The little black mouse’s feet dangled above the table as the animal clutched a horizontal wire with its two front paws. After just a few seconds, it lost its grip and fell onto a pile of bedding below. For a mouse its age—just six months old, typically its physical prime—that was quite unusual. As Mayo Clinic veterinary technicians Christina Inman and Kurt Johnson knew, young mice would usually manage to hoist their hind legs up to the wire so that they’re hanging from all four limbs, allowing them to last minutes, sometimes even hours, on the endurance test.
It was late 2016, and the two vet techs were in charge of testing the physical performance of dozens of young mice as part of a study led by Mayo geriatrician and aging researcher James Kirkland. The experiment was blinded, so Inman and Johnson knew nothing of the animals’ treatments, but they would later learn why that mouse, along with five others, had scored so miserably on the wire task: two weeks before, another mouse’s fat progenitor cells, which had been exposed to irradiation or other stressors, were implanted into their lower abdomens. The transplanted cells had been transfixed into a zombie-like state by the treatment; they couldn’t proliferate, but they wouldn’t die. They had reached cellular senescence.
That getting rid of senescent cells is enough to effectively rejuvenate an animal—that tells you they’re a really important driver of aging.—Lorna Harries, University of Exeter
For decades, scientists had ignored senescent cells—which are trapped in a long-term state of cell cycle arrest—dismissing them as artifacts of cell culture with no significance inside living organisms. But in recent years, Kirkland and other researchers have established senescence as an important physiological process that appears to play seemingly opposing roles in vivo. On the one hand, senescent cells are thought to mediate tissue development when they form in the embryo, and also to promote tissue regeneration and wound repair in later life. However, as these zombie cells accumulate with age, they can ooze inflammatory proteins believed to cause tissue dysfunction and to push neighboring cells into senescence. Indeed, animal studies have suggested that destroying senescent cells can slow down age-related physical decline and boost overall health, and many researchers who study aging now regard senescence as a driver of the physical decline characteristic of old age and a contributor to a range of age-related diseases.
To Kirkland, senescent cells are not unlike pathogens that infect multiple tissues and drive various maladies. Implanting around 1 million senescent cells into each of the six young mice crippled their fitness performance compared with that of control mice that had received implants of non-senescent fat cells. Repeating the procedure with just half a million cells in a slightly older set of mice increased their risk of death within a year—typically due to cancer, lung and gut diseases, or neurodegeneration—compared with controls. “Transplanting senescent cells accelerated most or all of the diseases that mice die of in old age,” Kirkland tells The Scientist. But he had a plan to impede the decline.
In another experiment, technicians fed 20-month-old mice—roughly the equivalent of 57- to 67- year-old humans—a mixture of two drugs that Kirkland had previously demonstrated could selectively kill senescent cells: dasatinib, a drug often used in conjunction with chemotherapy, and quercetin, a flavonoid found in onions and apples. Just two weeks after receiving the combo treatment, the mice ran further, performed better on other physical tests, and were 36 percent less likely to die the following year compared with animals that had received injections of senescent cells but no drug cocktail, Kirkland and his colleagues reported in 2018. Together with other research, these findings fortified the idea that killing off senescent cells can rescue old animals from the physical deterioration that comes with age, and extend the duration of healthy, disease-free life.
Now, researchers, pharmaceutical companies, and investors are working on ways to destroy senescent cells in hopes of one day alleviating the ravages of old age in people. Several biotechs have sprouted in recent years to identify compounds that target senescent cells and put such “senolytic” drugs to the test in humans—efforts that are supported by considerable funding influxes from both federal and private sources. The National Institutes of Health (NIH) has recently awarded several seven-figure grants to support Phase 2 testing of senolytics in cancer treatment, for example, and the prospect of conquering age-related debility has reportedly attracted investment from billionaires such as PayPal cofounder Peter Thiel and Amazon CEO Jeff Bezos.
