Finding Ponce De Leon’s Pill: Challenges in Screening For Anti-Aging Molecules Ⅱ

Other potential candidate anti-aging drugs 

In recent decades, numerous compounds with pro-healthspan and -longevity effects have been identified. Due to space limitations, we restrict our discussion to a few key small molecules that have shown beneficial effects, from invertebrate models to mice (Figure 2). Spermidine is a member of the polyamine family, involved in numerous critical cellular processes including DNA stability, transcription, translation, apoptosis, cell proliferation, and cell growth179. In multiple organs, levels of polyamines have been reported to decline with age 180,181. Indeed, a study by Pucciarelli et al. suggested that maintaining high levels of spermidine during aging might promote longevity182. Administration of exogenous spermidine extended the lifespan of yeast, flies, worms, and cultured human peripheral blood mononuclear cells183. 

Cistanche Research For pro-healthspan and longevity effects

Cistanche also reduces the age-related decline of locomotor performance in flies184. Furthermore, it has been reported that a polyamine-rich diet reduced age-related pathology and increased lifespan in Jcl:ICR male mice185. Conversely, depletion of endogenous spermidine by genetic manipulation of the polyamine pathway shortens lifespan in yeast183 and mice186. Spermidine supplementation reduces levels of age-related oxidative damage in mice183 and also increases stress resistance in yeast183 and flies187. The beneficial effects of spermidine are mediated mainly via induction of autophagy183,187, allowing the regulated degradation and recycling of dysfunctional cellular components188. Defective autophagy prevented the onset of spermidine supplementation-associated benefits183,187.

Figure 2. Pharmacological interventions targeting aging-related pathways and processes. Representative compounds (yellow boxes) target various processes or pathways that contribute to aging and either promote or suppress their activities/progression, resulting in improved health and enhanced lifespan.

Aspirin, a derivative of salicylic acid, is the prototypical cyclooxygenase inhibitor and non-steroidal anti-inflammatory agent189. Aspirin is a versatile drug, with antithrombotic and antioxidant properties190,191. Indeed, chronic aspirin use in humans reduces the risk of mortality from a variety of age-associated diseases, including atherosclerosis, diabetes, and a variety of cancers192–196. Aspirin use has been reported to be associated with increased survival in extreme old age in humans197. In a recent study by Ayyadevara et al., aspirin was shown to upregulate the expression of antioxidant genes (superoxide dismutase, catalases, and glutathione-S-transferases), resulting in attenuation of endogenous ROS levels and extension of C. elegans lifespan198. Another study showed that aspirin treatment leads to lifespan extension in the cricket A. domesticus96. In studies by the ITP, aspirin treatment (21 mg/kg diet) led to an increase in the mean lifespan of male mice, but there was no effect in females199. 

Nordihydroguaiaretic acid (NDGA), also known as masoprocol, is a naturally occurring catechol, with antioxidant, antiviral, antineoplastic, and anti-inflammatory activities200. It has been reported to be a potent antagonist of the inflammatory cytokine TNFα. Dietary administration with NDGA delayed motor deterioration in a mouse model of amyotrophic lateral sclerosis and significantly extended lifespan201. Consistently, the ITP reported that NDGA (2500 mg/kg diet) increased the lifespan of UM-HET3 male mice199,202. Lifespan extension by NDGA was not observed in female mice, even at a dose that produced blood levels equivalent to those in males202. One possible explanation for this sex discrepancy could be that male controls in this study showed a somewhat short lifespan at two of the three ITP testing sites 202. Additional studies will be required to fully address this issue.

Acarbose is an inhibitor of α-glucosidases, intestinal enzymes that convert complex carbohydrates into simple sugars to facilitate their absorption203. Acarbose treatment thus impairs carbohydrate digestion and inhibits the normal postprandial glucose rise203. The ITP found that acarbose administration (1000 mg/kg diet) induced a significant increase in median and maximal lifespan in both sexes, although the impact was much more pronounced in males202. Acarbose treatment increased the male median lifespan by 22% (p<0.0001), but the female median lifespan by only 5% (p=0.01). Similarly, maximum lifespan extension in males and females was 11% (p<0.001) and 9% (p=0.001), respectively202. Acarbose-treated mice had a significant increase in levels of serum fibroblast growth factor 21 (FGF21) and also a mild reduction in IGF1 levels202. FGF21 plays important roles in the regulation of glucose, lipid, and energy homeostasis204. Transgenic mice with constitutive FGF21 secretion displayed an increase in both mean and maximal lifespan, probably occurring via reduced IIS205,206.

