involved in the onset and progression of a variety of human diseases, the most widely studied of these is diabetes mellitus (33, 34). In addition, gelatinase has been reported to be one of the key ECM-degrading enzymes involved in tumor invasion and metastasis (35). Based on this, we will conduct a more comprehensive analysis and discussion on the exact roles of MMP-2 and MMP-9 in multiple myeloma, as well as their potential clinical implications.

There have been numerous studies on the expression of MMP-2 and MMP-9 in MM patients during the past two decades. Vacca et al. (36) showed that active MM patients express significantly higher levels of MMP-2 mRNA and protein compared to inactive MM and MGUS patients by in situ hybridizations of BM PCs and gelatin-zymography, while MMP-9 expression was similar in all groups. Furthermore, immunoassays of plasma cell extracts showed that levels of angiogenic basic fibroblast growth factor (FGF) - 2 were significantly higher in active MM patients than in inactive MM patients and MGUS patients. These findings suggest that the angiogenic and invasive potential of MM is partially dependent on FGF-2 and MMP-2 production. In addition, Marquez-Curtis et al. (37) established Long-term Marrow Cultures (LTMCs) system from acute myelogenous leukemia (AML) patients and normal donors BM cells. The LTMCs system contains multiple BM cells, including bone marrow stromal cells (BMSCs) and progenitor cells, and can maintain the growth and differentiation of various BM cells in vitro for a long period. Supernatants from this system were collected for enzyme spectrum analysis, and analysis of normal LTMC culture results showed that proMMP-9 concentration gradually decreased with the increase of culture time, reaching the lowest level by the 4 ± 5th week, while proMMP-2 concentration gradually increased with the increase of culture time. However, compared with normal LTMCs, AML LTMCs displayed higher levels of proMMP-9. Zdzisińska et al. (27) co-cultured RPMI8226 cells with BMSCs obtained from MM patients and healthy controls, and found that MM patient-derived BMSCs produced significantly higher levels of MMP-2 compared to BMSCs from healthy controls, while RPMI8226 cells alone did not produce detectable levels of MMP-2. Van et al. (38) established an MM mouse model to mimic the growth and progression of human multiple myeloma by transplanting 5T33MM cells into mice, and their study found that 5T33MM cells secreted MMP-9 in vivo. Specifically, upon injection of 5T33MMvt cells into immature mice, MMP-9 secretion was upregulated in MM cells isolated from the BM (5T33MMvt-vv) during tumor development. However, when these cells were re-cultivated in vitro, MMP-9 production declined and was eventually eliminated, suggesting that the production of MMP-9 was controlled by the BM microenvironment. Notably, in vitro, interaction of 5T33MM cells with BM endothelial cells resulted in upregulation of MMP-9 in the 5T33MM cells. These studies have shown that although MMP-2 and MMP-9 share similar structures and general functions, their expression in MM is vastly different. In MM, MMP-2 is mainly produced by BMSCs, whereas MMP-9 is primarily produced by malignant PCs, and its production is regulated by the BM microenvironment (27, 38).

In its physiological state, MMP-7 is mainly expressed in the human reproductive system, salivary glands, and prostate (57). Previous research has demonstrated that MMP-7 promotes in vivo osteolysis and is differently expressed in the tumor-bone microenvironment in breast and prostate cancer (58, 59). In MM, Barillé et al. (60) discovered that MMP-7 was able to stimulate BMSCs to produce MMP-2, which can degrade collagen I to promote bone absorption. Later, research by Thioloy et al. (58) demonstrated that MMP-7 produced from osteoclasts greatly aided tumor growth and tumor-induced osteolysis. All of their findings point to the possibility that blocking MMP activity might be a potential therapeutic approach for MM.

Nevertheless, another study soon after reached a completely different conclusion. In the study by Lwin et al. (61), a previously unknown role for MMP-7 in the pathogenesis of myeloma was discovered. The study found that MMP-7-deficient myeloma-bearing mice had a significantly higher tumor load and osteolytic bone disease compared to their wild-type counterparts, and clinical evidence supported the in vivo murine myeloma studies as well, which demonstrated a marked decrease in MMP-7 activity observed in MM patients. According to previous findings, MMP-7 is abundantly expressed by osteoclasts and able to cleave receptor activator for nuclear factor-κB (RANKL) expressed by osteoblasts can activate osteoclasts to a soluble active form (59). However, the data from Lwin et al.’s investigation indicate that there was no discernible change in RANKL levels between myeloma-bearing mice lacking MMP-7 and the wild-type control. Interestingly, they found an increase in myeloma cell viability by co-culturing with 2T3 pre-osteoblasts, whereas overexpression of MMP-7 in pre-osteoblasts suppressed this phenomenon (61). It follows that the promotion of osteolytic bone disease by loss of host-derived MMP-7 is not determined by a single mechanism.