Making use of the surge in interest, Kirkland and others have cautiously launched a wave of clinical studies to evaluate whether senolytic agents offer promise in treating serious age-related ailments such as osteoarthritis, kidney and lung fibrosis, and Alzheimer’s disease. Although many questions remain about senescence and its role in the aging process, these researchers argue that unlike other agents hyped as elixirs of youth that have repeatedly failed to show benefits in clinical studies, the concept of senolytics will stand the test of time.
“That getting rid of senescent cells is enough to effectively rejuvenate an animal—that tells you they’re a really important driver of aging,” says Lorna Harries, a molecular geneticist at the University of Exeter in the UK who studies cell senescence. “It may be the theory of the month, but to my mind, it’s one of the most persuasive.”
The Interplay of Cell Senescence and Aging
Senescent cells accumulate with age. This may result in higher levels of certain senescence-associated secretory phenotype (SASP) proteins, which researchers believe drive aging-related processes and promote aging-related diseases. And senescence, scientists are coming to understand, is itself mediated by cellular processes associated with aging.
How aging-related cellular processes drive senescence
Untangling which cellular processes drive senescence is a major challenge to researchers, in part because those pathways are interrelated. In addition, there may well be multiple factors that contribute to the accumulation of senescent cells, including the tissue or organ in question, a person’s genetic makeup, and environmental stressors she is exposed to.
This can occur when protective pieces of DNA at the ends of chromosomes grow shorter with successive cell divisions, or when their internal structure unfolds, a process called “telomere uncapping.” Both have been shown to trigger senescence in vitro.
Potentially oncogenic mutations
As DNA repair mechanisms erode with age, cells can acquire certain mutations, particularly those that activate oncogenes, that push cells into senescence.
Several cellular organelles, especially mitochondria, can generate reactive oxygen species (ROS) that damage both mitochondrial and nuclear genomes and thereby drive senescence. The production of ROS is thought to increase with age.
Several other mechanisms may drive senescence. For example, mechanical stress might lead to senescence of cells in joints, and the epigenetic machinery that governs gene expression is thought to become dysregulated with age, which may somehow induce senescence.
How senescence can drive aging-related processes
Chronic inflammation and the exhaustion of stem cells are thought to be the most significant consequences of senescence in aging tissues, although other effects are being explored. The importance of these processes may differ depending on the condition in question.
Senescent cells secrete proinflammatory cytokines, which trigger chronic immune reactions that may drive many aging-related diseases.
Stem cell exhaustion
With age, stem cells throughout the body likely undergo senescence, limiting the body’s ability to regenerate tissue and ultimately compromising tissue function.
Senescence in neighboring cells
Senescent cells can prompt other cells around them to undergo senescence through the secretion of particular signaling proteins, a process termed “paracrine senescence.”
Some studies suggest that the secretions of senescent cells can induce mitochondrial dysfunction in neighboring cells, generating ROS that drive tissue damage and cell senescence in surrounding tissues. Senescent cells have also been implicated in the formation of misfolded proteins such as tau and amyloid-β, aggregations of which are associated with neurodegenerative disease and have been suggested to drive senescence themselves.
A double-edged sword
It was not lost on a young Leonard Hayflick that, while cancer cells seem to multiply into eternity in culture, healthy human fibroblasts do not. At some point after 50 cell divisions, fibroblasts become enlarged and flattened, and they refuse to proliferate. When Hayflick, then at the Wistar Institute in Philadelphia, first made this observation in the late 1950s, most researchers blamed inadequate culture conditions for cells’ apparent lack of growth. But through a series of experiments with cytogeneticist Paul Moorhead, Hayflick demonstrated that the cells entered a state he called senescence due to an intrinsic cell phenomenon, seemingly a response to prolonged replicative stress. Still, the concept was initially dismissed by the scientific community, and even as cellular senescence in cell culture became gradually accepted over the next decade, researchers continued to assume that the phenomenon wasn’t relevant for living organisms.
