Our memories feel stable and secure. They’re such a critical part of our identities that losing them can feel like losing ourselves. But how do our brains achieve such permanence, when the molecules within them are constantly being degraded and recreated? How exactly do we store memories?
It’s likely that many molecules are involved, but over the last decade, one has emerged as a possible star player—an enzyme called PKM-zeta.
In 2006, Todd Sacktor from SUNY Downstate Medical Center managed to erase established memories in mice by injecting their hippocampus—a region involved in memory—with ZIP, a chemical meant to block PKM-zeta. A year later, Sacktor collaborated with Yadin Dudai at the Weizmann Institute in Israel to show that injecting ZIP into a different part of a mouse’s brain could erase month-old memories of an unpleasant taste. Many similar experiments followed suit, involving other animals, brain regions and labs. And in 2011, Sacktor managed to boost an old faded memory by increasing levels of PKM-zeta in the brain, by means of a virus loaded with copies of the gene.
As I wrote in a news piece for Nature, “these fascinating studies suggested that long-term memory, rather than being static and stable, is surprisingly fragile, and depends on the continuous activity of a single enzyme.”
But two teams of scientists have cast some doubt upon this neat tale, and upon PKM-zeta’s role as a memory molecule. Working independently, they deleted the gene for PKM-zeta in embryonic mice, producing adults that lacked the enzyme from birth.
And the rodents’ memories were fine. One group, led by Robert Messing at the University of California, San Francisco, showed that their mice formed persistent memories of fears, objects, movements and more. The other, led by Richard Huganir from Johns Hopkins University, showed that their mice had normal levels of long-term potentiation— a process that underlies learning and memory, where the connections between neurons become stronger. Both groups also found that injections of ZIP could still disrupt memories in their mice, suggesting that whatever this chemical is doing, it’s not acting on PKM-zeta (or at least, not doing so exclusively).
In my write-up for Nature, I go into more details about the studies, and discuss whether the results could be explained by other back-up systems that compensate for the loss of PKM-zeta. Do head over there for the full story.
Some navel-gazing about science writing and complexity
On a more personal note, the PKM-zeta story serves as a good reminder to resist easy explanations or tidy fables in science writing.
I have covered this molecule on this blog for years, including posts about two of the big splashy Science papers on memory erasure and memory strengthening. I gave the molecule a catchy epithet (“memory engine”). I wrote a long piece about its history and what it does.
Looking back at the coverage, I’m happy with the way the concepts are explained, but the pieces are rather breathless (“These were amazing results”, and “the implications of this are staggering,” quoth me.) And, most disappointingly, they’re largely one-sided. They present a hypothesis—which may or may not eventually turn out to be right—as hard fact. This piece, in particular, is 1,800 words long without a single outside voice.
Those voices were out there. In response to tweets about the new story, I’ve seen many comments that are variants of: “Did anyone seriously think that a single molecule would explain the maintenance of long-term memory?” Clearly, there was scepticism about the idea; I just didn’t look hard enough.
And of course, I should know better. I worked in a cancer charity for many years and I know full well that any attempt to explain a complex thing, whether cancer or obesity, through a single simple route is almost certainly wrong. And I’ve spent the last year lambasting examples of science writing that favour false simplicity over real complexity (oxytocin, anyone?). If the PKM-zeta story ends up being more complicated, with multiple redundant back-ups and many molecular players, that’s not going to shock many neuroscientists. (As one said to me: “[The studies] show that the situation is complicated—surprise!”)
The problem is that simple explanations are seductive, and they make for nice stories. But the ultimate story of science is that things are regularly complicated, and often bafflingly so. The best science writers embrace that complexity, rather than sweeping it under the rug for the sake of a clean narrative.
To clarify, regarding PKM-zeta, I’m not taking “sides” (if sides even exist to be taken). This story will roll on. I know that both Huganir and Sacktor have more experiments planned, and other neuroscientists have contacted me with their take on the contrasting studies. My job, and my desire, is to chronicle this meandering work in a fair and appropriately critical way, and I look forward to doing that in the future.
For more on my views about communicating complexity in science writing, see the video below. I’m the second to speak.