The above study implies that rather than exerting a direct antitumor effect, MMP-7 achieves either a tumor-promoting or an inhibitory effect through a variety of complex mechanisms in the specific tumor microenvironment. Furthermore, interactions between MMP-7 and other signaling molecules or proteins might differ across studies, contributing to contrasting conclusions.

MMP-13 is predominantly expressed in the lung, skin, prostate, small intestine, breast, testis, and bladder in normal humans, but more importantly, it is widely expressed in the mesenchymal stromal cell (MSC) lineage, including chondrocytes, osteoblasts, and osteocytes; however, osteoclasts do not appear to express MMP-13 (57, 62). To date, MMP-13 participates in the progress of MM mainly through its proteolytic activity and catalytic activity (62, 63). Many of the important functional roles of MMP-13 have been directly connected to its capacity to breakdown interstitial collagen which is a critical structural component of all connective tissues, including bone, and other matrix-associated targets (64, 65). However, Fu et al. (63) demonstrate that MMP-13 expression is induced by IL-6-mediated interactions between MM and BMSC and acts as a potent osteoclastogenic factor that can encourage the production of multinucleated, bone-resorbing osteoclasts. Different from the previous mechanism, MMP-13 exerts its pro-osteoclastogenic effects without a requirement for proteolytic activity by acting as a secretagogue inducing DC-STAMP, which is an important fusogenic factor whose absence impairs osteoclasts formation and function (63). Myeloma-bone interaction is a vicious loop that enhances the bioavailability of cytokines and growth hormones, which promote tumor growth and increase therapeutic resistance. Surprisingly, soluble factor analysis from MMP-13-null mice revealed decreased bioavailability of various osteoclastogenic factors of MSCs including CXCL7 which is capable to promote the recruitment and formation of osteoclast precursors. Therefore, CXCL7 can be regarded as a new MMP-13 substrate and osteoclast production regulator. Moreover, this effect is determined by MMP-13 catalytic activity rather than proteolytic activity since the selective inhibitor of MMP-13 could obviously prolong overall survival in MM-bearing mice (62, 66).

MT-MMPs are expressed almost in various tissues of the human urinary, reproductive, circulatory, and digestive systems in physiological status (57). MT-MMPs are a special subtype that can undergo pericellular proteolysis, which is considered an essential phase in the restructuring of paracancerous tissues (67). Among MT-MMPs, it has been reported that MT-MMP1 and MT-MMP2, also known as MMP14 and MMP-15, are mainly involved in MM progression (68–70). According to some research, MT1-MMP inhibits macrophage invasion by hydrolyzing ECM elements and cell surface chemicals and by triggering signal transduction pathways that affect motility and energy consumption (71). Several ECM proteins, including gelatin, fibronectin, laminin, and fibrillary collagens, are degraded by MT1-MMP and MT2-MMP, which can activate the pro-MMP2 on the cell surface as well (67, 72, 73).

MT1-MMP(MMP-14), the first membrane-type MMP discovered, was identified in 1994 as an activator of pro-MMP-2 (74). In MM, MMP-T1 has been confirmed to be expressed by malignant PCs and be conducive to CXCL12 promoting the invasion of myeloma cells via matrigel-reconstituted basal membranes and type I collagen gels (68). Several years later, Shimizu-Hirota et al. (71) identified in addition to acting as an ECM-degrading enzyme, MT-MMP1 also unexpectedly regulates inflammatory responses. They found that MT1-MMP^–/– macrophages produce excessive chemokine and cytokine responses to in vitro and in vivo immune stimuli. These researches prove that the role of MT-MMP1 in CXCL-12 promoting MM cells to invade the basement membrane may be related to macrophage immune response.

MT2-MMP (MMP-15), which shares 73.9% of its overall similarity with MT1-MMP, was initially identified from the human lung cDNA library (67, 75). It has been demonstrated that the human fibrosarcoma cell HT1080 expresses MT2-MMP. Moreover, it was shown that the migration and invasion of cancer cells were decreased following siRNA induction, which inhibited the endogenous expression of MT2-MMP (67). In comparison to B cell lines and other normal peripheral blood or BM-derived cells, the expression of MT2-MMP is markedly higher in MM cells, according to a recent study (70). When MT2-MMP was inhibited in U266 cells utilizing siRNA technology, it turned out to be the case that MT2-MMP siRNA considerably reduced its adhesion, invasion, migration, and proliferation capabilities (70). Taken together, the studies discussed above suggest that MT2-MMP may serve as a viable biomarker for both diagnosing and treating MM.

