Quercetin is Probably Not a Useful Senolytic

Senolytic compounds are those that preferentially destroy senescent cells. Since these cells are one of the root causes of aging, there is considerable interest in finding and then quantifying the effectiveness of senolytic compounds. The known and alleged senolytics vary widely in effectiveness and quality of evidence, and quercetin is one of the more dubious examples. I don't think that anyone expects quercetin, on its own, to have a useful level of impact on senescent cells and their contribution to degenerative aging. The study here comes to the plausible conclusion that quercetin really can't achieve that goal. Yes, it is true that the 2015 mouse study of the chemotherapeutic dasatinib and quercetin demonstrated that the two together cleared more senescent cells than dasatinib alone, but synergy with other compounds is a very different story from unilateral effects. Quercetin is a widely used and extensively tested supplement compound. Any significant effect on health resulting from quercetin alone would likely have been discovered many years ago.

Previously, quercetin was reported to be a senolytic in irradiation-induced senescent human umbilical vein endothelial cells (HUVECs). HUVECs are derived from the umbilical cord of newborn babies, and for a long time were the only model of primary human endothelial cells (EC); however, these cells are not the best model of diseases associated with human arterial aging. HUVECs have been shown to differ substantially from primary endothelial cells derived from adult human vasculature. In the current study, we investigated whether quercetin is a senolytic in adult EC, and evaluated whether quercetin 3-D-galactoside (Q3G; hyperoside) would be a more selective senolytic.

Quercetin's low therapeutic/toxic ratio in the HUVEC study raised the possibility that quercetin could significantly injure non-senescent cells. It was unclear whether the proliferation of non-senescent cells could be compensating for some of the quercetin-mediated cell death, thus masking its toxicity to the young cells at the lower concentrations found to be selectively cytotoxic to senescent cells. We used adult human coronary artery endothelial cells (HCAEC), which are microvascular cells, as a relevant model, and generated two groups of cells from them to better understand the effect of quercetin: EP (early passage; young) and SEN (senescent), as a model of an aging tissue.

Our key findings are that quercetin at a concentration that reduced SEN EC also caused significant EP EC cell death, and that there was no evidence of senescent cell-specific cell death mediated by quercetin. Thus, quercetin is not a selective senolytic in adult human arterial endothelial cells, where both EP and SEN cells responded similarly to quercetin's toxicity.

To circumvent quercetin's toxicity on healthy, non-senescent cells, we investigated Q3G, a derivative of quercetin with limited toxicity to endothelial cells, which is processed by senescence-associated beta-galactosidase (SABG) enriched in senescent cells to release quercetin in situ. Q3G could act as a selective prodrug in senescent cells. However, Q3G had no significant toxicity to either EP or SEN EC. The lack of Q3G's toxicity in the current study may be due to Q3G being unable to enter the beta-galactosidase-rich lysosomes, or alternatively, Q3G being able to translocate to the lysosomes to release quercetin, which is further processed into an inert compound.

Link: https://doi.org/10.1371/journal.pone.0190374


i never used that. i am currently using life extension geroprotect ageless cell which was designed with insillico medicine

Posted by: scott emptage at January 16th, 2018 6:27 AM

It is more likely that quercetin is masking certain parts of SASP, such as CD38 for example.

Flavonoid apigenin is an inhibitor of the NAD+ ase CD38: implications for cellular NAD+ metabolism, protein acetylation, and treatment of metabolic syndrome.

Flavonoid Apigenin Is an Inhibitor of the NAD+ase CD38

There is other supporting data but this will do for now. Quercetin does have a wide range of activity but I have been skeptical of its ability to remove senescent cells for quite a while now. It may have a place as a co-therapy in conjunction with true senolytics but alone I think it just masks some inflammatory cytokines.

Posted by: Steve Hill at January 16th, 2018 8:14 AM

This reinforces what I've been thinking. Finding safe and effective senolytics will involve many years of clinical trials.

Posted by: NY2LA at January 16th, 2018 11:05 AM

@NY2LA unity should be publishing phase 1 of their clinical trial this year

Posted by: scott emptage at January 16th, 2018 11:28 AM

The only active senolytic trial in clinicaltrials.gov is the Mayo Clinic's Phase II study for Chronic Kidney Disease with Dasatinib and Quercetin


This should answer some of the above questions more definitively than in-vitro models

I see no trials for Unity beyond their completed exploratory biomarker work in the osteoarthritis


Posted by: Qing Song at January 16th, 2018 12:43 PM

@qing song they are starting trial this year and they should complete phase 1 this year too. That's what I heard

Posted by: Scott emptage at January 16th, 2018 1:26 PM

@Scott emptage

That does not sound right

FDA requires 6+ month data on disease modification in osteoarthritis

As they do not even have registered trial, let alone recruitment active, I question such a time line

Any evidence for a filed IND?

Posted by: Qing Song at January 16th, 2018 1:42 PM

The 2015 mouse study claims that quercetin does have "unilateral effects" in eliminating senescent mouse bone marrow stem cells, even slightly better than the dasatinib and quercetin combo.

Posted by: Florin Clapa at January 16th, 2018 2:02 PM

@Florin Clapa: True, but it is hard to argue against the evidence of decades of human use and studies with no sign of benefits to any of the health issues you'd expect to be positively affected by senolytic activity.

Posted by: Reason at January 16th, 2018 2:27 PM

The data in this study does show that Q selectively kills SC, just not selectively enough to qualify as a true Senolytic by the established criteria. Given that this is an in vitro study, Q is cheap, available, and has almost no side effects, I am going to continue using it. And, IMO, it is not unfathomable to image that the non scenescent cells Q killed in the study were in some way weaked, just not yet Scenescent.

