Skip to Content


More on Metformin and Cancer (and Alzheimer’s)

Metformin: what a weird compound it is. Very small, very polar, the sort of thing you’d probably cross off your list of screening hits. But it’s been taken by untold millions of diabetics (and made untold billions of dollars in the process), because it really does reduce glucose levels. It does so though mechanisms that are still the subject of vigorous debate and which (I might add) were completely unknown when the drug was approved. (I keep running into people who think that mechanism-of-action is some sort of FDA requirement, but it most certainly is not. Not saying that it wouldn’t help, but what the regulatory agencies want are efficacy and safety. As they should).
And evidence has been piling up that the compound does many other things besides. The situation is murky. There was a report in 2009 that suggested that it might exacerbate the pathology of Alzheimer’s. But last summer there was a rodent study that showed (in obese mice) that the compound seemed to improve neurogenerative effects seen in the hippocampus. (Whether this operates in animals, or humans, who are not metabolically impaired is an open question, although metformin is right in the middle of the whole “Type III diabetes” debate about Alzheimer’s, which I’m going to cover in another post at some point soon). Meanwhile, human studies (in the large populations taking the drug) are not saying one way or another just yet. This British analysis suggested that there might be an association, but it’s not for sure.
Then there’s oncology. In 2010 I wrote about the evidence linking metformin use with lower incidence of some types of cancer, and one proposal for the mechanism. Now another paper is out suggesting that the compound works in this regard through modifying the inflammatory cascade. (Note that James Watson also highlighted this lab’s previous work in his recent paper, blogged about here). The summary:

. . .Taken together, our observations suggest that metformin inhibits the inflammatory pathway necessary for transformation and CSC formation. To link our results with previous work on metformin in the diabetic context, we speculate that metformin may block a metabolic stress response that stimulates the inflammatory pathway associated with a wide variety of cancers. . .
. . .We suspect that this glucose- and metabolism-mediated pathway operates in many different cell types, and hence might explain why metformin reduces incidence of different human cancers and why the combination of metformin and chemotherapy is effective on many cell types in the xenograft context. While this pathway is hypothetical and has not been described in molecular terms, our results suggest that components in this pathway might be potential targets for cancer therapy.

The pathway referred to is through Src and IkappaB (of the NF-kB pathway), among others; the paper goes into more detail for those who are interested. There’s a lot of stuff going on in the clinic with metformin added to different chemotherapy regimes, and I very much look forward to seeing the results. On the molecular level, I’d agree with the statement above – there’s a lot to dig into here. The whole intersection of metabolism and cancer is a very large, very complex (and very tricky) area, but you’d have to think that there’s a lot of really useful stuff to be found in it.

17 comments on “More on Metformin and Cancer (and Alzheimer’s)”