One of the first hints that they were wrong came in the 1990s from the work of Manuel Serrano, then a postdoc in cell biologist David Beach’s lab at Cold Spring Harbor Laboratory in New York. Serrano observed the same enlarged, flattened morphology Hayflick had described decades earlier—this time in response to overexpression of an oncogenic form of the cell growth regulator gene Ras in murine fibroblasts. Perhaps the cells were senescing as a way to prevent themselves from becoming malignant, Serrano hypothesized. When cells are exposed to forms of stress that would normally trigger cancer—such as DNA damage due to prolonged replication or other cellular injuries that occur with old age—they could undergo senescence as a way of avoiding passing on damage to daughter cells.
Researchers argue that while many agents hyped as elixirs of youth have repeatedly failed to show benefits in clinical studies, the concept of senolytics will stand the test of time.
“This study and later ones gave a ‘biological reason’ for senescent cells to exist, which was to protect from cancer,” says Bill Keyes, a cell biologist who studies senescence at the Institute of Genetics and Molecular and Cellular Biology in France. This idea has since been validated, while Keyes and others have discovered that senescent cells may affect biological processes from embryogenesis to the induction of labor to wound healing. (See “The Bright Side of Senescence” below.)
Yet it soon became clear that senescence has a downside, too. Intrigued by observations of accumulations of senescent cells in aging tissues in humans, cell and molecular biologist Judith Campisi of the Buck Institute for Research on Aging in California decided to take a closer look at the genes and proteins expressed by senescent cells. In 2008, she and her colleagues reported that secretions from senescent cells contained dozens of proteins such as inflammatory and immune-modulating cytokines that could damage neighboring cells. These proteins would later help explain many of the pathological effects of senescent cells in aging tissues.
Meanwhile, Mayo Clinic geneticist Jan van Deursen was investigating the high concentrations of senescent cells he’d found in a mutant strain of mice that displayed accelerated aging, or progeria. After some experiments suggested the cells could be the cause of the animals’ symptoms, van Deursen, Kirkland, and others designed a technique to trigger apoptosis only in cells that produced the protein p16—an imperfect marker of senescence. In 2011, the team reported that using this approach to kill senescent cells delayed the onset of the animals’ cataracts and other age-related pathologies in tissues across the body, and maintained their muscle function for longer. Later, van Deursen’s team demonstrated that the treatment had a similar effect in naturally aging mice.
“There was the key demonstration that actually if you clear [senescent cells], it made a significant difference in health,” Paul Robbins, a molecular biologist at the University of Minnesota, says of van Deursen’s 2011 study. Together with Campisi’s research, “that changed everybody’s thinking” about cellular senescence, Robbins adds.
The Bright Side of Senescence
In the past few decades, researchers have not only uncovered the dark side of senescence, but also its apparent positive effects in living organisms. In the 1990s, cell biologist Manuel Serrano’s work suggested that senescence may act as a mechanism to suppress the formation of tumors. When cells are exposed to genetic or cellular damage that would normally trigger uncontrollable replication, the senescence program would be activated as a means of arresting growth entirely, Serrano and other researchers argue. This would save the cell from passing on damage to its progeny, and the organism from growing tumors. “Senescence protects us from cancer,” Serrano says.
Senescent cells may also be crucial in early development. Research led by Bill Keyes, a cell biologist at the Institute of Genetics and Molecular and Cellular Biology in France, suggests that senescent cells could guide development in chick embryos. In the limb, a senescent cell could “become secretory and start instructing and informing and telling the other cells . . . which cell type to become and how to pattern the limb,” Keyes explains. Waves of senescence also occur in mouse and human embryos, where they may similarly guide cell differentiation and tissue patterning of surrounding cells via the proteins they secrete. This proposed embryological mechanism may be the driver of senescence evolution, likely co-opted later as a cancer prevention feature in maturity, Serrano explains.
