Matrix metalloproteinase-10 (MMP-10) is a zinc-dependent endopeptidase with the ability to degrade a broad spectrum of extracellular matrices and other protein substrates. The expression of MMP-10 is induced in acute kidney injury (AKI) and chronic kidney disease (CKD), as well as in renal cell carcinoma (RCC). During the different stages of kidney injury, MMP-10 may exert distinct functions by cleaving various bioactive substrates including heparin-binding epidermal growth factor (HB-EGF), zonula occludens-1 (ZO-1), and pro-MMP-1, -7, -8, -9, -10, -13. Functionally, MMP-10 is reno-protective in AKI by promoting HB-EGF-mediated tubular repair and regeneration, whereas it aggravates podocyte dysfunction and proteinuria by disrupting glomerular filtration integrity via degrading ZO-1.
1. Biology of Matrix Metalloproteinase-10 (MMP-10)
Matrix metalloproteinase-10 (MMP-10) belongs to a large family of MMPs consisting of 25 known members
[1]. MMP-10 shares common structural motifs with other MMPs. MMP-10 activities are modulated on several levels including transcription, pro-enzyme activation, or inhibition by its inhibitors.
1.1. Structure of MMP-10
As a secreted protein, MMP-10 is primarily synthesized as the 476 amino acid zymogen comprised of four domains: 17 amino acid N-terminal signal peptide, 81 amino acid pro-peptide domain, 165 amino acid catalytic domain, and 187 amino acid hemopexin domain
[2][3][4]. Interestingly, recent studies reveal that a nuclear localization sequence (NLS) might exist in the C-terminal, which contributes MMP-10 to be localized and targeted to the nucleus
[5][6][7].
The pro-peptide contains a cysteine switch motif
[8], an intramolecular complex formed between the sulfhydryl group of the cysteine residue in the pro-peptide domain and the essential zinc ion located in the catalytic domain. The cysteine switch motif blocks the active site and maintains the precursor structure
[2]. Consequently, disruption of this motif can release active enzyme
[9].
Two zinc ions (one catalytic and one structural) and three calcium ions are indispensable for the catalytic site. Three histidines from the zinc binding motif in the catalytic domain coordinate with catalytic zinc and constitute the catalytic site
[10]. Importantly, the substrate specificity is predominantly determined by the S1′ hydrophobic pocket which is adjacent to the catalytic site
[11].
The hemopexin domain with four hemopexin-like repeats presents four-bladed propeller folds which links with the catalytic domain by a flexible hinge region. In addition to regulating protein–protein interaction, the hemopexin domain could play an auxiliary role in helping substrate binding at catalytic sites
[12].
1.2. Transcriptional Regulation of MMP-10
In humans, the
MMP-10 gene is mapped to chromosomes at 11q22.2. A set of various
trans-activators can modulate the expression of
MMP-10 by regulating certain
cis-elements, including TATA boxes, activated protein-1 (AP-1) binding site, polyoma enhancer activator 3 (PEA3)-binding site, β-catenin/T-cell factor 4 (TCF4), and nuclear factor-ĸB (NF-ĸB)
[13][14][15]. The interplay of a variety of
cis-acting elements with cognate transcription factors confers the response to specific cues in various tissues and cell types. Moreover, DNA hypomethylation can also modulate transcriptional activity on the
MMP-10 promoter in a cell-specific manner
[16].
NF-κB and AP-1 are widely activated in all kinds of kidney disorders and participate in the progression of nephropathies
[17]. Transforming growth factor-β (TGF-β), a powerful profibrotic cytokine, may transactivate
MMP-10 by a specific interaction between the Smad proteins and AP-1
[18]. Furthermore, another study shows that
MMP-10 expression is regulated by TGF-β through mediating activation of myocyte enhancer factor 2A (MEF2A) and downregulating class IIα histone deacetylases (HDACs)
[19]. In addition, biomechanical stretching also induces
MMP-10 expression by mediating NF-ĸB activation in podocytes in vitro
[20].
Bioinformatics analysis shows that there are two putative TCF/LEF-binding sites (TBS) in the promoter region of
MMP-10 [21]. In various kinds of nephropathies, canonical Wnt/β-catenin signaling is activated and plays a pivotal role in mediating kidney injury, repair and pathogenesis
[22][23][24]. Recent studies indicate that Wnt7a induces MMP-10 expression through the canonical β-catenin pathway
[21]. Furthermore, overexpression of Wnt5b in vitro upregulates MMP-10 expression as well
[25]. As Wnt/β-catenin is activated in all kinds of AKI and CKD examined, these findings suggest a widespread induction of MMP-10 in various kidney disorders.
