Express Yourself: MMP-9 Saves the Day

Alzheimer’s Disease is the most common neurodegenerative disease, and is highly prevalent in the elderly population. The signature of this debilitating disease is the formation of amyloid-β (Aβ) oligomers. These oligomers that gradually appear with old age aggregates into insoluble amyloid plaque that can be found throughout the neural tissue. The soluble form of these oligomers is linked to cell toxicity and general neurodegenerative symptoms found in Alzheimer’s patients. Aβ oligomers are formed by the cleavage of the amyloid precursor protein (APP) via β-secretase and γ-secretase. As discussed in our course, Aβ oligomers have been associated with faulty insulin signaling in the brain (Duarte et al., 2012). Impairment of the insulin signaling pathway has been linked to cognitive defects in Alzheimer’s patients (de la Monte, 2009). Aβ oligomers bind to the membrane surface, which removes the insulin receptors (IRs) at the synapses and leads to stimulation of pro-apoptotic pathways activated by c-Jun N-terminal kinase (JNK) pathway. This activation phosphorylates insulin substrate-1 (IRS1) at Serine residue 636 (pIRS1-Ser636), which makes IRS1 inactive and suppresses insulin.

To counter this, the insulin signaling pathway down regulates oligomeric binding sites. Normal insulin function results in the binding of insulin to its receptor, which activates IRS1 by phosphorylating at Tyrosine residue 465 (pIRS1-Tyr465)(Kaminari et al.). This phosphorylation activates cell survival Akt/PKB kinase (Akt), which promotes insulin’s neurotrophic effects. In this state, Akt phosphorylates/ inactivates Glycogen Synthase Kinase-3β (GSK-3β), which is an enzyme associated with the hyper-phosphorylation of the Tau protein (Kaminari et al.). This is another hallmark characteristic of Alzheimer’s Disease.

Matrix metalloproteinase 9 (MMP-9), is a metalloproteinase, a family of zinc bound proteolytic enzymes that are linked to the remodeling of the extracellular matrix (Egeblad et al., 2002). MMP-9 has been associated with the configuration of synaptic connections. It has also been shown to breakdown Aβ in vivo and in vitro, which leads to decreased levels of neurotrophic factors like BNDF (Kaminari et al.). In their paper, “Overexpression of matrix metalloproteinase-9 (MMP-9) rescues insulin-mediated impairment in the 5XFAD model of Alzheimer’s disease”, Kaminari et al. conducted a series of experiments testing the effect of MMP-9 on the insulin signaling pathway of Alzheimer’s Disease 5XFAD (insulin impairment), TgMMP-9, 5XFAD/TgMMP-9, and wild type mice models .

First, they looked over the phosphorylation of IRS1 at Serine-636 (pIRS1-Ser636) in all mice models. Protein samples from primary and 3 month old hippocampal cell lysates showed that 5XFAD mice had the highest increase levels of pIRS1-Ser636 compared to wild type mice. Conversely, this was not observed in MMP-9 overexpressing mice in which pIRS1-Ser636 levels were reduced to standard levels in 5XFAD/TgMMP-9 and TgMMP-9 mice. They also observed a significant decrease in activating phosphorylation levels of IRS1 at Tyrosine-465 (pIRS1-Tyr465), which is consistent with insulin resistance associated with pIRS1-Ser636. Immunofluorescence labeling of hippocampal neurons from 5XFAD mice revealed levels of pIRS1-Tyr465 were reduced by 50% compared to the wild type mice. However, in the presence of MMP-9, 5XFAD/TgMMP-9 and TgMMP-9 mice’s levels of pIRS1-Tyr465 phosphorylation stayed within normal range. These findings proved that overexpression of MMP-9 enhanced IRS1 activation.

Next, the researchers observed the effects of MMP-9 in signaling kinases downstream of IRS1. They examined phosphorylation levels of Akt at Serine 473 (pAkt) and GSK-3β-pSer9 (pGSK-3β) in primary hippocampal neurons and in hippocampal extracts of 3 month-old mice. Hippocampal pAKT and pGSK-3β levels observed in 5XFAD mice declined significantly compared to wild type mice. Overexpression of MMP-9 in other mice models showed opposite results. The decrease in Akt and GSK-3β phosphorylation was not observed. This proved that overexpression of MMP-9 increased the phosphorylation levels of Akt and GSK-3β.