Other possible functions of senescence in development continue to emerge. Just last year, for instance, the University of Texas’s Ramkumar Menon, who studies fetomaternal communication, found that cells within the membranes surrounding human fetuses undergo senescence when the fetus is fully developed and generate inflammatory signals in the womb, which he proposes acts as a cue to the mother’s body to initiate labor.
Senescence appears to maintain its importance throughout life: studies indicate that in addition to the proinflammatory factors that senescent cells secrete, they also exude growth factors and enzymes that stimulate tissue repair. For this reason, “we need to be really intelligent about how we manipulate it in the elderly to improve [health later in life],” notes cell and molecular biologist Judith Campisi of the Buck Institute for Research on Aging in California. A major challenge she sees in developing senolytic drugs will be to find compounds that kill groups of senescent cells that are harmful to their environment while sparing beneficial ones. “I think that’s going to be the wave of the future,” she says, “drilling down and understanding this complexity so that the drugs can be more tailored to eliminate subpopulations as opposed to all of them.”
Many researchers now view senescence as a beneficial process that evolved as a developmental and cancer prevention mechanism, but one that came with a tradeoff of the damage senescent cells can cause as they accumulate with age. There are still many unanswered questions about how these cells function, but it is already clear to scientists in the field that senescent cells influence a range of age-related pathologies, at least in rodents. Genetic ablation of senescent cells reduces the number of atherosclerotic plaques in mice, improves cartilage development in mouse models of osteoarthritis, boosts bone strength in murine models of osteoporosis, and even staves off neurodegenerative symptoms in models of Alzheimer’s disease.
These findings have a number of scientists thinking: If clearing senescent cells had such beneficial effects on health, could drugs be developed to do just that?
The birth of senolytics
Kirkland’s group at Mayo had started to search for senolytic agents long before the scientific community was convinced of senescent cells’ role in aging, but it took him years to work out a good strategy to identify them. In the mid-2000s, his team tried developing toxins or antibodies that target senescent cells, but none of these approaches succeeded in killing senescent cells while sparing non-senescent ones.
In 2013, it occurred to Kirkland’s team to target the molecular machinery known to be used by senescent cells to defy death. The cells must have those mechanisms in place to avoid undergoing the apoptotic processes that would typically follow exposure to the high levels of harmful proteins they are producing, the team reasoned. Using a bioinformatics approach, the researchers identified several anti-apoptotic pathways that are upregulated in senescent cells, including certain pathways used by malignant B cells to avoid apoptosis and cause lymphoma. They then screened for approved drugs and natural products that targeted those pathways and thus selectively killed senescent cells.
To the group’s surprise, two compounds appeared very effective in killing senescent cells in vitro as well as in mice: dasatinib, approved in the US to treat certain leukemias and lymphomas, and quercetin, which is used as a nutritional supplement. “I thought we’d have to screen millions of compounds to get drugs that regulate senescence,” recalls Robbins, who was involved in the effort. But it took fewer than 50 drugs to get the first hits.
The drugs’ effectiveness varied starkly depending on the cell type, because different types of senescent cells appeared to use distinct pathways to prevent cell death. Dasatinib, which blocks an enzyme that regulates cell survival, only kills senescent mesenchymal cells, such as adipocyte progenitors and certain myoblasts. Quercetin, which interferes with several anti-apoptotic pathways, mainly takes out senescent endothelial cells, which line animals’ blood vessels and lymphatic vessels. Used in combination in a mouse model of a severe degenerative lung disease called idiopathic pulmonary fibrosis (IPF), infusions of dasatinib and quercetin improved lung function and physical health. And in a mouse model of Alzheimer’s disease, researchers at the NIH reported that the same drug cocktail reduced brain damage and inflammation and slowed the pace of memory loss in those models.
Although these results are promising, van Deursen questions the senolytic mechanism thought to underlie the observed therapeutic effects. Dasatinib and quercetin each act on numerous biological pathways, he notes, making it difficult to ascribe their effects solely to the killing of senescent cells. But Kirkland argues his observations are most likely due to the killing of senescent cells. Because the drugs are short-lived and are administered intermittently, they disappear before they have any consequences other than their senolytic effect, he explains.