1.3. Posttranscriptional Activation of MMP-10
MMP-10 is commonly regarded as an extracellular proteinase. However, recent finding shows that MMP-10 is also activated inside the cell and involved in the pathogenesis of diseases
[6]. The mechanism by which MMP-10 enters cells has not been fully elucidated. During the activation, pro-MMP-10 is enzymatically cleaved, and its conformation is modified. Pro-MMP-10 is secreted into the extracellular space after the N-terminal signal peptide is removed and can be subsequently activated by various factors
[2]. In extracellular milieu, the His81-Phe82 bond in the pro-peptide is often cleaved by other proteolytic enzymes, such as plasma kallikrein, trypsin, neutrophil elastase, or cathepsin G
[26][3]. As such, this procedure contributes to the conversion of 56 kDa latent MMP-10 to mature active forms, which are 47 kDa and 24 kDa, respectively
[8]. Along with this, the cysteine residue is dissociated from catalytic zinc that leads to cysteine-switch activation, and the catalytic site is exposed
[9].
It has been shown that pro-MMP-10 is labile and can undergo autolysis, leading to the production of mature MMP-10
[8]. Under nondenaturing conditions, MMP-10 can also be activated through conformational change without cleavage of the pro-peptide after exposure to SDS
[2].
1.4. Inhibitors of MMP-10
Inhibitors of MMP-10 are composed of two classes: endogenous and synthetic. Tissue inhibitors of metalloproteinases (TIMPs), consisting of TIMP1~4, are secreted proteins and function against multiple secreted or membrane-anchored metalloproteinases
[27]. In addition, TIMPs play a pivotal role in modulating the influence of the extracellular environment on cell phenotype
[28]. It has been demonstrated that each of the TIMPs can inhibit MMP-10 activity by forming a complex in a 1:1 stoichiometry
[29][30][31]. In spite of substantial structural differences between the TIMPs, most of the flexible loops of the MMP-10 catalytic domain are locked into nearly indistinguishable conformations upon TIMP binding
[30]. Disrupting the balance between MMP-10 and TIMPs will contribute to a broad spectrum of devastating diseases.
Although TIMPs can be considered as MMP-10 endogenous inhibitors, their actions are not merely on MMP-10
[27]. The development of a specific MMP-10 inhibitor with a pure inhibitory spectrum is urgent. Structurally, MMP-10 clearly resembles other MMPs, especially MMP-3, as they share 78% overall amino acid sequence identity and 86% sequence identity in their catalytic domain in humans. This makes it difficult to develop a selective and potent MMP-10 inhibitor
[3][10]. Nevertheless, there are several studies reporting that selective dual MMP-10/-13 inhibitors possess a clean inhibition profile and outstanding affinity for MMP-10
[32][33]. Furthermore, MMP-10 activity can be regulated by increased or diminished expression at the transcriptional or translational levels. Synthetic inhibitors generally contain a chelating group that binds the catalytic zinc atom at the MMP active site. Other inhibitors interact with various binding pockets on the MMP of interest resulting in specific inhibitory potentials for the given MMP. In addition, single-domain antibodies devoid of a light chain and CH1 region emerge as attractive inhibitors
[34]. A recent study reported a single-domain antibody, H3, which can selectively inhibit MMP-10 activity by binding to its active site
[35]. Further studies on developing specific inhibitors of MMP-10 are needed to provide a novel therapeutic strategy for various diseases with hyperactive MMP-10.
2. MMP-10 Expression in the Kidney
Contrary to its weak staining in normal human kidney tubular cells
[36], MMP-10 is scarcely expressed in normal murine kidneys. It is substantially induced under pathological conditions
[37][38]. Consistently, MMP-10-/- mice and transgenic mice with podocyte-specific expression of MMP-10 do not display overt renal abnormality
[37][39]. These findings indicate that MMP-10 is not required for murine kidney structure and function under normal physiological conditions. However, overexpression of MMP-10 renders transgenic mice susceptible to developing more severe podocytopathy after injury
[37].
In response to diverse insults, MMP-10 can be induced in various cells, including the renal tubular epithelium, glomerular podocytes, juxtaglomerular apparatus, as well as renal cell carcinoma
[37][38][36][40].
