[Alzheimer's disease](/diseases/alzheimers-disease) (AD) represents the most prevalent neurodegenerative disorder globally, affecting over 55 million individuals worldwide and imposing substantial socioeconomic burdens on healthcare systems [@alzheimers2023]. The disease is characterized by progressive cognitive decline, memory impairment, and accumulation of two hallmark pathological lesions in the brain: extracellular amyloid plaques composed of [amyloid-beta](/proteins/amyloid-beta) (Aβ) peptides and intracellular neurofibrillary tangles formed by [hyperphosphorylated tau](/proteins/tau-protein) protein [@selkoe2016]. Among these pathological features, the amyloid hypothesis has long dominated AD research, proposing that the accumulation and aggregation of Aβ peptides serve as the primary trigger for downstream neurotoxicity, synaptic dysfunction, and neuronal loss [@karran2011].
[Alzheimer's disease](/diseases/alzheimers-disease) (AD) represents the most prevalent neurodegenerative disorder globally, affecting over 55 million individuals worldwide and imposing substantial socioeconomic burdens on healthcare systems [@alzheimers2023]. The disease is characterized by progressive cognitive decline, memory impairment, and accumulation of two hallmark pathological lesions in the brain: extracellular amyloid plaques composed of [amyloid-beta](/proteins/amyloid-beta) (Aβ) peptides and intracellular neurofibrillary tangles formed by [hyperphosphorylated tau](/proteins/tau-protein) protein [@selkoe2016]. Among these pathological features, the amyloid hypothesis has long dominated AD research, proposing that the accumulation and aggregation of Aβ peptides serve as the primary trigger for downstream neurotoxicity, synaptic dysfunction, and neuronal loss [@karran2011].
Amyloid-beta peptides are derived from the sequential proteolytic cleavage of the amyloid precursor protein (APP) by β-secretase (BACE1) and γ-secretase, yielding Aβ fragments ranging from 38 to 43 amino acids in length [@haass2007]. The Aβ₁₋₄₀ and Aβ₁₋₄₂ species represent the most abundant isoforms in the brain, with Aβ₁₋₄₂ exhibiting greater aggregation propensity and forming the core component of amyloid plaques [@jarrett1993]. The balance between Aβ production and clearance determines its accumulation in the brain, and impaired clearance mechanisms contribute significantly to Aβ deposition in sporadic AD cases [@mawuenyega2010].
The clearance of Aβ from the brain occurs through multiple pathways, including enzymatic degradation by various proteases, receptor-mediated uptake, and transport across the [blood-brain barrier](/mechanisms/blood-brain-barrier-dysfunction) [@saido2012]. Among these clearance mechanisms, matrix metalloproteinases (MMPs) have emerged as critical enzymes capable of degrading Aβ peptides, representing an endogenous protective system that could be harnessed for therapeutic intervention [@vitek1996]. MMPs are a family of zinc-dependent endopeptidases that play fundamental roles in extracellular matrix remodeling, cell signaling, and tissue homeostasis [@nagase2006]. Recent research has revealed that several MMPs possess the capacity to cleave and degrade Aβ, thereby potentially reducing amyloid burden and mitigating neurotoxicity [@miners2008]
The study of amyloid-beta metalloproteinases in AD represents a rapidly evolving field that bridges enzymology, neuroscience, and clinical research. Understanding the mechanisms by which MMPs degrade Aβ, the factors regulating their expression and activity, and their potential as therapeutic targets holds significant promise for developing disease-modifying treatments for AD [@rosenblum1998]. This comprehensive review examines the current knowledge regarding the role of matrix metalloproteinases in amyloid-beta degradation and their implications for Alzheimer's disease pathogenesis and therapy.
Matrix metalloproteinases constitute a large family of zinc-dependent endopeptidases currently comprising at least 28 members in humans, each characterized by a conserved catalytic domain containing the zinc-binding motif HExGHxxGxxH [@overall2002]. MMPs are traditionally classified based on their substrate specificity and structure into collagenases (MMP-1, MMP-8, MMP-13, MMP-18), gelatinases (MMP-2, MMP-9), stromelysins (MMP-3, MMP-10, MMP-11), matrilysins (MMP-7, MMP-26), membrane-type MMPs (MT-MMPs), and other specialized forms [@hadlerolsen2013]. Under physiological conditions, MMPs participate in extracellular matrix turnover, wound healing, angiogenesis, and various signaling processes [@sternlicht2001].