  1. Lane Simonian says:

    It is interesting to compare the chemical structure of metformin with memantine (Namenda) in regards to the treatment of Alzheimer’s disease. Both contain methyl groups (that weakly scavenge the main oxidant in Alzheimer’s disease–peroxynitrites) and an amino group (NH2) which partially inhibits the peroxynitrite-mediated nitration of NMDA receptors. The nitration of NMDA receptors results in the efflux of glutamate and the influx of calcium that kills brain cells.
    Metformin may inhibit p38MAPK. P38MAPK leads to the formation of superoxide anions (via NADPH oxidase) and inducible nitric oxide (via Nuclear factor-kappa b) that combine to form peroxynitrites. Peroxynitrites in turn phosphorylate p38MAPK ensuring their continuous production (even after amyloid plaques are removed).
    Peroxynitrites contribute to the hyperphosphorylation of tau proteins and their nitration (which inhibits neurotransmission), inhibit the transport of glucose via lipid peroxidation, and oxidize receptors involved in short-term memory (muscarinic acetylcholine), sleep (melatonin), mood (serotonin and opioid), alertness (dopamine), smell (olfactory), and brain growth (adrenergic). The peroxynitrite-mediated nitration of the insulin receptor substrate may also impede brain growth.
    Metformin likely reduces the nitration of the insulin receptor substrate which may explain how it lowers blood glucose levels. Phenolic compounds will inhibit nitration, too. Methoxyphenols in particular not only inhibit nitration and partially reverse oxidation, they also are the best scavengers of peroxynitrites because they donate two electrons and two hydrogen atoms (ONOO- + 2e- + 2H+=H20 + N02-). Methoxyphenols such as eugenol in rosemary essential oil and ferulic acid, syringic acid, coumaric acid, and vanillic acid in heat-processed ginseng have partially reversed Alzheimer’s disease in human clinical trials.
    Peroxynitrites also cause DNA damage and inflammation and may play a role in the development and progression of several cancers.
    Several studies have indicated that the methoxyphenol eugenol could play a role in cancer treatment. The most recent is the following: Hyang Nam and Moon-Moo Kim, Eugenol with antioxidant activity inhibits MMP-9 related to metastasis in human fibrosarcoma cells.
    Peroxynitrites are not the only oxidant involved in various diseases, but they do play a role in many disease and thus the necessity of studying various peroxynitrite scavengers. I like to think that peroxynitrite scavengers will be to the 21st century what antiobiotics were to the 19th and 20th centuries.

  2. says:

    And an article linking metformin to lower peroxynitrite levels: Meriem Mahrouf-Yorgov, Metformin suppresses high glucose-induced poly (adenosine diphosphate-ribose)polymerase overactivation in aortic endothelial cells.
    Any compound that inhibits peroxynitrite formation delays the onset of Alzheimer’s disease. Any compound which effectively scavengers peroxynitrites treats Alzheimer’s disease. And what applies to Alzheimer’s disease likely applies to other peroxynitrite-mediated disease.

  3. Anonymous says:

    I’m a little confused on your comparison, Lane. Can you go more in to detail?
    – Why do you think the structures of Metformin and Memantine are comparable? The fact that they have methyl groups and primary amines is surely not unique. By your comparison, nearly any amino acid would react with peroxynitrite, and something like isoleucine would be the best! Methyl scavengers of peroxynitrites seems a bit far-fetched. Perhaps I am forgetting my chemistry, but where is the evidence for this? I would think a methyl would only be weakly reactive towards such a species.
    – Structurally, the two compounds are very different. I highly doubt they have similar specificity in any context. Additionally, would metformin even be BBB permeable? It may be protonated at physiological pH.
    – What is the evidence that peroxynitrites phosphorylate p38MAPK? Where does the phosphate group come from? You don’t mean that peroxynitrites themselves phosphorylate p38MAPK do you?
    – Same vein, how do they hyperphosphorylate tau? Also, you have a laundry list of effects caused by peroxynitrites, are they that long lived?
    – Finally, from my understanding, oxidative species found in AD and their associated amyloid formation seems to be a symptom of the disease and not necessarily what is causing the problems themselves. Some evidence suggests (I believe) that these large aggregates are actually protective and scavenge these oxidative species. If that is indeed the case, wouldn’t it seem that going to the source of the oxidant formation would be the best treatment option, and not scavengers of the oxidants themselves? It seems that they, themselves, may not be the causation of AD but are resultant of the disease itself.
    Thanks in advance for helping me understand your comment better!

  4. rico says:

    I would encourage people to read this:
    although they do not identify the receptor the MoA data is very compelling. Many of the other activities ascribed to metformin could be attributed to this MoA – on e thought – metformin targets glucagon receptor…but that might not explain the activities outside the liver…

  5. says:

