The NIA Interventions Testing Program Shows that Fisetin Does Not Extend Life in Mice
The latest results from the NIA Interventions Testing Program (ITP) were recently published. The ITP conducts the most rigorous of animal life span studies, frequently demonstrating that earlier promising results were incorrect. The most interesting outcome from this batch of different interventions is that fisetin, demonstrated to clear senescent cells in mice and improve health measures, did not extend life. In contrast, dasatinib and quercetin, the most well-studied senolytic, has been shown by other groups to extend life in mice, by 36% in one study. This is puzzling!
We might theorize that either fisetin at the senolytic doses used in the ITP study (more frequent dosing for a longer period of time than I might have chosen) produces meaningful harmful side-effects in comparison to less frequent dasatinib and quercetin dosing, or that an ITP-run life span study for dasatinib and quercetin treatment would show no benefit to life span. The former sounds more plausible than the latter, but the data is the data. The ITP researchers consider that the issue may be differences between mouse strains used in various fisetin studies, and this is also interesting if the case, that senescent cell burden and type might be different enough in different strains to produce quite different outcomes.
In genetically heterogeneous (UM-HET3) mice, the Nrf2 activator astaxanthin (Asta) extended the median male lifespan by 12%, while meclizine (Mec), an mTORC1 inhibitor, extended the male lifespan by 8%. Asta was fed at 1840 ± 520 (9) ppm and Mec at 544 ± 48 (9) ppm, stated as mean ± standard error (n) of independent diet preparations. Both were started at 12 months of age. The 90th percentile lifespan for both treatments was extended in absolute value by 6% in males, but neither was significant.
Five other new agents were also tested as follows: fisetin, SG1002 (hydrogen sulfide donor), dimethyl fumarate, mycophenolic acid, and 4-phenylbutyrate. None of these increased lifespan significantly at the dose and method of administration tested in either sex. Amounts of dimethyl fumarate in the diet averaged 35% of the target dose, which may explain the absence of lifespan effects. Body weight was not significantly affected in males by any of the test agents. Late life weights were lower in females fed Asta and Mec, but lifespan was not significantly affected in these females. The male-specific lifespan benefits from Asta and Mec may provide insights into sex-specific aspects of aging.
Senescent cells have been reported as important mediators of the pathophysiology of aging, and senolytics like fisetin (Fis) may play important roles in mediating their effects. Past researchers treated naturally aged or progeroid mutant mice with Fis and found that it reduced cells with senescent markers; for example, C57BL/6 mice at 23 ± 1 months old were given Fis or vehicle for 5 days by oral gavage. Three days later, in inguinal fat, controls averaged 8% SA-β-gal+ cells, while Fis-treated fat had 2%. They also fed mice 500 ppm Fis from 19 months of age and found that the median lifespan was 27 months in controls and 30 months in Fis-treated, with 3 of 8 treated mice outliving all 8 controls.
We elected to use 600 ppm Fis, starting at 20 months of age, since senescent cells are present in potentially harmful quantities starting at that age in mice. We fed either continuously or for 3 days every 2 weeks. Fis, using the doses and route described here, did not significantly lower the amount of p16Ink4a mRNA in UM-HET3 mouse liver, kidney, or brain. p16Ink4a whole tissue mRNA is one marker of senescent cell burden, but it is not a fully sensitive marker of senescence, for example, it is also expressed in other cell types such as activated macrophages. We had hoped that Fis would deplete senescent cells and thus test the important idea that the removal of senescent cells would lead to longer lifespan, but the absence of an effect on p16Ink4a-positive cells and the lack of inflammatory p21Cip1+ cells in older UM-HET3 mice prevented us from addressing this question. Further studies to analyze the types and location of senescent cells that might increase with age in UM-HET3 mice and how they differ from other mouse models in regard to their upregulated senescent cell anti-apoptotic pathways (SCAPs), as well as the use of Fis and other senolytic agents by gavage, might help to clarify these issues.