Following this, Kaminari et al. examined the effects of overexpression of MMP-9 on the activation of TrkB via Tyrosine phosphorylation, which oversees the activation of IRS1. Westernblots of hippocampal cultures showed that in 5XFAD mice, phosphorylation of TrkB was reduced compared to the wild type model while TgMMP-9 and 5XFAD/TgMMP-9 both maintained normal phosphorylation levels. Hippocampal cultures also revealed that 5XFAD mice had significantly lower levels of the neurotrophic factor, BNDF. This was the opposite in TgMMP-9 and 5XFAD/TgMMP-9 mice. In addition to this, the researchers gave a dose of BNDF to the primary cultured cells of 5XFAD mice to observe pIRS1-Tyr465 and pGSK3-β levels. A significant increase in pIRS1-Tyr465 and pGSK3-β levels were observed compared to untreated cells. These findings confirmed that IRS1 activation is regulated by BDNF.

In their last series of experiments, Kaminari et al. looked at the effects of overexpression of MMP-9 on the phosphorylation of JNK. JNK is a crucial stress-activated kinase in peripheral insulin resistance. It is also involved in IRS1 Serine phosphorylation. In 5XFAD mice, JNK phosphorylation levels were significantly increased compared to wild type mice. In 5XFAD/TgMMP-9 mice, where there was overexpression of MMP-9, JNK phosphorylation levels stayed at normal range and cell apoptosis was lower than in 5XFAD mice. These findings confirmed that overexpression of MMP-9 reduced hippocampal phosphorylation of JNK and significantly reduced occurrences of cell apoptosis.

These sets of experiments have opened a door to a treatment option worth testing for Alzheimer’s Disease patients who also have Type 2 Diabetes. Overexpressing MMP-9 in neural tissue may be a treatment worth pursuing.



Figure 1. Kaminari et al. proposed mechanism for MMP-9’s neuroprotective function in inhibiting amyloid β mediated impairment of the insulin survival pathway. Aβ oligomers stimulates TNFR signaling, which is linked to pro-apoptotic pathways. This activates KNJ kinase, which results in IRS1-Ser636 phosphorylation and GSK-3β activation. This leads to insulin resistance. MMP-9 reduces Aβ peptide accumulation, which effectively blocks Aβ oligomers from negatively affecting insulin signaling, and contributes to increased BDNF levels. BDNF binds to TrkB receptors, which induces Akt activation.



1. Kaminari, A., Giannakas, N., Tzinia, A. & Tsilibary, E. Overexpression of matrix metalloproteinase-9 (MMP-9) rescues insulin-mediated impairment in the 5XFAD model of Alzheimer’s disease. Scientific Reports 7, (2017).

2. Duarte, A. I., Moreira, P. I. & Oliveira, C. R. Insulin in central nervous system: more than just a peripheral hormone. Journal of aging research 2012, 384017 (2012).

3. de la Monte, S. M. Insulin resistance and Alzheimer’s disease. BMB reports 42, 475–481 (2009).

4. Egeblad, M. & Werb, Z. New functions for the matrix metalloproteinases in cancer progression. Nature reviews. Cancer2, 161–174 (2002).

10 Replies to “Express Yourself: MMP-9 Saves the Day

  1. Hi Aisha,
    I thought it was really interesting that the authors were able to identify a substrate that already exists in the cell as a possible way to block oligomerization. In cell bio, we learned that overexpression of MMPs is typically linked to metastasis and tumor growth, so I was initially confused that the authors were pursuing overexpression as treatment. Because it’s in the brain, I’m sure that the role of MMP-9 overexpression may not be the same as it would in the body. I was wondering if you saw any discussion elsewhere that may indicate concerns with a possible role in tumor growth or indicate why the role of MMPs in cancer is not a concern within the context of brain tissue.

    1. Hi Lena,
      I had the same thoughts when I first read the paper because we’re actually going through that unit in cell biology. However, I was able to find some research done by Reinhard et al. called “A delicate balance: role of MMP-9 in brain development and pathophysiology of neurodevelopmental disorders”. It’s been found that MMP-9 not only cleaves components of the extra cellular matrix, but plays a major part in restructuring of synaptic networks in the adult brain. However, unregulated MMP-9 has been linked to multiple neurodevelopment disorders. Researchers are finding that if they are able to properly regulate MMP-9, it will help regain a state of plasticity and reverse some cognitive deficits.

