Autophagy: Drugging the yin and yang of the cell

At the beginning of 2017, Keith Dionne was looking for the next big idea in biotech. He had just joined the life sciences–based venture capital firm Third Rock Ventures, and the investors there were building a company centered on the cellular process of autophagy, the subject of the 2016 Nobel Prize in Physiology or Medicine, announced just months before.

Autophagy literally means self-eating, but the cellular housecleaning it accomplishes is more important than its cannibalistic connotations. Cells build structures called autophagosomes around junk they want to recycle—for example, old or damaged organelles, fat deposits, clumps of poisonous proteins, or even microbial invaders. Once that garbage is bagged, cells break it down into reusable building blocks and nutrients.

A year earlier, Beth Levine, an autophagy researcher at the University of Texas Southwestern Medical Center, and a few other experts in the field had come to Third Rock to pitch a biotech start-up that would develop drugs to boost that cellular housecleaning. The Third Rock investors heard that a wide variety of seemingly unrelated diseases all share an underlying problem: a buildup of cellular garbage. Aggregations of amyloid proteins, for example, are associated with Alzheimer’s disease. An accumulation of lipids in the liver leads to nonalcoholic fatty liver disease. And toxic molecules pile up in a swath of rare genetic enzyme deficiencies. The list kept going, Dionne says. “The thought was if there is a common pathology, then potentially there’s a common solution.”

Dionne learned that over the past 2 decades, researchers had shown that boosting autophagy did seemingly everything, from clearing out the toxic proteins that accumulate in neurodegenerative diseases to increasing the life spans of worms and mice. Levine laid out evidence that autophagy could be important for fighting infections and maybe even preventing cancer. It wasn’t just about taking out the trash; it was a process of rejuvenation.

Enhancing autophagy, it would seem, could be a panacea. Yet Dionne had known almost nothing about it. “How could I be ignorant of this and yet it have that much potential impact?” Dionne recalls thinking. And why, he wondered, weren’t other researchers trying to develop drugs for this pathway?

Actually, a number of scientists had tried. But their therapeutic thesis had run counter to Levine’s: they were all bent on trying to inhibit autophagy in cancers that appeared to hijack the process to their own advantage. For years, that approach overshadowed the idea of enhancing autophagy. Moreover, conflicting and lackluster results in those autophagy-blocking studies eventually soured drugmakers’ interest, and industry retreated from the field.

Third Rock, however, was convinced that autophagy enhancement was worth exploring. In May 2018, the firm backed Levine’s broad vision for doing so by launching Casma Therapeutics and funding it with $58.5 million. Dionne, now the start-up’s CEO, says that Casma already has six drug-discovery programs to treat diseases afflicting the brain, gut, liver and muscles.

And although Casma’s therapeutic goals are particularly large, it’s not the only company designing autophagy-boosting therapies. Autophagy-focused biotech firms are working on treatments for rare genetic diseases, Alzheimer’s, and many other indications. They even think those same drugs could extend our life spans. Yet as investment floods into the field, researchers caution that cellular self-eating has a dark side: autophagy-boosting drugs might have the unintended effect of helping silently lurking tumors sprout and spread. That link to cancer has kept some researchers focused on the older idea of developing drugs that block the process.

Despite the lingering controversy of whether it’s better to enhance or inhibit autophagy, the momentum has those who have worked in the field for years excited that autophagy’s time has finally come. “The literature has just grown and grown, and now it’s explosive,” says David Perlmutter, an autophagy researcher and dean at Washington University School of Medicine in St. Louis. “And every single new observation validates how important the system is in our biology.”

Cancer controversy

Long before autophagy was linked to human disease, biologists recognized the process as a cellular survival mechanism. When scientists deprived cells of nutrients in the lab, large vesicles would emerge to engulf, digest, and recycle a cell’s innards.

Conceptually, this wasn’t surprising. The human body’s ability to survive in periods of extreme, even self-imposed starvation was well known. Inspired by this notion, in 1859 a French physiologist experimented with what he called artificial autophagy, in which he drew blood from malnourished animals and fed it back to them. His intention was to devise strategies for shipwrecked people to sustain themselves while awaiting a prolonged rescue. Upon hearing the gossip, the New York Times called it “cannibalism reduced to a civilized and humanitary institution!”

Autophagy, as scientists conceive it today, was first named by Belgian biologist Christian de Duve in 1963. The process that de Duve described in cells, and others later refined, is as follows: during starvation or under other stress, cells begin to form membranes around protein clumps, organelles, or anything else the cell considers to be debris or otherwise dispensable. This developing membrane, called the phagophore, extends around the debris, eventually engulfing it. That newly formed cellular trash bag is called an autophagosome. Finally, an acid-and-enzyme-filled vesicle called the lysosome fuses with the autophagosome to dismantle the debris into molecular building blocks and nutrients for the cell to use and build anew.