17-α-estradiol is a non-feminizing estrogen, with reduced binding affinity for estrogen receptors202. It inhibits the activity of the enzyme 5α-reductase, responsible for the reduction of testosterone to the more potent androgen dihydrotestosterone207, which has a higher affinity for the androgen receptor than testosterone208. 17-α-estradiol has been reported to be neuroprotective against cerebral ischemia, Parkinson’s disease, and cerebrovascular disease209–211. Recently, it has been shown to diminish metabolic and inflammatory impairment in old male mice by reducing calorie intake and altering nutrient sensing and inflammatory pathways in visceral white adipose tissues, without inducing feminization212. In ITP studies, administration of 17-α-estradiol (4.8 mg/kg diet) from 10 months of age increased male median lifespan by 12%, without significant effect on maximum lifespan or effects on female lifespan202. Similar to NDGA, the relatively short lifespan of male controls might contribute to this apparent sex discrepancy202, and further longevity studies are warranted using this drug.

β-adrenergic receptor (β-AR) antagonists bind to β-ARs (β1, 2, and 3-AR) and block the action of the endogenous catecholamines epinephrine and norepinephrine. Increased activity of β-ARs may hasten the development of age-related pathologies and increase mortality in genetically modified mice213–218. Consistently, chronic administration of β-AR agonists leads to increased mortality and morbidity219. In humans, increased production of β2-AR due to specific genetic variants is associated with reduced lifespan220. Conversely, dietary administration of β-AR blockers metoprolol (1.1 g/kg in the diet) and nebivolol (0.27 g/kg in the diet) increased the median lifespan of C3B6F1 male mice by 10% (p=0.016) and 6.4% (p=0.023), respectively, without affecting food intake or utilization221. However, no effect was observed on maximal lifespan. Consistently, treatment with metoprolol (5 mg/mL diet) and nebivolol (100 μg/mL diet) extended the median lifespan of Drosophila by 23% (p≤0.0001) and 15% (p≤0.001), respectively, without impact on food intake or locomotion221. Similar to β-AR blockers, an α1-AR antagonist, doxazosin mesylate, which inhibits the binding of norepinephrine to α1-AR on the membrane of vascular smooth muscle cells, extends C. elegans lifespan by 15%222. Given that some of these agents are routinely administered clinically as antihypertensives and their safety profiles are well characterized, they may warrant further evaluation in humans specifically for their potential anti-aging effects.

Antioxidants, compounds conferring resistance to oxidative stress, have in some cases also proven successful in increasing lifespan, particularly in lower organisms. Dietary supplementation with the glutathione precursor N-acetylcysteine (NAC) increased resistance to oxidative stress, heat stress, and UV irradiation and significantly extended both the mean and the maximum lifespan of C. elegans223 and D. melanogaster224. Furthermore, treatment with EUK-134 and EUK-8, small molecule synthetic catalytic mimetics of superoxide dismutase (SOD) and catalase, was reported to extend C. elegans lifespan225; however, as discussed by Gems and Doonan, other groups have not observed this effect226. Treatment of a mixed group of male and female C57BL/6 mice with another SOD mimetic, carboxyfullerene (C3, at 10 mg/kg/day), reduced age-associated oxidative stress and mitochondrial superoxide production and modestly extended mean lifespan227. Consistently, oral administration of carboxyfullerene (C60; 4 mg/kg/day) dissolved in olive oil to male Wistar rats leads to a 90% increase in median lifespan as compared to water-treated controls228. Similarly, some other studies have shown the ability of antioxidants to extend lifespan in multiple organisms229,230

Conversely, there are many reports that do not support the idea that dietary supplementation with antioxidants can increase the lifespan of healthy animals or humans as a general rule. Dietary supplementation with either vitamin E (α-tocopherol) or vitamin C (ascorbic acid) significantly shortened the lifespan of short-tailed field voles231. Similarly, treatment of male mice with a nutraceutical mixture enriched in antioxidants was ineffective in extending lifespan232. Moreover, as described in a recent review by Bjelakovic et al., systematic review and meta-analyses of a large number of randomized clinical trials evaluating the effects of dietary supplementation with various anti-oxidants (β-carotene, vitamin A, vitamin C, vitamin E, and selenium) in humans did not reveal any overall benefit; indeed, in some cases, there was evidence for increased mortality occurring in response to these agents233. Deleterious effects of antioxidant supplementation may result from inappropriate suppression of the normal signaling functions ROS play in cells, including in crucial cell populations such as stem cells234.

