Understanding the “zombie cells” that won’t die

Here is a line you do not hear every day in cancer research: How can we control that zombie?

In response to various forms of stress after birth, such as cancer, human cells can shift into a defensive crouch called senescence, which is marked by sharply reduced activity. In this altered state, the cells stop dividing, grow in size, become somewhat disorganized, and start pumping out an array of functionally diverse factors. Much like zombies, they also refuse to die easily.

As researchers are finding, these dynamic but poorly understood “zombie cells,” as they have been dubbed, are full of contradictions. In one form, they can hold some cancers at bay for years. A premalignant colorectal polyp, for example, can be composed of a clump of senescent cells that remain relatively stable over time. Certain oncogenes, however, can thwart the defense by reanimating the cells and forcing them to resume their uncontrolled replication.

As an alternative to another cancer defense known as programmed cell death, or apoptosis, senescence is far less predictable. “Apoptosis is more like live or die; it’s binary, right? But senescence is nothing like it,” says Masashi Narita, MD, PhD, a professor of senobiology at the University of Cambridge in the United Kingdom. “It’s a progressive and a heterogeneous phenotype, so it’s very difficult to say [whether] senescence is a good thing or a bad thing.”

Depending on the context, it may be both. In April 2023, at the annual meeting of the American Association for Cancer Research, Dr Narita and other experts spoke at a special session about senescent cells’ “double-edged sword.” Senescent cells induced by oncogenes such as MYC do not die easily, Dr Narita notes, in part because they are resistant to apoptosis. That transition, in other words, undermines another means of cellular defense. The early stages of the slowdown can also increase plasticity in a way that promotes cancer development through other mechanisms.

Some cancer interventions yield therapy-induced senescence, only to be undone by a mechanism called senescence-associated secretory phenotype (SASP). As part of SASP, cells release a stew of factors, including proinflammatory proteins that can promote tumorigenesis. Conversely, interventions such as chimeric antigen receptor T-cell therapy hinge on cancer’s hallmark of rapidly dividing cells; this means that senescence can limit the therapy’s effectiveness.

Despite the many complexities, or perhaps because of them, the field of cancer-associated senescence is booming. Ricardo Iván Martínez Zamudio, PhD, an assistant professor of pharmacology at Robert Wood Johnson Medical School and a research member of the Rutgers Cancer Institute of New Jersey in New Brunswick, recalls that the phenomenon was once labeled a cell culture artifact with no medical relevance. “But it’s just exploding! There are now more and more people interested in senescence,” he says.

A quarter-century ago, researchers found that the abnormally fast replication of cancerous cells brought about by certain oncogenes can trigger a senescence slowdown. During that seeming lull, however, the cells can display a dynamic range of behaviors: Some encourage their own removal by immune cells, whereas others seem to discourage that degradation and promote cancer and other age-related diseases. Dr Martínez Zamudio believes that the transitional state and its SASP mechanism may give the stressed cells a chance to assess their surroundings and, if possible, shift back to a seminormal or at least more stable existence. As a side effect of that potential quest for normalcy, however, some of the SASP factors can encourage mitosis and thereby drive the return of cancer cell proliferation.

In a recent study, Dr Martínez Zamudio showed how low-activity senescent cells could be revived in colorectal cancer and thus shed new light on one way in which malignancies can exploit the defense system.¹ First working in human fibroblast cells overexpressing the RAS oncogene, he and his colleagues found that cell senescence triggers a loss of regulation by a family of related transcription factors that would otherwise control which genes are turned on and off. “What we found is that the AP1 family of transcription factors in our model system seems to be very critical to mediate both the entry and the exit out of senescence,” he says.

AP1 transcription factors, in conjunction with another transcription factor called POU2F2, drive the senescent cells out of their slumber by reactivating cell cycle, reprogramming, and inflammation-promoting genes. In colorectal cancer cells, the researchers saw the same process.

More directly, if further research continues to implicate POU2F2 in the escape of colorectal cancer cells from senescence, it could prompt a search for drugs that bind to and disable the protein and thus prevent the return of cell proliferation. “With more validation in a more relevant system and in some tissue samples, I think POU2F2 could be a target,” he says, at least for a subset of oncogene-associated cancers. Dr Martínez Zamudio cautions, however, that the complexity of SASP could limit broader applicability of the research findings. “It’s such a complicated thing. The state of the senescent cell, it really depends on what the initial trigger was, and on the cell type,” he says. “I don’t think there’s going to be a Holy Grail drug to kill senescent cells.”

Despite the challenges, other groups are starting to untangle separate parts of the process. In their own line of research, Dr Narita and his colleagues are seeking to clarify how the RAS oncogene can trigger cell senescence. Initially, critics suggested that the phenomenon might be a laboratory artifact until a series of studies detected it in vivo. In their research, Dr Narita and his team are hoping to quantify the approximate physiological dose of RAS required to induce senescence.

On the basis of their results and those of other laboratories, Dr Narita speculates that cells carrying a slightly higher than normal level of oncogenic RAS might receive a small survival benefit. Over time, cells with higher levels of the oncogene could start to dominate. At some point, however, that level would exceed a threshold; the cellular stress would either trigger the start of senescence or cause the cells to become cancerous. “So, we are proposing that oncogenic senescence has to be seen as a spectrum, and can then be an integral part of tumor evolution,” he says.

As with cellular differentiation, he says, senescence may yield a vastly different gene expression profile over time. The plasticity of early senescence suggests that a cell in that transitional state might yet become cancerous. “If it’s very deep senescence, it may not be able to come back. We don’t know; no one knows,” says Dr Narita. One possibility, which he hopes to test, is that an epigenetic mechanism altering the DNA’s chromatin structure may eventually cause a senescent cell to reach a point of no return.

Depending on what scientists discover about the progressive stages of senescence, they may understand better where the pressure points are and whether they can be realistically targeted. “There’s a huge impetus on developing drugs that can either eliminate senescent cells or modulate the senescent states,” Dr Martínez Zamudio says. Another big priority, he says, is to generate a comprehensive catalog of what senescence means in different contexts. In his own laboratory, he hopes to analyze various cell types and senescence inducers to see if there is any commonality of factors driving the phenotype. On the basis of some of his own research, he believes that the AP1 family of transcription factors may yet oversee some important general mechanisms.

Dr Narita says that the field of research is entering a much more difficult phase, however: Unlike cellular differentiation during development, he sees the senescence process as being much more stochastic, or less predictable. Like Dr Martínez Zamudio, however, he is hopeful that some larger patterns might yet emerge. If so, the field may be on its way toward controlling a powerful, dangerous, and mercurial cellular zombie.