Until 1993, no one knew how this process was controlled. That year, Yoshinori Ohsumi described 15 genes in yeast that were essential for orchestrating autophagy. That work—the basis for his 2016 Nobel Prize—meant researchers could finally study the machinery that made autophagy tick, and perturb it to see what happens when the process is broken.

The first connection between autophagy and human disease came in 1999 when UT Southwestern’s Levine serendipitously discovered that a protein called beclin 1 promoted autophagy and blocked cancer growth in human breast cancer cell lines (Nature 1999, DOI: 10.1038/45257). It turned out that some 40–75% of women whose breast and ovarian cancers weren’t driven by an inherited mutation were missing a single copy of the beclin 1 gene. In a subsequent experiment, she showed that mice missing both copies of their beclin 1 genes died as embryos, while mice missing a single copy of the gene lived, only to develop cancer within 12 to 18 months.

To Levine, all of this amounted to a smoking gun: properly functioning beclin 1, and perhaps autophagy at large, was a critical tumor-suppression system in the body.

Several other research teams, however, would soon call that model into question. One of those teams was headed by Eileen White at the Rutgers Cancer Institute of New Jersey, whose lab discovered that cancer cells, particularly ones that refused to die, had suspiciously high autophagy levels. “Our hypothesis was that tumor cells were usurping a natural survival pathway to promote their own survival,” says White, who later started an autophagy-inhibitor company called Vescor Therapeutics.

White’s lab showed that inhibiting autophagy—either with drugs or by deleting autophagy genes like beclin 1—weakens cancer cells and halts their growth, making them easier to kill. Meanwhile, University of Pennsylvania oncologist Ravi Amaravadi discovered that the antimalarial drugs chloroquine and hydroxychloroquine—already known to inhibit autophagy by neutralizing lysosomes—made tough cancer cells more sensitive to chemotherapy.

Those studies, which were at direct odds with Levine’s hypothesis, spurred several large pharmaceutical companies to begin developing autophagy inhibitors for cancer. Amaravadi even began testing antimalarial drugs with anticancer therapies in Phase I and II clinical trials. So far, the results look encouraging, but they’re not a home run, he says.

The big pharma programs didn’t last long enough for new autophagy inhibitors to be tested in humans. Scientists at Novartis and Pfizer independently conducted autophagy-inhibition experiments in dozens of cancer cell lines, and they teamed up to publish a study in 2015 concluding that the strategy didn’t work (Proc. Natl. Acad. Sci. U.S.A. 2015, DOI: 10.1073/pnas.1515617113). That paper, which was edited by Levine, has proved contentious. “It is a very damaging paper for the field,” says Amaravadi, who is still testing hydroxychloroquine in people with cancer.

To complicate matters further, studies in mice have shown that permanently inhibiting autophagy in healthy cells and cancer cells kills tumors but causes death from rapid neurodegeneration. Conversely, permanent autophagy enhancement protects mice from neurodegeneration and even increases their life spans, but there are scattered reports of increased tumor development. “There is always a yin and yang here, a positive and a negative effect of what autophagy is doing,” says Andrew Thorburn, an autophagy and cancer researcher at the University of Colorado Denver.

Thorburn, who is submitting a paper to refute the 2015 Novartis-Pfizer study, still thinks that briefly inhibiting autophagy to weaken tumors is a good strategy. “Autophagy is doing both good things and bad things,” he says. “And what we’re doing right now—saying we’ve just got to boost it or inhibit it—is way too crude.”

Getting selective

After the Nobel Prize, investors were more receptive to taking meetings with scientists with ideas for autophagy companies. “There is definitely a growing industry excitement around autophagy enhancement,” says Sebastian Aguiar from Apollo Ventures, a venture capital firm focused on the biology of aging.

Aguiar is also a cofounder and the chief operating officer of an Apollo-backed biotech start-up called Samsara Therapeutics, which is developing drugs that enhance autophagy to, among other things, extend life span. In February, the start-up’s four scientific cofounders identified a molecule, called 4,4′-dimethoxychalcone, that appears to do just that. Found in a Japanese herb, the compound is consumed on an island known for its supercentenarians—people who live to 110 years and beyond (Nat. Commun. 2019, DOI: 10.1038/s41467-019-08555-w).

The Nobel Prize also seems to have spurred investors to go looking for autophagy experts themselves. In January, an antiaging conglomerate called Life Biosciences raised $50 million to fund multiple subsidiaries that are each focused on tackling a different aspect of aging. And for help with autophagy, the company reached out to Ana Maria Cuervo, an autophagy researcher at the Albert Einstein College of Medicine.