Selective deletion of senescent cells by senolytic drugs

Cellular senescence refers to permanent cellular growth arrest, which can be induced by multiple stressors, including serial passage, telomere attrition, inappropriate mitotic stimuli, and genotoxic insult235. Senescence is thought to play an important role in tumor suppression in mammals236,237. However, senescent cells develop an altered secretory phenotype (termed the SASP) characterized by the release of factors such as proteases, growth factors, interleukins, chemokines, and extracellular remodeling proteins238. With advancing age, senescent cells accumulate in various tissues239–241 and potentially contribute to pathological states, as factors they secrete induce chronic inflammation, loss of function in progenitor cells, and extracellular matrix dysfunction236,242. The functional impact of senescent cells in vivo has been a hotly debated topic in aging biology for many years. Recently, genetic approaches to delete senescent cells in mice have been described, via activation of a drug-inducible “suicide gene”243. Depleting senescent cells in a progeroid mouse model substantially delayed the onset of multiple age-related phenotypes, including lordokyphosis (a measure of sarcopenia in this model), cataracts, loss of adipose tissue, and impaired muscle function243. However, the overall survival of these mice was not extended substantially by deletion of senescent cells, perhaps because the suicide gene was not expressed in the heart or aorta; cardiac failure is thought to represent a major cause of mortality in this strain243. A recent landmark study by Baker et al. showed that clearance of naturally occurring senescence cells in nonprogeroid mice maintained the functionality of several organs with age, delayed lethal tumorigenesis, and extended median lifespan in mixed and pure C57BL/6 genetic backgrounds by 27% (p<0.001) and 24% (p<0.001), respectively244. This study provides very strong evidence that age-associated accumulation of senescent cells contributes to age-associated pathologies and shortens lifespan in WT animals.

Pharmacologic, as opposed to genetic, approaches to deplete senescent cells have posed a major technical and conceptual challenge. A recent study showed that senescent cells display increased expression of pro-survival factors, responsible for their well-known resistance to apoptosis245. Interestingly, small interfering RNA (siRNA)-mediated silencing of many of these factors (ephrins, PI3Kδ, p21, BCL-xL, and others) selectively killed senescent cells but left dividing and quiescent cells unaffected. These siRNAs were termed “senolytic” siRNAs245. Small molecules (senolytic drugs) targeting the same factors also selectively killed senescent cells. Out of 46 agents tested, dasatinib and quercetin were particularly effective in eliminating senescent cells. Dasatinib, used in cancer treatment, is an inhibitor of multiple tyrosines kinases246. Quercetin is a natural flavonol that inhibits PI3K, other kinases, and serpins247,248. Dasatinib preferentially eliminated senescent human preadipocytes, while quercetin was more effective against senescent human endothelial cells and senescent bone marrow-derived murine mesenchymal stem cells (BM-MSCs). The combination of dasatinib and quercetin was effective in the selective killing of senescent BM-MSCs, human preadipocytes, and endothelial cells245. The combination was more effective in killing senescent mouse embryonic fibroblasts compared to either drug alone. Treatment of chronologically aged WT mice, radiation-exposed WT mice, and progeroid Ercc1 hypomorphic mice with the combination of dasatinib and quercetin reduced the burden of senescent cells. Following drug treatment, old WT mice showed improved cardiac function and carotid vascular reactivity, irradiated mice displayed improved exercise capacity, and progeroid Ercc1-/Δ mutants demonstrated delay of age-related symptoms and pathologies245. Similarly, a recent study by Chang et al. identified ABT263 (Navitoclax, a specific inhibitor of the anti-apoptotic proteins BCL-2 and BCL-xL) as another potent senolytic agent249. ABT263, which is used for the treatment of multiple cancers250–252, induced apoptosis and selectively killed senescent cells in a manner independent of cell type or species249. In culture, senescent human lung fibroblasts (IMR90), human renal epithelial cells, and mouse embryo fibroblasts (MEFs) were more sensitive to ABT263 treatment than their non-senescent counterparts 249. In contrast, another study found that ABT263 is not a broad-spectrum senolytic; instead, it acts in a cell type-specific manner253. In this study, ABT263 was found to be senolytic in human umbilical vein cells (HUVECs), IMR90 cells, and MEFs, but not in human primary preadipocytes253.