2.1. AKI
Induction of MMP-10 is a common pathological finding in various models of AKI. Studies show that MMP-10 protein is substantially upregulated in various animal models of AKI, including ischemia-reperfusion injury (IRI), cisplatin-induced nephrotoxic AKI and glycerol-induced rhabdomyolysis-associated AKI
[38][41]. Notably, the mRNA level of
MMP-10 is induced as early as 4 h and sustained at least to 48 h after IRI
[38]. Data mining from the Kidney Interactive Transcriptomics database created by single-nuclei RNA-sequencing reveals the dynamics and landscape of
MMP-10 expression in mouse AKI induced by IRI (
http://humphreyslab.com/SingleCell, accessed on 20 December 2021). Furthermore, proximal tubules, especially the S3 segment, are the main cell population that express
MMP-10 following IRI
[42]. Immunohistochemical staining reveals that MMP-10 is strongly stained in the injured area of the kidney and predominantly localizes in the renal tubular epithelium
[38][41]. The mechanism underlying MMP-10 induction in AKI, as well as the triggers for induction, remains unclear. Notably, the canonical Wnt/β-catenin signaling, which protects the kidney from AKI
[43], is activated in this setting. This may account for the regulatory mechanism of MMP-10 expression in response to AKI.
2.2. CKD
MMP-10 can also be detected in the glomerular podocytes of CKD patients with diabetic kidney disease (DKD), IgA nephropathy (IgAN), and focal segmental glomerulosclerosis (FSGS)
[37]. Consistent with this finding, MMP-10 protein is not only induced significantly in adriamycin (ADR) mice in a time-dependent fashion, but is also upregulated in other proteinuric CKD models, such as diabetic
db/db mice, remnant kidney model after 5/6 nephrectomy (5/6NX), Alport mice, CD151 null mice and α-actinin-4 knockout mice, which are a Alport syndrome model and a model of FSGS, respectively
[37][20][40][44]. Furthermore, compared with the controls, zymographic analysis also reveals an increase in MMP-10 proteolytic activity in the kidney after ADR
[37]. In experimental models, MMP-10 is not only localized in glomerular podocytes but also in the juxtaglomerular apparatus
[37][40][45].
2.3. RCC
Upregulation of
MMP-10 is also common in numerous cancers
[46]. In patients with renal cell carcinoma (RCC), immunohistochemical staining reveals that induction of MMP-10 is primarily detected in tubular cancer cell cytoplasm
[36]. Meanwhile, MMP-10 expression exhibits a tight correlation with the grade and pT stage. Cancer cells of the invasive front strongly express MMP-10, and the majority of sarcomatous cancer cells shows moderate or strong intensity
[36]. As such, sarcomatous change represents a transformation to a higher-grade malignancy and portends a worse prognosis in renal cell carcinoma
[47]. Consequently, these findings suggest that MMP-10 is relevant to RCC invasion.
MMP-10 appears to be the target of diverse factors involved in RCC invasion. In favor of this notion, ghrelin, a peptide hormone, binds to the ghrelin receptor and results in MMP-10 expression by upregulating Aurora A, which results in RCC invasion
[48]. In vitro, MMP-10 is upregulated by IL-1β, mechanistically due to activation of transcription factor CCAAT enhancer binding protein β (CEBPβ)
[49]. In addition, there is a study reporting that activation of
MMP-10 might be involved in the secreted frizzled-related protein 1 (SFRP1)-caused cell invasion in metastatic RCC
[50].
3. Mechanism of MMP-10 Action in Kidney Disease
3.1. HB-EGF
HB-EGF is a membrane-anchored growth factor, which is widely expressed in glomerular and tubular compartments of adult kidneys at a very low level. In response to AKI, HB-EGF is markedly induced primarily in tubular epithelial cells
[51]. As a unique epidermal growth factor receptor (EGFR) ligand, HB-EGF is activated post-translationally after cleavage of its ectodomain by diverse proteinases
[52]. Activated HB-EGF then binds to EGFR and promotes renal epithelial cell repair, proliferation, and regeneration after injury
[51][52]. Recent studies identify MMP-10 as the key enzyme responsible for mediating the cleavage of HB-EGF in human proximal tubular cells. As such, MMP-10 releases an active HB-EGF fragment capable of binding to EGFR, which prevents human proximal tubular epithelial cells from hypoxia-reoxygenation (H/R)-induced injury in vitro
[38]. The beneficial role of MMP-10 in AKI is in harmony with many earlier reports in which activation of EGFR signaling is vital for renal tubular injury repair and regeneration
[53][54][55]. In agreement with this notion, the protective role of MMP-10 in AKI can be eradicated by erlotinib, a specific EGFR tyrosine kinase inhibitor
[38][56].
Like EGFR, the role of the MMP-10/HB-EGF/EGFR axis in renal pathology is somehow contradictory and context-dependent, as both beneficial and deleterious effects have been reported in the kidney
[57][58][59]. In addition to the protective role illustrated above, mounting evidence suggests that persistent tubule-specific EGFR activation contributes to renal fibrosis in response to chronic injury
[60][61][62]. Furthermore, transgenic mice that selectively express human HB-EGF in the proximal tubule develop spontaneous, early onset, progressive tubulointerstitial fibrosis
[63]. Mechanistically, activation of EGFR stimulates signal transduction and activator of transcription (STAT3), which interacts with the promoter region of homeodomain-interacting protein kinase 2 (
HIPK2) gene and drives the progression of AKI to CKD
[64]. These findings raise an intriguing possibility that MMP-10 facilitates tubulointerstitial fibrosis by cleaving HB-EGF, leading to activation of EGFR in the renal proximal tubule.