Cuervo has spent most of her career studying a specialized process called chaperone-mediated autophagy, in which individual proteins are flagged and transported one by one into a specialized set of lysosomes for destruction. Animal studies suggest that the process is important for cleaning up toxic proteins that accumulate in neurodegenerative diseases.

In collaboration with her Einstein colleague Evripidis Gavathiotis, Cuervo has discovered small molecules that directly boost chaperone-mediated autophagy without touching any of the other autophagy pathways. A protein called retinoic acid receptor α inhibits chaperone-mediated autophagy, and small molecules inhibiting that protein boost the process (Nat. Chem. Biol. 2013, DOI: 10.1038/nchembio.1230).

The work didn’t make much of a splash at first. Suddenly, after the Nobel Prize, “the concept became cool,” Cuervo says. “We had at least 10 groups of investors talking to us.”

Cuervo and Gavathiotis ultimately teamed up with Life Biosciences to found Selphagy Therapeutics, which is currently developing the duo’s compounds for degenerative diseases characterized by misfolded protein accumulation. The long-term goal is to treat brain diseases like Alzheimer’s and Parkinson’s, but Selphagy will start by testing its compounds in eye diseases because clinical trials will be cheaper and faster to run.

Targeting chaperone-mediated autophagy and other forms of selective autophagy could also provide a means for avoiding the potentially harmful effects of broadly boosting autophagy. Of interest is mitophagy, which uses autophagosomes to remove mitochondria that have been molecularly marked for destruction, and has been implicated in Parkinson’s. Mitochondria, the power generators of a cell, break down and wear out frequently and can leak toxic molecules into the cytosol if not replaced. That makes mitophagy particularly important in brain and heart muscle cells, which stay with you for life.

In fact, mutations in genes that regulate mitophagy, PRKN and PINK1, are some of the strongest predictors of early-onset Parkinson’s disease. Translating that knowledge to drug-discovery programs, or simply finding drugs that specifically enhance mitophagy, has been difficult. “There are lots of things that boost mitophagy,” but they do so by damaging the mitochondria, says Roberta Gottlieb, who studies the process at Cedars-Sinai.

A few companies are trying, however. In November 2018, a start-up called Rheostat Therapeutics launched with $23 million to boost mitophagy and autophagy for neurodegenerative diseases. Other firms, including Mitobridge and Proteostasis Therapeutics, are designing inhibitors of deubiquitinating enzymes, which remove the molecular markers that destine old mitochondria for destruction. The inhibitors, in theory, could keep pushing mitochondria through mitophagy.

Other diseases also feature clear defects in an autophagy pathway. For instance, several rare genetic diseases are characterized by defective lysosomes. And in 2017, neuroscientist David Rubinsztein from the Cambridge Institute for Medical Research discovered that autophagosome formation is impaired in Huntington’s disease and genetically similar conditions.

The idea that impaired autophagy is a common thread across so many serious diseases is tantalizing and suggests that a single drug program, in theory, could lead to many cures. “In some cases you are repairing a defect, but I don’t think you need a defect for many of these diseases to justify the strategy,” Rubinsztein says.

Tantalizing targets

As researchers consider how best to control this process, some are looking first to older drugs that have inadvertently been activating or inhibiting autophagy all along. For instance, Amaravadi and University of Pennsylvania chemist Jeffrey Winkler have developed derivatives of hydroxychloroquine that are 1,000 times as potent as the version Amaravadi is testing in the clinic. And they’ve founded a company, Pinpoint Therapeutics, to try to commercialize the work.

Repurposing existing drugs holds potential for autophagy enhancement too. Rubinsztein’s group discovered that an antihypertensive called rilmenidine boosts autophagosome formation in mice. More recently, Rubinsztein’s lab found that another antihypertensive drug, called felodipine, enhanced autophagy in three animal models of neurodegenerative disease (Nat. Commun. 2019, DOI: 10.1038/s41467-019-09494-2). And Cuervo’s lab found that the rare-disease drug lonafarnib helps clear out toxic tau proteins in mice by enhancing lysosome activity (Sci. Transl. Med. 2019, DOI: 10.1126/scitranslmed.aat3005).

Perhaps the most well-studied autophagy activator, however, is rapamycin—a drug with a storied history and a long list of side effects, including immune suppression. Rapamycin inhibits a protein called mTOR, which is part of a molecular rheostat that controls metabolism. “We know that if you inhibit mTOR, you will induce autophagy,” Casma’s Dionne says. “The challenge is you will induce 20 other pathways as well.”