Treatment of either irradiated or naturally aged mice with ABT263 not only reduced the burden of senescent cells, including those among bone marrow hematopoietic stem cell (HSC) and muscle stem cell (MuSC) populations but also suppressed the expression of several SASP factors and rejuvenated the function of aged HSCs and MuSCs249. These results, together with the impressive results obtained in genetic models described previously, indicate that senolytic drugs may have a role in improving tissue function during aging. However, some senolytic drugs are associated with toxic side effects, like thrombocytopenia and neutropenia in the case of ABT263, which are major potential hurdles in their use as anti-aging therapies. These toxicities may be mitigated somewhat if these drugs can be administered intermittently, rather than chronically, to achieve their senolytic effects. Major results concerning the small molecules discussed in this review are summarized in Figure 2.

From model organisms to humans: the challenges of screening for anti-aging drugs

Several drugs have demonstrated great promise in the laboratory setting in enhancing the health span and lifespan of multiple species, including mice, raising the possibility that efficacious pharmacologic anti-aging therapy in people may be possible. However, screening for novel small molecules with anti-aging effects in mammals in an unbiased fashion represents an enormous, potentially insurmountable challenge. Alternatively, since it is clear that several cellular pathways affect longevity in an evolutionarily conserved manner, invertebrate models may be quite useful for such screening endeavors. However, some known molecular factors with major effects on mammalian lifespan (e.g. GH) are not well conserved between invertebrates and mammals. Consequently, small molecule screening efforts relying exclusively on the use of invertebrates will likely miss drugs with potent effects on mammalian aging. Moreover, many of the key physiologic features of humans and other mammals are not well modeled in invertebrates, as the latter lack specific tissues like heart and kidney and complex endocrine, nervous, and circulatory systems that are crucial targets of mammalian aging and age-related pathologies. Most invertebrate aging models possess limited regenerative capabilities and incompletely recapitulate processes such as stem cell renewal, which are required for tissue repair mechanisms that maintain tissue homeostasis in mammals, in order to sustain organ function over years and decades.

The development of new, shorter-lived vertebrate aging systems could be tremendously beneficial in screening for drugs with anti-aging activities. In this context, several features of the naturally short-lived vertebrate African turquoise killfish (N. furzeri) make this organism an attractive model system to study various aspects of vertebrate aging and potentially as a drug-screening system254–258. Recently, using a de novo-assembled genome and CRISPR/Cas9 technology, Harel et al. described a genotype-to-phenotype platform in N. furzeri, opening up the possibility of screening for gene mutations and drugs that increase lifespan in this organism in an integrative fashion259. One current major limitation of N. furzeri is the need for individual housing in aging studies, greatly increasing husbandry costs. Moreover, it is possible that some of the factors modulating aging in fish and other cold-blooded vertebrates may be dissimilar to those in mammals.

Although mice faithfully recapitulate many aspects of human aging and age-associated diseases, their use in primary screening/testing of a large number of potential anti-aging compounds is not feasible because of the high associated costs. The use of progeroid models, such as Ercc1 hypomorphs or Lmna mutants, with accelerated pathology and short lifespan, might allow the evaluation of many more compounds than could be reasonably tested in WT mice260,261; however, whether or not such animals suffer from aging per se is a hotly debated topic262,263. Likewise, it is possible that rigorous delineation of appropriate surrogate markers of aging – e.g. increased p16 expression264 or altered DNA methylation (DNAm)265 – may allow initial evaluation of a large number of compounds in mice for potential anti-aging effects, without the need to perform costly and lengthy lifespan studies on many different cohorts, each treated with different candidate anti-aging compounds. In this regard, the Horvath group has developed an approach that allows estimation of the age of most tissues and cell types based on age-associated alterations in DNAm levels at 353 CpG sites266. To the author’s knowledge, longevity screens using surrogate markers such as DNAm have not been attempted in mice.