Posted by: JohnD at January 16th, 2018 3:41 PM

Reason, I doubt any human studies got the dosage, frequency, and age ranges right. Even if all of those variables lined up correctly somehow, the effect of just using Q might be too subtle to detect (a somewhat lower cancer incidence over several decades, for instance), especially if you didn't know what to look for.

Posted by: Florin Clapa at January 16th, 2018 5:00 PM

I am taking Prosta-Q from Farr Laboratories. It contains Quercetin. Unfortunately, it's part of a proprietary blend of ingredients, so I don't know how much I've been taking.

Posted by: NY2LA at January 16th, 2018 6:38 PM

@Florin: their argument here is (a) the original Mayo Clinic study reported Q as senolytic in human umbilical cord vascular cells (HUVEC), which (i) "are derived from the umbilical cord of newborn babies" and are thus on their face" not the best model of diseases associated with human arterial aging", and (ii) "have been shown to differ substantially from primary endothelial cells derived from adult human vasculature," whereas these guys used the more relevant primary endothelial cells; and (b) as JohnD says, their selectivity is weak. Indeed, even the original Mayo Clinic report found a fairly narrow therapeutic window, and the new report said it was too small to be useful. Selectivity is really a core feature of a true senolytic: an agent that kills lots of senescent cells and also kills nearly as many normal ells at the same time is not really a senolytic, but a toxin.

Posted by: Michael at January 16th, 2018 9:35 PM

Michael, while Q might not be useful at eliminating senescent HCAECs (I never claimed that it was), it still might be useful in eliminating senescent BM-MSCs. As for Q's toxicity, I doubt it can kill a significant number of normal cells to matter (no toxic side effects were reported for mice, for instance), and even if Q isn't that selective at killing senescent BM-MSCs, it seems to kill enough of them to matter.

Posted by: Florin Clapa at January 16th, 2018 10:50 PM

@Florin: well, the only BM-MSC in which Q was tested Q were derived from Ercc1[-/Δ] mice, which have defects in DNA repair, great increases in the rate of senescence, and shortened lifespan - and, in particular, abnormal bones and bone marrow. So I wouldn't put too much stock in that result without more supporting data. As to the lack of side effects: well, we don't have the suitable controls of D or Q monotherapy in vivo. Possibly the dose of Q is too low to have much effect on either normal or senescent cells and D is responsible for all the observed effects, or possibly the effects would have been more positive without Q dragging them down - or the benefits of Q were balanced by deleterious effects, leaving it a wash. And they only did brief treatments, unlike some of the longer-term work with either genetic approaches, Navitoclax, or UBX0101.

Posted by: Michael at January 17th, 2018 12:32 PM

@Reason I periodically take Q (500mg) but am somewhat concerned about its specificity for only getting rid of scenescent cells. Q does have some history of health benefits in that two very healthy food products used for hundreds if not thousands of years are known to have high Q content, including apples (an apple a day keeps the Dr. away) and onions. The peels of apple are have the highest concentration of Q, and I eat 1-3 apples a day.

Posted by: Biotechy at January 17th, 2018 12:36 PM

I hope everyone feel that small molecular approach in rejuvenation biothechology (and even in medicine as a whole) is a dead end. They are very expensive to design, to develop, to test and they hugely interfere with metabolism and thus have many side effects. Now it is just a waste of time and money to screen in the hope to find anything useful. We need to look forward and make things like Oisin targeted very efficient clearance systems.

Posted by: Ariel at January 17th, 2018 2:09 PM

Michael, SCs are SCs regardless of mouse strain, right? And as I've mentioned before, Q actually killed off more senescent BM-MSCs than either D or D+Q. While it's true that Q monotherapy wasn't tested in vivo, it's unlikely that there would be any big negative side effects at low doses and infrequent administration. So, the bottom line for me is that Q is low risk, and potentially high reward; it may (or may not) reduce senescent BM-MSCs without negative side effects.

Posted by: Florin Clapa at January 17th, 2018 7:25 PM

Well, "SCs are SCs" is a bit of an oversimplification even inside a single wild-type mouse, but leaving that aside, the main point is that a senescent cell with a hypomorphic Ercc1 gene (and in a milieu generated by other Ercc1 hypomorphic cells in a mouse that is actually rather young in narrow chronological terms) is not the same as a senescent cell with an intact gene and situated in an otherwise-normal but aging milieu. This was a question even in the original Van Deursen BubR1 hypomorphic mouse study, but at least in that study you could be confident that they were exclusively hitting p16-expressing cells because the mechanism of ablation was directly tied to p16 expression. Here, the idea is that the supplement is supposed to be working by selectively hitting the anti-apoptotic machinery on which senescent cells differentially rely, but the reality is that Q is a quite dirty drug with multiple MoAs, and could quite possibly be differentially toxic in cells that are both senescent and have sub-par nucleotide excision repair.

Posted by: Michael at January 19th, 2018 2:01 PM

The bioavailability of Q in humans is pretty bad. If the active senolytic species in rodents is the aglycone, then it's probably not going to work in humans. There are non-trivial difference in metabolism of polyphenols between humans and rodents. Humans knock out polyphenols very efficiently through glucuronidation and sulfation.

Posted by: george at March 15th, 2018 10:54 PM

I understand that the digestive tract breaks down quercetin very quickly to compounds that have no senolytic property. Is that true?

Posted by: Lenka at August 17th, 2018 6:04 AM

Post a comment; thoughtful, considered opinions are valued. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.