The involvement of MMPs in Alzheimer's disease has been increasingly recognized over the past two decades, with accumulating evidence suggesting both beneficial and detrimental roles depending on the specific MMP, cellular context, and disease stage [@yong2005]. MMPs are expressed in various cell types within the central nervous system, including [neurons](/cell-types/neurons), [astrocytes](/cell-types/astrocytes), [microglia](/cell-types/microglia), and endothelial cells, and their expression can be regulated by inflammatory cytokines, oxidative stress, and Aβ itself [@cunningham2005]. The blood-brain barrier also expresses several MMPs, which may influence Aβ transport and clearance [@rosenberg1998].
Initial studies revealed elevated MMP levels in AD brain tissue and cerebrospinal fluid, suggesting their involvement in disease pathogenesis [@backstrom1996]. However, the interpretation of these findings proved complex, as MMPs can exert both protective effects through Aβ degradation and harmful effects through promoting [neuroinflammation](/mechanisms/neuroinflammation), disrupting blood-brain barrier integrity, and contributing to synaptic dysfunction [@van2002]. The dual nature of MMPs in AD has made it challenging to define their precise role in disease progression.
MMP-9, also known as gelatinase B, has been particularly associated with AD pathophysiology. Elevated MMP-9 expression has been detected in AD brain tissue, particularly in association with amyloid plaques and around blood vessels [@asahina1998]. Studies have demonstrated that MMP-9 can degrade Aβ peptides in vitro and in vivo, suggesting a potential protective role [@van1992]. However, MMP-9 also contributes to neuroinflammation and can exacerbate neuronal damage under certain conditions [@gu2000]. The balance between these opposing effects may determine the net impact of MMP-9 on disease progression.
MMP-2 (gelatinase A) similarly demonstrates complex involvement in AD, with both reported protective and detrimental effects [@lorenzl2003]. MMP-2 is constitutively expressed in the brain and can degrade various substrates including Aβ, but its activity may be reduced in AD, contributing to impaired Aβ clearance [@byman2021]. The regulation of MMP expression and activity in AD involves multiple signaling pathways, including those activated by inflammatory mediators and Aβ itself, creating a complex network of interactions that influence disease processes [@li2020].
The enzymatic degradation of amyloid-beta peptides by matrix metalloproteinases represents a significant endogenous clearance pathway that has attracted considerable research attention. Several MMPs have demonstrated the capacity to cleave Aβ peptides at multiple sites, generating various cleavage products that may possess different biological activities and aggregation properties [@liao2019]. The degradation of Aβ by MMPs occurs through hydrolysis of peptide bonds within the Aβ sequence, potentially reducing amyloid burden and mitigating downstream neurotoxic effects.
MMP-2 (gelatinase A) effectively degrades Aβ₁₋₄₀ and Aβ₁₋₄₂ peptides in vitro, with cleavage occurring at multiple sites within the Aβ sequence [@roher1994]. Studies have identified specific cleavage sites for MMP-2, including positions ¹⁶-¹⁷ and ³⁴-³⁵ within the Aβ sequence, generating truncated peptides that exhibit reduced aggregation propensity compared to full-length Aβ [@tsubuki2005]. The degradation of Aβ by MMP-2 appears to be enhanced in the presence of certain cofactors and may be influenced by the aggregation state of Aβ, with soluble oligomers potentially being more accessible to proteolytic cleavage than aggregated fibrils [@yin2006].
MMP-9 (gelatinase B) has also demonstrated significant Aβ-degrading activity in multiple studies. Research has shown that MMP-9 cleaves Aβ₁₋₄₀ and Aβ₁₋₄₂ at multiple positions, generating fragments that are less prone to aggregation and neurotoxicity [@wang2018]. The cleavage efficiency of MMP-9 varies depending on the Aβ isoform, with Aβ₁₋₄₂ potentially being more resistant to degradation due to its greater hydrophobicity and tendency to form stable aggregates [@hoshino2022]. MMP-9 activity against Aβ can be modulated by various factors, including tissue inhibitors of metalloproteinases (TIMPs), oxidative stress, and post-translational modifications [@baker2002].
MMP-3 (stromelysin-1) represents another important Aβ-degrading enzyme with distinctive characteristics. MMP-3 can degrade both soluble and aggregated Aβ, and its expression can be induced by inflammatory stimuli and Aβ itself [@lim2017]. Interestingly, MMP-3 activation has been associated with microglial activation and neuroinflammation, suggesting complex interactions between Aβ clearance and inflammatory processes [@deb2003]. Studies have demonstrated that MMP-3 cleavage of Aβ generates fragments that may have reduced toxicity and could potentially serve as diagnostic biomarkers [@yoshiyama2000].