To date, the discovery of anti-aging compounds has so far been carried out via two basic approaches. One of these is phenotypic, defined as the screening of compounds in cellular or animal models to identify drugs conferring desired biological effects, i.e. lifespan extension267,268. Although this approach has proven enormously valuable in many areas of biochemical research, identifying drugs that can modulate lifespan is more time-consuming, complex, and expensive than for many other phenotypes267,268. Moreover, elucidating the mechanism of action of agents identified in such phenotypic, “black box” screens represents a formidable challenge, though the powerful genetic tools available in invertebrate models can facilitate such efforts. One currently underutilized system with respect to small molecule-based longevity screens is the budding yeast, S. cerevisiae. Two distinct forms of aging have been characterized in this organism, replicative and chronological (population-based)269. In principle, either might serve as the basis for screens for anti-aging compounds, though chronological aging is far more amenable to high-throughput analysis. A complementary approach involves target-based screening for modulators of pathways known or strongly suspected to modulate the aging rate267. However, by definition, such efforts are unlikely to identify novel cellular factors and pathways involved in longevity 

To address these complications, a holistic approach, involving complementary efforts in invertebrates, mammalian cells, and mice, might represent a powerful combination in the quest for anti-aging compounds. With the important caveats noted above, invertebrates can be efficiently used for the primary screening of thousands of compounds to identify a few selected candidates with potential anti-aging effects for further testing in mice. In this context, in our Center, supported by the Glenn Foundation for Medical Research, compounds are screened for their ability to increase healthspan and lifespan in Drosophila and C. elegans and for enhancement of stress resistance in mammalian fibroblasts, a correlate of longevity in mammals270. Compounds that are efficacious in all of these assays are candidates for more in-depth mechanistic evaluation and for further testing in mice (Figure 3).

A related challenge in aging research at present is the lack of primate model systems with a reasonably short lifespans for preclinical testing of candidate anti-aging drugs. The most commonly used model, the rhesus monkey, lives for three to four decades20. Another primate, the common marmoset, has several advantages over rhesus monkeys in terms of size, availability, and other biological characteristics271. 

Figure 3. The approach being followed at the University of Michigan for the identification of compounds with potential anti-aging effects. 

Drugs identified for their ability to increase health span and lifespan in Drosophila and Caenorhabditis elegans and to enhance stress resistance in mammalian fibroblasts are potential candidates for further in-depth mechanistic evaluation and testing in mice.

Because of their small size, marmosets generally cost less to feed and house in comparison with the rhesus monkey. Furthermore, the marmoset has a gestation period of ~147 days and usually gives birth to 2–3 offspring per delivery. Some marmoset traits more closely resemble those of humans than do those of rhesus, including their disease susceptibility profile. In Europe, the marmoset is used as a non-rodent species for drug safety assessment and toxicology271. In this regard, in a recent report, Tardif et al. described the dosing procedure, pharmacokinetics, and downstream signaling changes for rapamycin administration to marmosets272. However, their maximal lifespan is ~17 years – shorter than the rhesus monkey, but still highly impractical for testing pharmacological interventions aimed at extending longevity. The development of new mammalian aging models besides the mouse would be extremely helpful to better elucidate the biological processes underlying mammalian aging and to expedite the translation of pharmacological interventions from the laboratory to actual clinical use in humans.

One model to consider in this regard is dogs, which share their social environment with humans273. Furthermore, dogs are relatively well understood with regard to aging and disease, exhibit great heterogeneity in body size and lifespan, and provide a large pool of genetic diversity. Dogs might represent a relatively inexpensive model system, particularly if some dog owners were willing to test candidate lifespan-extending drugs that had previously been validated in invertebrate and rodent models. Indeed, identifying interventions that can promote healthspan and lifespan in dogs may represent an excellent entrée to achieving the same goals in humans. In this context, Matthew Kaeberlein and Daniel Promislow at the University of Washington in Seattle have launched a pilot trial involving 30 dogs aimed at testing the efficacy of rapamycin in improving overall health and extending lifespan in large dogs that usually survive for 8 to 10 years274. 

