PITY the poor prion protein. If ever biology had a case of “give a dog a bad name”, this is it. Prions are known almost exclusively for their role in BSE, CJD and other prion diseases. Their principal claim to notoriety is the ability to morph from an inoffensive shape into a rogue one that causes other prion molecules to follow suit, setting off a deadly chain reaction.
But perhaps it won’t be long before this most maligned of molecules enjoys a change of image. A number of scientists are starting to believe that the ability to shape-shift might not be so sinister after all. For example, a Nobel prizewinning neuroscientist has stumbled upon a prion-like protein that, instead of causing disease, might change shape in order to play a vital function in memory (see “Memory makes its mark”). And a biochemist thinks he has uncovered a whole family of proteins that shift from one form to another. These proteins appear to share a number of characteristics with the prion protein. Is it possible, then, that the ability to acquire and transmit a wholesale change in shape is not unique to prions and not always a bad thing – perhaps even a part of normal biology?
Biochemist James Morré from Purdue University in West Lafayette, Indiana, certainly thinks so. He has become convinced that our cells contain thousands of proteins that flip-flop from one form to another and back again, playing a key role in the body’s timekeeping. The reason these proteins can keep time, he says, is because they can “learn and teach” – they can sense their neighbour’s shape and adopt it themselves, so in the end all family members in the cell are flip-flopping in unison. To Morré, this modus operandi is not too dissimilar from that of the prion protein, and he believes his enzymes, as well as prions themselves, might be members of a larger family of shape-changing proteins.
Advertisement
It’s a controversial view, and all the experts we asked to comment on his work say his arguments are still far from watertight. But most acknowledged that his findings are intriguing. Molecular neurobiologist Jerry Yin of the University of Wisconsin-Madison calls Morré’s results “interesting and provocative”, and says he hopes that others will try to repeat them. “The data are quite preliminary,” adds Rudolph Glockshuber, an expert in protein folding at the Swiss National Institute of Technology in Zurich, but he too describes Morré’s interpretation as “rather fascinating”.
Morré’s family of proteins, found in plants and animals, is called ECTO-NOX. His team first described them in the 1980s as cell-surface enzymes with the ability to transfer electrons from a molecule called NADH to oxygen. It wasn’t until the late 1990s, though, that ECTO-NOX started delivering surprises. First, the team found that the proteins had a second function. As well as the ability to oxidise NADH (the origin of the “NOX” in the name), it was able to carry out a reaction called “disulphide-thiol interchange” in which electrons are exchanged between sulphur atoms in proteins. This dual role was unexpected, as enzymes are supposed to be specialists.
There were more bizarre findings to come. When Morré’s team took a closer look at the NADH-oxidising activity of ECTO-NOX proteins, they noticed what appeared to be a cyclic pattern. Whichever member of the ECTO-NOX family they looked at, they saw peaks of activity repeating very precisely. Some members did so every 22 minutes, while others cycled every 24 minutes. The oscillations were also there for the disulphide thiol activity, and seemed to happen in a mirror-image fashion: when NADH oxidase went up, disulphide thiol interchange went down, and vice versa. Somehow, it seemed, the proteins had the ability to oxidise NADH and perform disulphide-thiol interchange in an alternating pattern with exact intervals (Biochemistry, vol 41, p 3732).
For any biochemist, the suggestion that something like this could happen is almost heretical. No known molecule has ever been described as oscillating between two distinct functions, points out Francisco Laurindo of the University of São Paulo in Brazil. An NADH oxidase expert himself, Laurindo was sceptical about this claim when he heard Morré present it several years ago. But after taking a closer look at the data and following Morré’s work for years he now believes the oscillations are real.
What is still contentious, says Laurindo, is what the oscillations mean. Morré believes they are the basis for how a cell keeps time, and two experiments appear to bolster that argument. In one study his team made eight mutant versions of t-NOX and tested their ability to cycle between the two functions. In four of the mutants the activity had shifted: in three, the peaks now came every 36 minutes and in one every 42 minutes. When added to cells, the mutants also shifted the normal 24-hour clock to 36 and 40 to 42 hours (Biochimica et Biophysica Acta, vol 1594, p 74).
In addition, there is some evidence that ECTO-NOX proteins might exhibit a property called “entrainment”, which any system that acts as a biological clock must have. For the clock idea to make sense, there has to be way for individual oscillators to keep time on a larger scale. In other words, a population of thousands of ECTO-NOX proteins can only be involved in timekeeping if they can all oscillate together – if they can “entrain” one another to cycle in unison.
Morré has published a number of articles where he claims that this is indeed what ECTO-NOX proteins do. In one of the studies, his team took several different samples of mammary tissue membrane harvested from cow’s milk – which they knew to be a good source of an ECTO-NOX protein with activity that cycled every 24 minutes. At the start of the experiment, the samples were not cycling in sync, but when they mixed two preparations and left them for several hours, the activity of the mixture became synchronised. Two identical preparations stored separately for the same amount of time and mixed just before testing remained out of sync (Biochimica at Biophysica Acta, vol 1559, p 10). This shows, Morré says, that when two out-of-phase samples of ECTO-NOX are mixed they can entrain each other to cycle together.