    Excellent questions and I will try to answer them as best as I can. I will provide citations as I am not a chemist, although I can understand most of the primary research. Plus you may find mechamisms that I may have missed.
    At least some methyl groups appear to weakly scavenge peroxynitrites via a one electron transfer (Kim, et al. Selective peroxynitrite scavenging activity of 3-methyl-1,2-cyclopentanedione from coffee extract).
    I will copy and paste the various oxidation and nitration reactions of peroxynitrites. Though peroxynitrites are short-lived, they are fast reacting and readily cross biological membranes.
    Amino Acids
    December 2003, Volume 25, Issue 3-4, pp 295-311 Peroxynitrite reactivity with amino acids and proteins
    B. Alvarez, R. Radi
    … show all 2 hide
    » Look Inside » Get Access Summary.
    Peroxynitrite, the product of the fast reaction between nitric oxide (•NO) and superoxide O2 •− radicals, is an oxidizing and nitrating agent which is able to traverse biological membranes. The reaction of peroxynitrite with proteins occurs through three possible pathways. First, peroxynitrite reacts directly with cysteine, methionine and tryptophan residues. Second, peroxynitrite reacts fast with transition metal centers and selenium-containing amino acids. Third, secondary free radicals arising from peroxynitrite homolysis such as hydroxyl and nitrogen dioxide, and the carbonate radical formed in the presence of carbon dioxide, react with protein moieties too. Nitration of tyrosine residues is being recognized as a marker of the contribution of nitric oxide to oxidative damage. Peroxynitrite-dependent tyrosine nitration is likely to occur through the initial reaction of peroxynitrite with carbon dioxide or metal centers leading to secondary nitrating species. The preferential protein targets of peroxynitrite and the role of proteins in peroxynitrite detoxifying pathways are discussed.
    Peroxynitrites (via NO) nitrate primary and secondary amines (Tannebaum, DNA damage and cytotoxicity caused by Nitric Oxide) and the phenol group in tyrosine. I believe this is the primary way by which primary amines and phenolic compounds inhibit peroxynitrite-mediated nitration (by in essence acting as sacricifical compounds).
    Here is the evidence that p38MAPK stimulates the production of peroxynitrtes: Byoung Kwon Yoo, et al. Activation of p38 MAPK induced peroxynitrite generation in LPS plus IFN-y-stimulated rat primary astrocytes via activation of iNOS and NADPH oxidase.
    The following article discusses peroxynitrites and p38 phosphorylation, although I cannot see the mechanism identified in the abstract: SW Rabkin P38 MAP kinase in valve interstitial cells is activated by angiotensin II or nitric oxide/peroxynitrite, but reduced by Toll-like receptor-2 stimulation.
    Peroxynitrites play a critical role in the hyperphosphorylation of tau proteins: Zhang, et al. Peroxynitrites induce Alzheimer’s-like tau modifications and accumulation in rat brain and its underlying mechanisms. A couple of ways in which peroxynitrites-mediate tau hyperphosphorylation is by oxidating g protein-coupled recetors and the nitration of tyrosine kinase receptors, the upshot of which GSK3 is not deactivated and instead hyperphosphorylates tau proteins.
    Amyloid plaques contribute to the formation of peroxynitrites in part through the entombment of heme, copper, and zinc. Peroxynitrites are anti-microbial. However, they cause multiple damage to people’s brains. If you remove amyloid plaques, you do not stop the production of peroxynitrites nor reverse the damage they have already done. That is why removing amyloid plaques seems to have only very limited benefit in the treatment of Alzheimer’s disease and then only when attempted early.
    Oxidation is the likely cause of Alzheimer’s disease. Phospholipase C (gamma and beta) are the likely triggers for peroxynitrite formation. Phospholipase C gamma can be inhibited with various phenolics compounds found in several fruits, vegetables, spices, and essential oils and the activity of both enzymes can be inhibited with polyunsaturated fats. That is why several studies indicate that a diet in phenolic compounds and polyunsaturated fats can delay the onset of Alzheimer’s disease (see Cole, Valente, Fernandez-Fernandez).
    There are probably dozens or perhaps hundreds of compounds that will appear effective against Alzheimer’s diseae in vivo or in mice, but some do not cross the brain blood barrier, some are not absorbed well into the bloodstream, some are not absorbed well into the brain, some are excreted or metabolized before they can do any good. Even given that there are a number of compounds that have already worked in human beings.
    I do not know enough about metformin to specifically judge its efficacy (or lack thereof) in treating Alzheimer’s disease. All I can look for is how it might effect specific pathways and enzymes that may either make it helpful or not helpful in the treatment of the disease. I do know enough about methoxyphenols in the treatment of Alzheimer’s disease both in terms of chemistry (thankfully for me the chemistry is not overly complicated or extensive–most of it has to do with electron and hydrogen donation) and biology and from human clinical trials to know that they are often effective peroxynitrite scavengers and can likely be used to treat Alzheimer’s disease.