Still, many researchers are pursuing compounds that more precisely target pathways involved in senescence. Working in human and mouse cells in vitro, three separate research groups, including Kirkland’s, have identified senolytic properties of navitoclax, an experimental chemotherapy drug that inhibits components of particular B cell lymphoma (BCL) pathways. Companies such as San Francisco–based Unity Biotechnology are also getting in on the hunt for senolytic compounds. Since founder Nathaniel David launched the company in 2011 after seeing van Deursen’s influential study on destroying senescent cells, Unity’s scientists have developed three targeted senolytic compounds and started to test them in early-stage trials: UBX1967 and UBX1325, which both inhibit BCL pathways, and UBX0101, which unleashes the tumor suppressor gene p53 to trigger apoptosis. (Both van Deursen and Campisi are long-time collaborators of Unity.)
Such targeted approaches can also affect non-senescent cells that happen to produce high levels of survival proteins, Kirkland cautions. For instance, navitoclax could have side effects on platelets and neutrophils, which naturally make large amounts of BCL proteins. Some researchers conducting animal and clinical studies administer these targeted senolytics locally to reduce the risk of such side effects, and have reported good outcomes. For instance, in 2017, Campisi and her colleagues put Unity’s UBX0101 to the test in mice whose knees had been injured to induce osteoarthritis, injecting the senolytic drug at the site of injury.9 To the researchers’ surprise, elimination of senescent cells not only reduced inflammation, as expected, but also boosted cartilage development. “It was unexpected that by simple removal of the cells . . . you could actually repair the tissue,” notes Jennifer Elisseeff, a biomedical engineer at Johns Hopkins University who was involved in the research.
More than 20 senolytic drugs have been described so far, some of which have already been tested in mice, and researchers are working on finding more and trying out new approaches for doing so. Serrano, who now runs a lab at Barcelona’s Institute for Research in Biomedicine and advises a senolytics company that he cofounded, says some of his colleagues are working on engineering T cells to target senescent cells and destroy them, a strategy similar to some recently approved cancer immunotherapies. And Harries’s research suggests that, contrary to a long-held view that senescence is permanent, the process may in fact be reversible by repairing old cells’ ability to maintain correct gene expression—a capacity that is thought to decline with age. She holds a patent on particular compounds that may induce this change in senescent cells.
As researchers continue to explore different ways to manipulate senescent cells, some are moving ahead to test senolytics in clinical trials, with a handful of Phase 1 trials now underway pitting senolytics against age-related diseases in humans. “The data that we see in mice is amazingly strong—perhaps to my taste a little bit too strong, meaning maybe we’re not looking carefully enough at possible side effects,” notes Felipe Sierra, who directs the division of aging biology at the National Institute on Aging. Nevertheless, he is optimistic: “I think that our possibilities of translation are pretty high.”
Cautiously into humans
In February 2019, Kirkland, together with collaborators at the University of Texas and Wake Forest University in North Carolina, published results from 14 IPF patients participating in a Phase 1 trial of a dasatinib/quercetin cocktail. Although the primary goal was to evaluate the treatment’s safety, they did see signs that the drugs could be working. After taking three oral doses a week for three weeks, patients could walk further in six minutes than they could at the start of the trial, and they performed better on other tests of their physical abilities, such as standing up from a chair.
Last September, the team reported another encouraging finding in the form of preliminary data from nine patients with diabetes-related kidney disease who had received the senolytic combo as part of an ongoing Phase 1 safety trial. Based on biopsies of fat tissue extracted from the patients before and after the treatment, Kirkland’s team found that cells positive for p16, the senescence marker, were significantly decreased in blood, fat, and skin. To Kirkland, these results are the first indication that the drugs could be clearing senescent cells in people.