Beyond these major MMPs, additional family members have demonstrated Aβ-degrading activity, including MMP-1 (collagenase-1), MMP-7 (matrilysin), and MMP-13 (collagenase-3) [@lee2021]. Each enzyme exhibits unique substrate specificity and cleavage patterns, suggesting that multiple MMPs may contribute to overall Aβ clearance in the brain. The redundancy in Aβ-degrading enzymes could provide a protective buffer against Aβ accumulation, but this system may become overwhelmed or dysregulated in AD [@vitek1996a].
The regulation of MMP-mediated Aβ degradation involves multiple layers of control, including transcriptional regulation, post-translational activation, and inhibition by endogenous inhibitors such as TIMPs [@brew2000]. TIMP-1, TIMP-2, TIMP-3, and TIMP-4 constitute the endogenous tissue inhibitors of metalloproteinases that regulate MMP activity by forming complexes with active MMPs [@murphy1995]. The balance between MMPs and TIMPs determines the net proteolytic activity available for Aβ degradation, and alterations in this balance have been reported in AD brain tissue [@van2008].
MMP-2, also known as gelatinase A or 72 kDa type IV collagenase, is constitutively expressed in various brain cell types, including neurons, astrocytes, and endothelial cells [@conant1999]. Unlike many other MMPs, MMP-2 is produced as a proenzyme (pro-MMP-2) that requires activation by membrane-type MMPs (MT-MMPs) or other proteases [@sato1995]. The activation of pro-MMP-2 occurs at the cell surface through a complex mechanism involving MT1-MMP (MMP-14) and TIMP-2, which serve as a receptor and inhibitor, respectively [@kinoshita2000].
In the context of Alzheimer's disease, MMP-2 has been implicated in both protective and pathogenic processes. Studies have demonstrated reduced MMP-2 activity in AD brain tissue compared to age-matched controls, potentially contributing to impaired Aβ clearance [@cudaback2015]. This reduction in MMP-2 activity may result from decreased expression, impaired activation, or increased inhibition by TIMP-2 [@del2014]. The decreased MMP-2 activity represents a potential therapeutic target, as enhancing MMP-2 expression or activity could restore Aβ clearance capacity.
Research has shown that MMP-2 can degrade Aβ₁₋₄₀ and Aβ₁₋₄₂ peptides, generating non-toxic or less-toxic fragments [@yan2006]. The cleavage sites for MMP-2 in the Aβ sequence include positions within the hydrophobic C-terminal region, which is critical for Aβ aggregation [@ishida1999]. Importantly, MMP-2 can degrade both soluble and fibrillar Aβ, although aggregated forms may be less efficiently cleaved due to limited accessibility [@ma2018].
Beyond direct Aβ degradation, MMP-2 influences AD pathogenesis through additional mechanisms. MMP-2 participates in synaptic plasticity and memory formation, and its dysregulation may contribute to cognitive deficits in AD [@verslegers2015]. MMP-2 also modulates neuroinflammation and can influence blood-brain barrier integrity, with implications for Aβ transport and clearance [@choudhury2012]. The multifaceted roles of MMP-2 in AD highlight its potential as a therapeutic target, although careful consideration of its diverse functions is necessary.
MMP-9, also designated gelatinase B or 92 kDa type IV collagenase, is inducibly expressed in response to various stimuli, including inflammatory cytokines, growth factors, and Aβ itself [@opdenakker2001]. MMP-9 is synthesized as a proenzyme (pro-MMP-9) that can be activated by various proteases, including MMP-3, plasmin, and certain membrane-type MMPs [@nagase1997]. The activity of MMP-9 is tightly regulated by TIMP-1, which forms a stoichiometric complex with active MMP-9 [@waring2019].
Elevated MMP-9 expression and activity have been consistently reported in AD brain tissue, cerebrospinal fluid, and plasma [@bjerke2016]. This upregulation appears to be driven by inflammatory stimuli and Aβ accumulation, suggesting a potential compensatory response to increased amyloid burden [@kim2007]. However, the elevated MMP-9 in AD may not necessarily translate to enhanced Aβ clearance, as the enzyme may be functionally impaired or inhibited by excess TIMP-1 [@lindberg2015].