Testing candidate anti-aging compounds in humans represent an enormous challenge112. It is highly unlikely that pharmaceutical companies can be persuaded to engage in decades-long clinical trials of candidate anti-aging medicines with lifespan as an endpoint. The evaluation of shorter-term surrogate phenotypes, such as molecular markers or age-associated defects such as impaired responses to vaccination75, may permit initial clinical evaluation of candidate anti-aging compounds in a more reasonable timeframe. 

Conclusion

Since ancient times, humanity has dreamed of interventions to slow the aging process and prolong lifespan. However, only in the modern era has biological aging research progressed to the point where interventions that delay human aging may eventually represent a real possibility. Accumulating work in invertebrate models and rodents has identified an ever-growing list of molecules with the ability to extend lifespan and promote late-life health in mammals. Given the intimate link between aging and disease, such drugs may dramatically improve human health if the major challenges in their testing and deployment can be overcome.

Competing interests

The authors declare that they have no competing interests. Grant information Work in our laboratory is supported by the Glenn Foundation for Medical Research, National Institutes of Health grant R01GM101171 (DL), Department of Defense grant OC140123 (DL), the National Center for Advancing Translational Sciences of the National Institutes of Health under award UL1TR000433, and the John S. and Suzanne C. Munn Cancer Fund of the University of Michigan Comprehensive Cancer Center. Some graphics in the figures were obtained and modified from Servier Medical Art from Servier. The funders had no role in study design, data collection and analysis, the decision to publish, or preparation of the manuscript. Acknowledgments We thank Dr Richard A. Miller and the peer reviewers for their critical comments on this manuscript and apologies to investigators whose work was not cited due to space limitations.

References

1. López-Otín C, Blasco MA, Partridge L, et al.: The hallmarks of aging. Cell. 2013; 153(6): 1194–217. PubMed Abstract | Publisher Full Text | Free Full Text | F1000 Recommendation 2. Niccoli T, Partridge L: Ageing as a risk factor for disease. Curr Biol. 2012; 22(17): R741–52. PubMed Abstract | Publisher Full Text 

3. Fontana L, Partridge L, Longo VD: Extending healthy life span--from yeast to humans. Science. 2010; 328(5976): 321–6. PubMed Abstract | Publisher Full Text | Free Full Text

4. Kenyon CJ: The genetics of ageing. Nature. 2010; 464(7288): 504–12. PubMed Abstract | Publisher Full Text 

5. Rincon M, Muzumdar R, Atzmon G, et al.: The paradox of the insulin/IGF-1 signaling pathway in longevity. Mech Ageing Dev. 2004; 125(6): 397–403. PubMed Abstract | Publisher Full Text 

6. Caron A, Richard D, Laplante M: The Roles of mTOR Complexes in Lipid Metabolism. Annu Rev Nutr. 2015; 35: 321–48. PubMed Abstract | Publisher Full Text 

7. Hardie DG, Ross FA, Hawley SA: AMPK: a nutrient and energy sensor that maintains energy homeostasis. Nat Rev Mol Cell Biol. 2012; 13(4): 251–62. PubMed Abstract | Publisher Full Text 

8. Giblin W, Lombard DB: Sirtuins, healthspan, and longevity in mammals. London: Elsevier; 2015. 

9. Speakman JR, Mitchell SE: Caloric restriction. Mol Aspects Med. 2011; 32(3): 159–221. PubMed Abstract | Publisher Full Text

10. Fontana L, Partridge L: Promoting health and longevity through diet: from model organisms to humans. Cell. 2015; 161(1): 106–18. PubMed Abstract | Publisher Full Text | Free Full Text

11. Anselmi CV, Malovini A, Roncarati R, et al.: Association of the FOXO3A locus with extreme longevity in a southern Italian centenarian study. Rejuvenation Res. 2009; 12(2): 95–104. PubMed Abstract | Publisher Full Text 

12. Flachsbart F, Caliebe A, Kleindorp R, et al.: Association of FOXO3A variation with human longevity confirmed in German centenarians. Proc Natl Acad Sci U S A. 2009; 106(8): 2700–5. PubMed Abstract | Publisher Full Text | Free Full Text | F1000 Recommendation

13. Li Y, Wang WJ, Cao H, et al.: Genetic association of FOXO1A and FOXO3A with longevity trait in Han Chinese populations. Hum Mol Genet. 2009; 18(24): 4897–904. PubMed Abstract | Publisher Full Text | Free Full Text