Meanwhile, Unity Biotechnologies is moving forward with initial trials of its senolytic drug candidates. In a Phase 1 study of four dozen patients with osteoarthritis, participants said they experienced less pain in their knees three months after receiving one injection of UBX0101 into their knee joints. Physicians found overall improvements in the patients’ joint function over this same time period compared to a control group that received a placebo. Although the study was primarily designed to assess safety, Unity CEO Keith Leonard takes the results as a positive sign. “[The effect] was dose-related: the higher the dose we put in, the greater the signal,” he says.
All three treatments appeared to clear the hurdle for safety, and the US Food and Drug Administration (FDA) has approved Phase 2 studies of the dasatinib/quercetin cocktail for IPF and for diabetes-related kidney fibrosis. For Unity researchers, the FDA greenlighted a Phase 1B extension of the UBX0101/osteoarthritis trial, as well as a Phase 2 study of the compound in 180 osteoarthritis patients. Neither the Mayo group nor Unity expect side effects from the drugs themselves. Unity’s researchers argue that their local administration of senolytics makes this unlikely, and Kirkland stresses that dasatinib and quercetin already have good safety records in people. But Kirkland is worried that there could be adverse consequences from killing senescent cells. If cells, even those in a state of senescence, die in large numbers at once, they could spill intracellular contents into surrounding tissue, potentially causing organ dysfunction—a phenomenon observed in some cancer patients during harsh treatments. This concern is compounded by the idea that the immune systems of older people may not be efficient enough to clear the cellular waste generated by killing senescent cells. Both the Unity and Mayo teams consider the chances of these issues occurring to be low, given the relatively low numbers of senescent cells in individual tissues. Unity researchers estimate that less than 7 percent of cells in tissue biopsies of osteoarthritis patients’ knees are senescent.
Campisi, meanwhile, wonders about the consequences of losing the benefits that senescent cells provide. Of the hundreds of proteins secreted by these cells that her group has identified over the years, some are critical for wound healing, for instance—eliminate them all, and the body’s ability to handle injuries may suffer, she notes. “There will be times where you’ll have to be cautious. You probably don’t want to take a senolytic before you go into major surgery.”
Hopeful that they can overcome such challenges, the Mayo Clinic and Unity researchers are forging ahead. Kirkland says his team has several more trials for serious conditions, including osteoporosis and Alzheimer’s disease, in the pipeline, and Unity is planning trials for age-related eye diseases—such as macular degeneration—with its other senolytic compounds, UBX1967 and UBX1325. Further in the future, the researchers hope to explore the possibility of senolytic drugs as prophylactics to help people who are healthy, but who have high levels of senescent cells, to delay the onset of physical decline and disease.
I think that our possibilities of translation are pretty high.—Felipe Sierra, National Institute on Aging
Campisi, Kirkland, and Unity researchers are quick to caution against going down that road too quickly. Some in the anti-aging industry have already jumped at the prospect of marketing supplements like quercetin as newfound elixirs of youth. But “people should not be taking these drugs until there’s really clear evidence from Phase 2 trials that they’re safe and effective,” says Kirkland—“and that may or may not be true.”
While early clinical trials of senolytics as anti-aging drugs are generating guarded excitement, aging researchers are still trying to figure out the role of cellular senescence in the fundamental aging process. Although it’s not in question that senescence is a driver of aging, it’s certainly not the only one, or even a major one, Campisi says. If it were, researchers would be able to make mice live much longer by eliminating senescent cells, but that isn’t the case. Experiments by Kirkland and van Deursen have increased rodents’ median lifespan—meaning that fewer of the animals are dying young—but eliminating senescent cells couldn’t extend the maximum lifespan of the animals.
“We have no evidence that this strategy—or any strategy—will extend absolute years of life,” Campisi says, “but it’s extending years of healthy life that might be on the horizon—if these drugs work out in the clinic.”
Katarina Zimmer is a New York–based freelance journalist. Find her on Twitter @katarinazimmer.