  2. Hi Aisha.Wow, the implications of this paper seem very clear. As Lena pointed out, it’s interesting how MMP-9 is already present in the neural synapse and can also enhance IRS1 activation and increased the phosphorylation levels of Akt and GSK-3β. I’m wondering if MMP-9 levels are found to be decreased in Alzheimer’s, just as low insulin levels due to type II diabetes can be found in Alzheimer’s.

    1. Hi Melanie,
      MMP-9 enzymatic levels were found to be lower than normal in neurodegenrative disorders due to lack of expression. In a series of experiments in the article I sent to Lena ( the researchers found that genetic manipulation of MMP-9 affected expression and of course enzymatic activity. However, since this is fairly new, more research needs to be done.

  3. Hi Aisha,

    First of all, I want to applaud you. This post was unbelievably informative. The wealth of background information you provided was extremely helpful for understanding the implications of the article you discussed, so I am very thankful that you took the time to include it.

    I have one question regarding the background information. Do the oligomers have any normal function that is unrelated to AD? As far as I was aware, they are solely implicated in the disease state. If they only function in causing AD, then why do you think that cells would naturally express binding sites for the oligomers (they must do so in order to be able to downregulate the sites)?

    As for the article itself, it seems pretty obvious that the authors stumbled on something amazing – that MMP-9 overexpression solves many of the problems responsible for the AD disease state. However, as for treatment options, how would it be possible to go about forcing brain cells to overexpress MMP-9? Could another treatment option be supplementation with BNDF? Or would the molecule be unable to cross the blood-brain barrier?

  4. Hey Aisha,
    I really enjoyed reading your article. If MMP-9 is a zinc-dependent, works by breaking down extracellular matrix and remodels damaged tissues, so wouldn’t that require the migration of different cellular components ( for example: leukocytes (granulocytes, lymphocytes, macrophages, etc.))? Wouldn’t these components be dangerous to have in great amounts within the brain tissues? How does the body entirely deal with overexpressed MMP-9 and how to restore trace element stability that was altered by increased MMP-9? Would MMP-9 recognize everything as damaged and then recruit all types of cells to attack its own tissues in addition to Aβ oligomers causing further brain damage?

  5. Hey there Aisha,
    Good job in unpacking the article in a more understandable way. Based off the large amount of technical terms in your review I can imagine how much was in the actual article itself. There’s just a couple of clarification questions I want to ask to make sure I have a solid understanding.
    You mention this neurotrophic factor BNDF a few times. What exactly is this and does it have any relevance to the functionality the authors were measuring or was it just a notable change in overexpressing MMP-9?
    One other thing, while talking about the initial experiment, you mentioned that the authors measured phosphorylation levels of Akt at Serine 473. You originally mention that phosphorylation at serine 636 inhibits insulin intake and phosphorylation at tyrosine 465 leads to normal insulin pathway stimulation. So what was the significance of looking at Serine 473?

  6. While the study will surely help lead to the development of new therapies for people with diabetes and Alzheimer’s disease, I’m unsure of what MMP-9 might hold for people who have Alzheimer’s but not diabetes or insulin resistance. As MMP-9 affects so many enzymes that are involved in insulin signaling pathways, couldn’t treatment with MMP-9 mess up these pathways in people who do not have diabetes? If so, treatment with MMP-9 might cause even more problems for people who have Alzheimer’s but not diabetes. This article demonstrates just complex and interconnected cellular signaling is. While cellular signaling’s complexity and interconnectedness could hold the key for some further treatments, it also seems like it can create side effects and other unpleasantries that can result from treatments that alter cellular signaling.

  7. Alzheimer’s is such a debilitating disease that affects so many, so it is great to see any sort of advancement in the field of treating this disease. I am curious as to how we can get Alzheimers patients to overexpress MMP-9. Would we be able to do this with a drug? And what exactly would this drug target?

  8. Hi Aisha! This was a very interesting spotlight! I appreciate how you were able to connect the article to your CUE project. As we learned lecture, the connection between insulin resistance and Alzheimer’s Disease is a relatively new concept. I think it’s exciting how researchers are finding more and more connections between the two disease mechanisms. I, like Beth, am interested in how MMP-9 could be used as protein therapy. I don’t really understand how it works and would love to know more!

Leave a Reply

Your email address will not be published. Required fields are marked *