TIMP1 was first discovered and defined in 1972, which constitutively expresses in many mammalian tissues (16, 76). Nowadays TIMP1 has been found in a lot of tissues, including the heart, brain, colon, arteries, liver, kidneys, lungs, bladder, breasts, skin, lymph nodes, ovaries, uterus, prostate, and testes (57). Proinflammatory cytokines such as tumor necrosis factor-alpha (TNF-α) and interleukin-1beta(IL-1β) can increase the production of TIMP1 mRNA (77). TIMP1 has a certain correlation with cell migration. For example, TIMP1 regulates the MMP-mediated degradation of vascular endothelial cadherin (VE-cadherin) at intercellular junctions and thereby inhibits microvascular endothelial cell migration (78). In addition, TIMP1 seemed to mediate bronchiole epithelial cell migration after injury by inhibiting the MMP7-dependent cleavage of syndecan-1, a heparan sulfate glycoprotein, whose shedding is associated with increased cell migration (79, 80). TIMP-1 inhibits many types of MMPs, including MMP-1, MMP-2, MMP-3, MMP-7, MMP-8, and MMP-9, it is worth mentioning that it exhibits a high binding affinity to both pro-MMP-9 and active MMP-9 and has been seen as the most effective inhibitor of MMP-7 in four TIMP families (57, 81). Of note, significantly elevated TIMPI levels were consistently found in both mice and patients with MM bone disease, while MMP-7 levels decreased (61). In addition, it has been observed that human MSCs are capable of secreting TIMP-1, which aids in reducing the levels of MMP-9 produced by MM cells and consequently prevents their migration (82). Furthermore, interactions between TIMP-1 and the hemopexin domain of pro-MMP-9 lead to the formation of a compound that hinders MMP-9 activation, which in turn inhibits the production of MMP-9 (83). These phenomena also indicate that TIMP1 has the potential to be a novel therapeutic target for MM and osteolytic bone disease.

The cloning and sequencing of TIMP-2 were initially reported in 1990, wherein the cDNA library of A2058 human melanoma cells was utilized for this purpose (84). More importantly, in 1991, it was demonstrated that TIMP-2 functions as a potent inhibitor of collagenase activity by forming a stable complex with a 1:1 molar ratio between TIMP-2 and the enzyme (85). The lymph nodes, brain, heart, arteries, colon, kidneys, liver, breasts, ovaries, prostate, and testes are among the tissues that express TIMP-2 (57). Previous studies (57, 86) have shown that TIMP-2 also regulates MT1-MMP, MMP-2, and MMP-9 activity. Furthermore, a recent study has confirmed a correlation between lower expression of TIMP-2 and/or higher expression of MMP-9 in colon cancer tissues and poorer overall survival. Additionally, both in vitro and in vivo experiments have shown that TIMP-2 efficiently prevents the invasion and migration of HCT 116 cells through the regulation of MMP-9 (87). In non-small cell lung cancer(NSCLC), TIMP-2, as a protective factor, is up-regulated and related to the prognosis (88). A recent study has confirmed that miR-483-5p exerts a tumorigenic effect in MM by targeting TIMP-2, leading to its attenuation in MM cells, which in turn promotes cell proliferation and inhibits apoptosis, thereby facilitating MM progression (89).

TIMP-3 was cloned in chickens and humans in 1992 and 1994, respectively (90–92). TIMP-3 expression has been detected in various tissues, including the brain, heart, colon, kidneys, lungs, liver, breasts, ovaries, prostate, and testes (57). Mechanistically, TIMP-3 has been shown to function as a potent activator of apoptotic pathways by inhibiting the release of death receptors like Fas from the cell surface, which is mediated by its N-terminal domain, thereby promoting the cleavage and activation of downstream apoptotic signaling molecules (93). Given that TIMP-3 is not secreted like the other three TIMP families but rather exists in the matrix, it is not unexpected that TIMP3 can affect ECM-cell signaling mediated focal adhesion kinase (FAK) and fibronectin (FN) (94). TIMP-3 is associated with a variety of cancers, including kidney cancer and brain cancer, and the methylation-related silencing of the TIMP-3 gene indicates its inhibitory effect on cancer progression (95). Besides that, its N-terminal domain can interact with the active sites of MMP-2 and MMP-9 to exert its MMP inhibitory effect (57). It is well known that Interleukin-6 (IL-6) is the main growth factor of human myeloma cells, which acts through its receptor IL-6R. In vivo, IL-6R is expressed on the surface of myeloma cells and released into the serum to become a soluble form (sIL-6R). Clinical research has suggested that the level of serum IL-6R correlates with the poor prognosis of patients. Importantly, TIMP-3 and BB-94, a metalloproteinase inhibitor based on hydroxamic acid, can prevent the release of sIL-6R. In other words, TIMP-3 mediated inhibition of sIL-6R release may be one of the methods for treatment in MM (96).

TIMP-4 was cloned from a human heart cDNA library by using the expressed sequence tag sequencing approach in 1996 (97). It has been discovered in several organs and tissues, including the brain, heart, kidneys, breasts, uterus, pancreas, colon, ovaries, testes, prostate, and adipose tissue (57). TIMP-4 is the main MMP inhibitor in human platelets, and it affects platelet recruitment and aggregation (98). In addition, TIMP-4 has also been known to regulate the activity of MMP-2 by acting as the negative regulator of MT1-MMP (99). As mentioned above, MT1-MMP and MMP-2 are closely related to MM and osteolytic osteopathy.