3.2. ZO-1
ZO-1 is a tight junctional protein that constitutes the framework of podocyte junctions and plays an indispensable role in nephrogenesis and maintaining renal function
[65][66][67]. The expression of ZO-1 in podocytes and its interaction with slit diaphragm components, such as nephrin, NEPH1 and NEPH3, are critical for maintaining the integrity of the filtration barrier of glomeruli
[66][68]. A recent study indicates that MMP-10 specifically cleaves ZO-1, thereby impairing the integrity of slit diaphragm and resulting in podocyte dysfunction and proteinuria
[37]. This finding is in line with an earlier report demonstrating that specific loss of ZO-1 in podocytes leads to proteinuria and renal dysfunction with impaired glomerular filtration
[69].
As a key tight junctional protein, ZO-1 has a function in modulating paracellular permeability by interacting with ZO-1-associated nucleic acid binding protein (ZONAB), a Y-box transcription factor, normally bound to the SH3 domain of ZO-1 at intercellular tight junctions
[70][71][72]. In this regard, ZO-1 might also function as an inhibitor of ZONAB
[73]. The interaction between ZO-1 and ZONAB is essential for maintaining barrier function of epithelia and regulating epithelial cell proliferation and differentiation
[74][75]. Destruction of ZO-1/ZONAB complex leads to nuclear translocation of ZONAB from tight junctions, resulting in target gene expression
[76].
Although prior work has focused primarily on the role of ZO-1 in slit diaphragm, it has been demonstrated that ZO-1 is also abundantly expressed in renal tubule epithelial cells
[66][77][78]. The ZO-1/ZONAB signal pathway is indispensable for modulating epithelial phenotype. Certainly, following separation of ZONAB from ZO-1, nuclear ZONAB is positioned to shift the switch from differentiation to proliferation in proximal tubule epithelial cells in a cell density-dependent manner
[79]. Furthermore, ZONAB also promotes the stabilization and enhanced translation of p21 mRNA involved in cellular senescence
[80][81]. In nephropathic cystinosis, overproduction of reactive oxygen species (ROS) contributes to the destruction of tight junction integrity and ZONAB nuclear translocation, which is disastrous for renal tubular epithelial function
[73][82]. Based on these findings, it is tempting to speculate that the demolition of the tight junction and activation of a signaling cascade involving ZONAB in tubular cells, due to cleavage of ZO-1 by MMP-10, might be another mechanism in the evolution of CKD.
3.3. Other MMPs
As a proteolytic enzyme, MMP-10 is capable of cleaving pro-MMP-7 in a dose- and time-dependent manner, resulting in the release of mature MMP-7
[8]. Moreover, several direct lines of evidence suggest a role for MMP-10 in activating pro-MMP-9 in a dose-dependent manner, which rapidly produces four stable small fragments
[83][8]. Of note, the proteolytic capacity of MMP-10 to pro-MMP-9 is more powerful than to pro-MMP-7. Recombinant pro-MMP-1, pro-MMP-8, as well as pro-MMP-13 could also be processed to the mature form by incubating with MMP-10
[26][84]. Of particularly interest, in contrast to a previous report that MMP-10 fails to directly digest pro-MMP-2
[8], a more recent study demonstrates that MMP-10 can cleave the collagen α2 (I) chain through activating an endogenous collagenolytic MMP with the molecular weight similar to activated MMP-2
[85]. Nevertheless, whether this collagenase is MMP-2 needs further investigation.
The expression of MMP-1, MMP-2, MMP-7, MMP-8, MMP-9, MMP-10, as well as MMP-13 are upregulated after kidney damage
[1][86]. Up to now, studies show all of them implicated in the pathophysiology of kidney disorders
[87][88][89]. MMP-8 and MMP-13 have function in reducing kidney fibrosis by promoting matrix degradation after damage
[90][91]. Contrary to this conclusion, there is a study showing that a selective MMP-13 inhibitor is able to alleviate renal fibrosis
[92]. Taken together, these studies indicate that the ability of MMP-10 to super-activate other MMPs, including pro-MMP-1, pro-MMP-7, pro-MMP-8, pro-MMP-9, pro-MMP-10 and pro-MMP-13, might be another mechanism of MMP-10 involved in renal pathology.
This entry is adapted from the peer-reviewed paper 10.3390/ijms23042131