The role of MMP-9 in Aβ degradation has been extensively studied, with evidence supporting its capacity to cleave multiple Aβ species [@mizoguchi2009]. MMP-9 efficiently degrades Aβ₁₋₄₀ and Aβ₁₋₄₂, generating truncated fragments with reduced aggregation propensity [@yao2010]. Studies in animal models have demonstrated that MMP-9 can reduce amyloid plaque burden and improve cognitive function when overexpressed or activated [@talamaga2018]. Conversely, MMP-9 deficiency in mice results in increased Aβ accumulation and accelerated cognitive decline, supporting a protective role [@wang2019].
However, MMP-9 also contributes to deleterious processes in AD. MMP-9 can degrade components of the extracellular matrix and basal lamina, potentially disrupting blood-brain barrier integrity [@lee2020]. MMP-9 activity has been linked to neuronal damage and synaptic dysfunction in various contexts, and its upregulation in AD may reflect both protective and harmful responses [@gu2015]. The net effect of MMP-9 in AD likely depends on the cellular context, disease stage, and balance between Aβ-degrading activity and pro-inflammatory effects.
MMP-3, also known as stromelysin-1 or transin, is expressed in various brain cells and can be induced by multiple stimuli, including cytokines, growth factors, and Aβ [@chandler1997]. MMP-3 plays important roles in extracellular matrix remodeling and can activate other MMPs, including pro-MMP-1, pro-MMP-7, pro-MMP-8, and pro-MMP-9 [@yu2008]. This activation function positions MMP-3 as a key regulator of the MMP cascade with implications for Aβ metabolism.
The involvement of MMP-3 in AD has been supported by multiple lines of evidence. MMP-3 expression is elevated in AD brain tissue, particularly in association with amyloid plaques and activated microglia [@yoshiyama2000a]. Studies have demonstrated that MMP-3 can directly degrade Aβ peptides and can also activate pro-MMP-9, indirectly enhancing Aβ clearance [@kim2018]. The activation of MMP-9 by MMP-3 creates a functional link between these enzymes in Aβ metabolism.
MMP-3 also influences AD pathogenesis through effects on neuroinflammation and cell survival. MMP-3 can activate pro-inflammatory cytokines and contribute to microglial activation, potentially exacerbating neuroinflammation [@mcclure2020]. Conversely, MMP-3 has been implicated in apoptotic pathways and may contribute to neuronal loss in AD [@van2007]. The dual roles of MMP-3 in both Aβ degradation and neuroinflammation make it a complex therapeutic target.
Genetic studies have investigated the relationship between MMP3 polymorphisms and AD risk, with some studies suggesting associations between certain MMP3 variants and disease susceptibility [@chen2015]. However, the results have been inconsistent, and further research is needed to clarify the role of MMP3 genetic variants in AD pathogenesis.
The recognition of matrix metalloproteinases as endogenous Aβ-degrading enzymes has opened promising therapeutic avenues for Alzheimer's disease treatment. Strategies aimed at enhancing MMP-mediated Aβ clearance could potentially reduce amyloid burden, slow disease progression, and improve clinical outcomes [@saido2012a]. However, the multifaceted nature of MMPs and their involvement in various physiological processes necessitate careful consideration of potential risks and benefits.
One therapeutic approach involves direct enhancement of MMP expression or activity. This could be achieved through pharmacological agents that increase MMP gene transcription, promote pro-MMP activation, or reduce endogenous inhibition by TIMPs [@bonomi2017]. Several compounds have been identified that can upregulate MMP expression, including certain flavonoids, non-steroidal anti-inflammatory drugs, and natural products [@vines2019]. However, the broad-spectrum activation of MMPs could lead to unintended consequences, including enhanced inflammation and tissue damage.
Another strategy focuses on specific MMPs with favorable Aβ-degrading profiles. Given the complexity of MMP functions, targeting individual enzymes such as MMP-2, MMP-9, or MMP-3 may offer more selective therapeutic benefits [@zhou2021]. Peptide-based or small-molecule activators of specific MMPs could potentially enhance Aβ clearance while minimizing off-target effects. Additionally, delivery of recombinant MMPs directly to the brain represents an alternative approach that has shown promise in preclinical studies [@sierau2022].
Inhibition of TIMPs to enhance MMP activity represents another therapeutic strategy. Since TIMPs naturally inhibit MMP activity, reducing their levels or blocking their interaction with MMPs could increase available proteolytic activity for Aβ degradation [@amour1998]. However, TIMPs also have MMP-independent functions, and their inhibition could have broader physiological consequences that require careful evaluation.
