Dynamic Differences

Tumor suppressor protein displays subtle variations in behavior across species despite sequence conservation

In this time-lapse movie taken over 24 hours, damaging radiation triggers simultaneous and repeated pulses of p53 (green) in individual cells. Animation: Alexander Loewer

Since it first evolved in the common ancestor to multicellular animals some one billion years ago, the p53 family of genes has played the role of genomic guardian for countless organisms.

So essential is its task—to protect the DNA of cells from damage and stress—that some form of p53 can be found in almost all animals, from sea anemones and insects to fish and mammals.

Get more HMS news here.

“The more you understand the mechanisms of a system like p53, which plays such a central role in controlling the cellular response to DNA damage, the better chance you have to identify new targets and design better therapies” - Galit Lahav

In humans, p53 famously serves as a master tumor suppressor, and for decades scientists have worked to better understand its biological functions with the goal of identifying new cancer treatment strategies.

Among the myriad tools used in this endeavor are animal models such as mice, which, despite roughly 75 million years of evolutionary divergence, express p53 protein that is around 90 percent similar to human p53.

This high level of evolutionary conservation has been thought to imply a correspondingly high degree of similarity in how p53 behaves across species.

But a new study from Harvard Medical School, published in Cell Systems, reveals subtle differences that suggest caution when investigating human disease and biology in animal models, the researchers say.

The team, led by Galit Lahav, professor of systems biology at HMS, and postdoctoral fellow Jacob Stewart-Ornstein, performed the first systematic cross-species analysis of a dynamic, time-dependent behavior of p53 known as oscillation, and found that it can vary depending on species.

“Our study shows that p53 behaves differently in human cells compared with mouse cells, and identifies the mechanisms involved,” Lahav said. “While it is still a mystery why these differences have been preserved by evolution, there may be more immediate biomedical implications.”

Variations in p53 behavior raise important questions about extrapolating observations made in animals—for example, whether conclusions related to toxicity and side effects could be affected when testing drugs on human tumors in mice—according to Lahav.

“Sequence data, which represent the genetic instructions, may not be sufficient if we are to understand how critical genes like p53 actually behave in different species,” she said.

As the central regulator of the DNA damage response, p53 is responsible for activating a cell’s DNA repair machinery. If the damage is irreparable, p53 initiates programmed cell death—effectively suicide—to eliminate potentially cancerous or otherwise harmful cells. Mutations to p53 are seen in nearly half of all cancers.

Levels of p53 protein within a cell rise dramatically when DNA damage is detected, but this increase also triggers the production of a protein known as MDM2, which degrades p53. The feedback between the proteins causes levels of p53 to oscillate over time.

Double Time

Previous studies by Lahav and colleagues have shed light on numerous aspects of this still- opaque behavior, showing that p53 oscillation timing can affect cell survival, drug response and cell fate. In the current study, the team expanded the investigation across species by analyzing cell lines from five different animals: mice and rats, humans and monkeys, and dogs, to represent a nonrodent and nonprimate group.

Human p53 (top) oscillates slower than mouse p53 (bottom) in response to DNA damage. Grey lines represent repeated trials. Image: Stewart-Ornstein et al, 2017.
To test whether this variation arose from the small differences in p53 amino acid composition between species, the team introduced human p53 into mouse cells and mouse p53 into human cells. Human p53 showed rapid, mouse-like oscillation timing when expressed in a mouse, and mouse p53 showed slow, human-like oscillations when expressed in human cells.

Further analyses revealed that mouse cells degrade p53 faster than human cells, and that human cells are slower to express MDM2, supporting the importance of cellular context as the determinant of p53 oscillatory behavior.

“We have great genomic data for many signaling pathways, and we’ve generally assumed that if there is strong conservation across species—the proteins look the same, the binding sites are the same, and so on—then it means that the behaviors will be similar,” said Stewart-Ornstein. “Our study shows that p53 is an example where those are all true, but we nevertheless observe different dynamic behaviors that are not predicted by genomic or proteomic data alone.”

Why different species exhibit variations in p53 oscillation timing remains unknown. Lahav, Stewart-Ornstein and their colleagues are now working to better understand the functional role of these differences and to further explore p53’s behavior in more species.

“I think the more you understand the mechanisms of a system like p53, which plays such a central role in controlling the cellular response to DNA damage, the better chance you have to identify new targets and design better therapies,” Lahav said.

There are many other biological pathways that display oscillatory or other temporal behaviors, Lahav added.

“We hope that our findings will prompt other researchers to look at the dynamics of their favorite proteins and investigate whether they are conserved across species.”

This research was supported by The Harvard Ludwig Cancer Research Center and NIH grants GM083303, GM116864 and CA207727. HMS graduate student Jacky Ho Wa Cheng is an additional author on the study.