  6. Vader says:

    Metformin: Effective against diabetes, associated with a lower risk of poor cardiovascular outcomes, dirt cheap, and with an excellent safety record. What’s not to like?

  7. sandy says:

    My Father has been taking Metformin for approx 30 yrs to control Diabetes and has been suffering from Vascular Dementia for the last 10 years, maybe coincidental, but I think more research into this is needed, sooner rather than later.

  8. Christine says:

    Ditto Sandy, my husband also has been taking Metformin for 30 years and is suffering from vascular dementia, diagnosed 10 years ago and I agree that this should be researched.

  9. RKN says:

    It does so though mechanisms that are still the subject of vigorous debate
    The wikipedia page you linked to indicates Metformin is an agonist of AMPK, the increased activity of which up-regulates SHP, which in turn down regulates PEPCK & GL-6-Pase to reduce gluconeogenesis in the liver. Is this controversial?

  10. Derek Lowe says:

    #9 RKN: Check out this recent paper for some references to the controversy. AMPK certainly seems to be part of the story, but not all of it:

  11. Morten G says:

    regarding mechanism:
    Isn’t there generally an inverse relationship between drug size and polypharmacology?
    Of course there are outliers like glucocorticoids but in general?

  12. Helical_Investor says:

    Looking forward to the diabetes-Alzheimer’s post. Probably have to tie in the good/bad oxidative stress debate as well. My fear is that Alzheimer’s is indeed tied to oxidative stress (and thus related to poor glucose control) through a mechanism not unlike a crystallization ‘seeding’ event (which chemist should all relate to). If so, it will continue to be a very very very difficult disease to get a handle on, and one perhaps best addressed for many via lifestyle vs. therapeutic intervention (until its too late, like with diabetes-II?)

  13. Lane Simonian says:

    I left at least one question unanswered. Peroxynitrites appear to contribute to the phosphorylation of p38 MAPK via the nitration of NMDA receptors (and the subsequent influx of calcium).
    This article examines a potential downside of Metformin before the development of Alzheimer’s disease (and perhaps other forms of dementia).
    Antidiabetic drug metformin (GlucophageR) increases biogenesis of Alzheimer’s amyloid peptides via up-regulating BACE1 transcription.
    Metformin increases the activity of phospholipase C gamma (y) which increases the release of BACE1 which increases the production of amyloid plaques. However, Metformin also appears to lower peroxynitrite levels. The authors of the above article suggest the drug should not be used as monotherapy for elders (without Alzheimer’s disease).
    High glucose levels are indeed one of the causes of oxidative stress in Alzheimer’s disease, although it is much more complicated than simply insulin resistance in the brain. I look forward to the next post on this as well.

  14. Anonymous says:

    CNS penetration of metformin: looked for this very thing a few yrs back: I recall finding a paper where they measured it in rat CSF. They reported that CSF levels were ~5% of serum levels. So it does get in – sort of!

  15. By how much does metformin reduce those levels, though? Is it really so much better than anything else on the market?

  16. anonymous says:

    @14 – Couldn’t agree more…metformin is liver targeted via organic cation transporter uptake. Much lower exposure in any peripheral tissue that lacks OCTs, not to mention that pesky blood brain barrier.

  17. Anonymous says:

    @16: Makes sense. Thanks!

Comments are closed.