Gene therapy approaches targeting MMP expression have been explored in preclinical models. Viral vector-mediated delivery of MMP genes to the brain could potentially provide sustained expression and activity [@zhang2023]. Studies in animal models have demonstrated that MMP-9 gene delivery can reduce amyloid plaque burden and improve cognitive function, supporting the therapeutic potential of this approach [@li2018].
Challenges in developing MMP-based therapies include the blood-brain barrier, which limits delivery of therapeutic agents to the central nervous system [@pardridge2005]. Strategies to enhance brain penetration, including nanoparticle delivery systems and receptor-mediated transport, are actively being investigated. Additionally, the optimal timing of intervention remains unclear, as MMP-based therapies may be most effective in early disease stages before extensive amyloid accumulation has occurred.
Current research on amyloid-beta metalloproteinases in Alzheimer's disease encompasses diverse approaches, from basic science investigations to translational studies and clinical trials. Significant progress has been made in understanding the mechanisms of MMP-mediated Aβ degradation, identifying novel regulators of MMP activity, and developing therapeutic strategies targeting the MMP-Aβ axis.
Preclinical research continues to elucidate the molecular mechanisms governing MMP expression, activation, and Aβ degradation. Studies have identified signaling pathways that regulate MMP expression in brain cells, including the NF-κB, MAPK, and PI3K/Akt pathways [@chakraborti2003]. Additionally, research has characterized the structural determinants of Aβ cleavage by specific MMPs, providing insights that could inform the design of engineered enzymes with enhanced Aβ-degrading activity [@huang2020].
Animal models of AD have proven invaluable for evaluating MMP-based therapeutic strategies. Studies in transgenic mouse models have demonstrated that MMP activation can reduce amyloid plaque burden, decrease soluble Aβ levels, and improve cognitive performance [@jha2019]. However, the translation of these findings to human disease has proven challenging, highlighting the need for better model systems and more rigorous preclinical validation.
Clinical research on MMPs in AD has included biomarker studies, genetic associations, and therapeutic trials. Cerebrospinal fluid and plasma MMP levels have been investigated as potential biomarkers for AD diagnosis and progression [@horup2022]. While some studies have reported elevated MMP-9 in AD patients, the results have been variable, and MMPs have not yet been validated as reliable clinical biomarkers.
Genetic studies have examined polymorphisms in MMP genes as potential risk factors for AD. Several MMP polymorphisms have been associated with altered AD risk in meta-analyses, including variants in MMP9 and MMP3 [@lin2019]. However, the effect sizes are generally modest, and replication across populations has been inconsistent. Further research is needed to clarify the role of MMP genetic variants in AD susceptibility.
Clinical trials targeting MMPs in AD have been limited. The complexity of MMP biology and concerns about potential adverse effects have hampered clinical development. Trials of broad-spectrum MMP inhibitors in other indications have revealed musculoskeletal side effects, highlighting the challenges of targeting these enzymes [@coussens2002]. More selective approaches targeting specific MMPs or their activation may offer improved safety profiles.
Emerging research areas include the development of MMP-based gene therapies, engineered Aβ-degrading enzymes, and small-molecule modulators of MMP activity [@wang2024]. Additionally, combination approaches that target multiple Aβ clearance mechanisms, including MMPs, may prove more effective than single-target strategies. The identification of biomarkers that predict response to MMP-targeted therapies could enable personalized treatment approaches.
Recent advances in single-cell transcriptomics and proteomics have provided new insights into MMP expression in specific brain cell types in AD [@mathys2019]. These technologies hold promise for identifying cell type-specific MMP dysregulation and developing targeted interventions. Furthermore, computational approaches to predict MMP-substrate interactions and design enzyme variants with enhanced Aβ specificity are areas of active development [@lavecchia2019].
The integration of MMP research with broader AD therapeutic development efforts reflects the growing recognition that multiple mechanisms must be addressed to effectively treat this devastating disease. While significant challenges remain, the progress in understanding MMP-mediated Aβ clearance provides hope for developing novel disease-modifying therapies that could complement existing approaches.
PMID: 36870981(https://pubmed.ncbi.nlm.nih.gov/36870981/)
The following diagram shows the key molecular relationships involving amyloid-beta-metalloproteinases discovered through SciDEX knowledge graph analysis: