Clinical strategies and animal models for developing senolytic agents
Introduction
Aging is associated with increasing predisposition to multiple chronic diseases: atherosclerosis, cancers, dementias, diabetes, arthritis, and many others (Goldman et al., 2013, Kirkland, 2013a, Kirkland, 2013b, Research, A.f.A., 2012). Chronological age is the biggest risk factor for many of these diseases and in some cases is a better predictor than all other known risk factors combined. One fundamental aging process that is often operative at sites of pathology underlying chronic age-related diseases is cellular senescence (Kirkland, 2013a, Tchkonia et al., 2013).
Cellular senescence refers to the essentially irreversible cell cycle arrest caused by potentially oncogenic and metabolic insults (Tchkonia et al., 2013). Senescent cells can acquire a senescence-associated secretory phenotype (SASP) that involves release of pro-inflammatory cytokines, chemokines, pro-thrombotic factors, and extracellular matrix proteases that cause tissue damage, as well as extracellular matrix proteins that can contribute to dysfunctional tissue architecture or fibrosis. Thus, the adverse pathogenic mechanisms at the tissue level that could be promoted by cellular senescence include chronic inflammation, loss of functional progenitor cells, clotting, and extracellular matrix dysfunction.
Senescent cells accumulate in multiple tissues with advancing age (Tchkonia et al., 2013, Waaijer et al., 2012). Senescent cell burden is, in turn, associated with lifespan. At 18Ā months of age, long-lived Ames dwarf, Snell dwarf, and growth hormone receptor knockout (GHRKO) mice have fewer senescent cells in their fat tissue than age-matched control wild-type animals, while short lived growth hormone over-expressing mice have more (Stout et al., 2014). Caloric restriction sufficient to increase lifespan in mice is associated with decreased expression of p16Ink4a, a senescence marker, in multiple tissues compared to ad libitum-fed animals (Krishnamurthy et al., 2004). Progeroid mice, including mouse models of Werner and HutchinsonāGuilford progerias, as well as Klotho-deficient, Erccā/ā, and BubR1H/H mice have increased senescent cells (Baker et al., 2008, Chen et al., 2013, Eren et al., 2014b, Tchkonia et al., 2013). In comparisons across longer- vs. short-lived mouse cohorts, senescent cell accumulation in liver and intestinal crypts predicts mean and maximum lifespan (Jurk et al., 2014). Cellular senescence can occur at any point during life, even in blastocysts (Meuter et al., 2014) and in the placenta (Rajagopalan and Long, 2012). Indeed, senescence is important in remodeling during embryogenesis (Munoz-Espin et al., 2013, Storer et al., 2013).
These associations between cellular senescence, aging, and age-related pathologies prompted testing if senescent cell clearance ameliorates dysfunction. Genetically targeting senescent cells in INK-ATTAC;BubR1H/H progeroid mice that express a drug-activatable āsuicideā gene only in senescent cells enhanced healthspan (Baker et al., 2011), the portion of the lifespan during which freedom from pain, disability, and dependence is enjoyed (Kirkland and Peterson, 2009). Even clearing only around 30% of senescent cells from these mice led to partial reversal of age-related lipodystrophy and decreased progression of frailty, sarcopenia, and cataract formation (Baker et al., 2011, Tchkonia et al., 2013).
These findings have spurred development of small molecule senolytic agents and other approaches to decrease senescent cell burden, including peptides, RNA interference, and vaccines. For this effort to succeed: 1) appropriate animal models of human age-related diseases need to be developed. 2) Models have to be made to prove that beneficial effects are actually caused through clearing senescent cells by putative senolytic agents. Without this proof, it would be possible that the candidate agent leads to senescent cell clearance, but that phenotypic improvement is due to āoff-targetā effects on non-senescent cells, not directly through senescent cell clearance. 3) Models are needed in which possible side effects of senolytic agents can be tested. It should be noted that even though continual genetic clearance of senescent cells from mice did not lead to any overt side effects during 20Ā months of observation (Baker et al., 2011), there is evidence that cellular senescence has beneficial effects under some circumstances. For example, cellular senescence protects against cancer development, helps to resolve tissue fibrosis during healing, is involved in immune responses, and can contribute to tissue remodeling (Krizhanovsky et al., 2008, Tchkonia et al., 2013, Xue et al., 2007).
Section snippets
Associations between diseases in humans and cellular senescence
Cellular senescence is associated with many of the chronic diseases and disabilities that are the leading drivers of morbidity, mortality, and health costs (Kirkland, 2013a, Tchkonia et al., 2013, Zhu et al., 2014). Senescent cells have been identified at sites of pathology in a number of these conditions and may have systemic effects that predispose to others. These include: 1) metabolic conditions (diabetes, obesity, metabolic syndrome, and age-related lipodystrophy (Minamino et al., 2009,
Potential scenarios for initial proof-of-concept studies of senolytic agents
Initial clinical studies of senolytic agents will most likely involve indications in which short term administration leads to measurable clinical benefits in already symptomatic subjects, rather than studies of lifespan or healthspan. The first clinical studies may need to be publicly funded in academic settings if they involve repurposed agents that are off-patent. Existing agents still under patent, patentable methods of administration (e.g., aerosol, topically, or ophthalmic drops), novel
Animal models of cellular senescence-associated diseases
Animal models reflective of the potential indications for senolytics in humans considered above are needed for the preclinical studies required before proceeding to proof-of-principle human trials. In some cases, genetically modified mice or relevant disease-inducing genetic, pharmacological, or dietary manipulations are available. However, in many cases there are either no models or only ones with imperfections. In the case of human progerias or other syndromes that result from single gene
Conclusions
Indications we feel could be particularly promising for initial proof-of-principle clinical studies as senolytic or SASP-protective agents become available will be in mitigating short- or long-term effects of chemotherapy or radiation, aerosol delivery for idiopathic pulmonary fibrosis, treatment of primary biliary cirrhosis, and injection for osteoarthritis. Disease-emulating animal models reflecting each of these conditions, proof-of-concept testing in animals bred by crosses between these
Conflicts of interest
The authors and Mayo Clinic have a financial interest related to this research. This research has been reviewed by the Mayo Clinic Conflict of Interest Review Board and is being conducted in compliance with Mayo Clinic Conflict of Interest policies.
Acknowledgments
The authors wish to acknowledge the assistance of J. Armstrong in preparing this manuscript. Support was provided by NIH grants AG013975 and AG044396 (Geroscience Network). The authors are grateful for advice and conversations about strategies for translating agents that affect fundamental aging mechanisms into clinical interventions with colleagues during the retreat series supported by the Geroscience Network.
References (83)
- et al.
Decreased osteogenesis, increased cell senescence and elevated Dickkopf-1 secretion in human fracture non union stromal cells
Bone
(2009) - et al.
Monitoring tumorigenesis and senescence in vivo with a p16(INK4a)-luciferase model
Cell
(2013) - et al.
Expression of senescence-associated beta-galactosidase in enlarged prostates from men with benign prostatic hyperplasia
Urology
(2000) - et al.
Inflammatory networks during cellular senescence: causes and consequences
Trends Mol. Med.
(2010) - et al.
Expression of Chr9p21 genes CDKN2B (p15(INK4b)), CDKN2A (p16(INK4a), p14(ARF)) and MTAP in human atherosclerotic plaque
Atherosclerosis
(2011) - et al.
Inflammation, coagulation, and the pathway to frailty
Am. J. Med.
(2009) Translating advances from the basic biology of aging into clinical application
Exp. Gerontol.
(2013)- et al.
Hyperglycemia causes cellular senescence via a SGLT2- and p21-dependent pathway in proximal tubules in the early stage of diabetic nephropathy
J. Diabetes Complications
(2014) RPE cell senescence: a key contributor to age-related macular degeneration
Med. Hypotheses
(2012)- et al.
Senescence of activated stellate cells limits liver fibrosis
Cell
(2008)