A sizable list of other drugs has also been shown to activate autophagy, including the diabetes drug metformin, the nutritional supplement resveratrol, and aspirin. The problem with many of these molecules, Casma’s chief scientific officer, Leon Murphy, says, is that we have no idea if they are activating autophagy directly, through an indirect pathway, or simply by stressing cells. “We really haven’t been rigorous about evaluating these molecules.”

Casma, like many of the recently formed autophagy start-ups, is focused on drugs that more specifically target particular proteins involved in the process. One promising strategy is activating a protein called TFEB, the master lock for autophagy and lysosomal genes. Studies using gene therapy to permanently boost TFEB in mice show that it can reduce the poisonous proteins in liver and neurodegenerative diseases and even reduce obesity.

At least two groups think a TFEB gene therapy could make for a powerful treatment in humans. Harvard University geneticist George Church and a former postdoc from his lab, Noah Davidsohn, have founded a company called Rejuvenate Bio that is testing antiaging gene therapies in dogs. An expansive patent held by the duo includes TFEB and other autophagy-related genes in a table of what might be included in the gene therapy. Church is also working with scientists at Lund University in Sweden to develop a version of gene editing called CRISPR activation that would keep the TFEB gene turned on. By boosting autophagy this way, they hope to reduce the amyloid-β and tau proteins linked to Alzheimer’s disease.

Dionne is wary of the gene-therapy approach, partly because some studies show that constant TFEB activation is linked to the development of renal cancer. In contrast, using a small molecule to activate TFEB could allow people to periodically boost autophagy. That may be a safer approach, and there appears to be more than one way to do it.

TFEB is a transcription factor that is normally inactive and resting in the cell’s cytosol. During starvation and other autophagy-triggering events, TFEB travels to the nucleus, where it turns on autophagy and lysosomal genes. Several proteins control this process: enzymes called kinases phosphorylate TFEB to keep it in the cytosol, activation of a calcium channel called TRPML1 helps trigger TFEB’s journey to the nucleus, and a protein called exportin 1 expels TFEB from the nucleus. These proteins are all druggable targets for companies looking to boost autophagy.

Pipeline in a pill

Research and investment might be proliferating, but at the moment, there are few molecules available to precisely probe autophagy. No drugs specifically designed to boost or inhibit autophagy have entered the clinic, and the first trials from Casma and Selphagy are likely a year or more away.

And some think there are still too many unknowns for drug development to move forward. “Some of these things need to simmer for a while,” says Vojo Deretic, who studies autophagy’s role in infections at the University of New Mexico. “Somebody may get lucky, but I think it is going to be very difficult to do this.”

The excitement for basic and applied autophagy science is yielding new resources to make it happen. Last year, the National Institutes of Health awarded Deretic $11 million to launch the new Autophagy, Inflammation, and Metabolism Center of Biomedical Research Excellence. This year, the NIH funded a $37 million effort, by UT Southwestern’s Levine and five other institutions, to develop approaches that use autophagy to fight infections. And in February, Washington University in St. Louis used a $10 million donation to launch the Philip and Sima Needleman Center for Autophagy Therapeutics and Research.

“That money is just the beginning,” says Perlmutter, leader of Washington University’s new autophagy center. “We are going to be raising more for the center and investing from an institutional perspective. This is a big priority for us.”

And though autophagy-focused biotech companies have ambitions to take on big diseases such as Alzheimer’s and nonalcoholic steatohepatitis, most plan on testing their drugs in rare diseases first. For instance, Casma is working on therapies for a rare brain disease called Gaucher disease type 3. “But there’s no reason to believe that an inducer of TFEB wouldn’t work just as well on a rare genetic disease as it would in a more common disease like Parkinson’s,” Dionne says.

Likewise, although Samsara will likely start by testing its autophagy boosters in a rare disease, Aguiar says the company plans on applying the same drugs to multiple age-related conditions. “Drugs that slow the aging process are the ultimate pipeline-in-a-pill archetype,” Aguiar says.

Some scientists are already dreaming up ideas for once-a-day autophagy-enhancing pills to clean up the accumulation of toxic proteins in the brain. Rubinsztein thinks that an approach like this could help prevent neurodegenerative diseases in the same way that statins are used to help prevent heart disease.

For an idea like that to ever work, however, companies will need to show that long-term autophagy enhancement is safe in humans. For now, that remains an open question.

“It has enormously palpable potential,” Perlmutter says. Yet when asked what autophagy’s biggest disease-curing potential is, Perlmutter’s answer is one that is echoed across the field. “Autophagy is important in a lot of things,” he says. “None of them stand out because all of them look interesting.”