But how can the proteins entrain one another? Morré says spectroscopic evidence from his lab suggests it’s via a shape change. He believes something inherent in the shape of ECTO-NOX proteins makes it easy for them to shift from one job to the other, like miniature pendulums. If this is right, then the properties of ECTO-NOX proteins would be unprecedented. Not surprisingly, Morré’s entrainment experiments are the ones most heavily challenged by outside experts. “It is so unusual that of course your first reaction is to ask what kind of artefact could have happened,” Laurindo says. There is nothing to suggest the experiments were not done properly, he adds. “It’s just a little bit too much to swallow.” Morré is aware of the scepticism. “This is way out of the box,” he admits.
Intriguing links
To top things off, Morré’s laboratory has found a few intriguing links between shape-changing ECTO-NOX proteins and prions. It all started in 1998, when his team noticed another unexpected behaviour in the enzymes. They were studying t-NOX, a member of the ECTO-NOX family present in cancer cells, when they found that at high concentrations, its NADH-oxidase activity mysteriously disappeared. As the protein’s activity declined, they noted, a grainy substance appeared in the test tube.
The grains initially looked like amorphous gunk, but when the team examined them under an electron microscope a different picture emerged. It turned out the proteins were forming rods about 10 nanometres in diameter and 100 nanometres long, very similar in shape to the characteristic “amyloid rods” that form in the brains of people with Alzheimer’s and prion diseases. And like amyloid rods, t-NOX rods were unaffected by extreme pH, detergents and heat (Archives of Biochemistry and Biophysics, vol 358, p 125).
Now Morré has found other hints of a connection between ECTO-NOX proteins and the prion protein found in humans and other mammals. He recently showed, for example, that ECTO-NOX proteins contain a region that specifically binds copper. Very similar copper-binding domains are found in theprion protein and other proteins that form amyloid rods.
What is more, Morré’s team recently showed that, just like ECTO-NOX, the prion protein from mice has NADH oxidase and disulphide-thiol interchange activities that oscillate every 24 minutes (Biochemical and Biophysical Research Communications, vol 315, p 1140). Interestingly, one of the few known defects in mice that lack the prion protein altogether is a problem with their circadian clock. There is also a hereditary human disease called fatal familial insomnia, caused by a mutation in the prion protein. At the onset of the disease, victims often report disturbed sleep patterns and other circadian disruptions. “I do think this is how it’s ultimately going to be,” Yin says. “There are going to be proteins in the cell that do timekeeping, and the prion protein could be one of them.”
Morré now thinks that ECTO-NOX and the prion protein may be just two examples of a larger class of proteins with the ability to flip-flop between shapes, which could give them a role in cellular timekeeping. The capacity to morph so easily might be what makes these proteins prone to locking into a more extreme, irreversible shape, he thinks. In the case of t-NOX, this is the form Morré saw as rods in his experiment. In the case of the prion, it’s the infamous form we associate with disease.
Morré still has work to do to stand his hypothesis up. On the other hand, it’s hard to ignore the signs that he might be onto something. And together with the work by Nobel prizewinning neuroscientist Eric Kandel and colleagues, the conventional wisdom on what shape-changing proteins do inside our cells might soon change. It’s not time to unveil a new image for the prion protein just yet. But the makeover may have begun.
MEMORY makes its mark
Think of your favourite childhood song. Do you hear it in your head? It’s a remarkable feat that we can somehow remember certain things for long periods of time.
Scientists don’t fully understand how this happens, but they do have some hints as to how it all starts. Lasting memories are engraved in the brain through a process that physically strengthens the connections between neurons. This process, known as marking, requires the manufacture of new protein near the synapse. The nature of the proteins involved in synapse marking, and exactly how it is that the marking takes place, are still shrouded in mystery. But Nobel laureate Eric Kandel from Columbia University in New York has made the bold proposal that prion-like proteins might be involved.
Prions are the only proteins thought to change shape and transmit that change to others. Kandel is now proposing that synapse marking depends on a similar shape-shifting protein (Cell, vol 115, p 879).
If this turns out to be true the implications are immense. It suggests that prion-like behaviour in proteins could be a widespread phenomenon in our cells. “We might be looking at the tip of the iceberg. This could be how normal biology works,” says prion expert Witold Surewicz from Case Western Reserve University in Cleveland, Ohio.
Kandel shared a Nobel prize in 2000 for work on the sea slug Aplysia, a model organism for studying memory. In search of the proteins that might be used for synapse marking, Kandel’s team looked for an Aplysia equivalent of a mammalian gene called CPEB, whose protein has been postulated as a memory marker molecule in higher animals. The team did find the sea slug’s version of CPEB, but it came with a surprise: unlike the gene from other organisms, this one seemed to have a sequence very similar to one present in the prion gene PrP.
Working with Susan Lindquist from the Whitehead Institute for Biomedical Research in Cambridge, Massachusetts, Kandel’s team confirmed that Aplysia’s CPEB protein behaves very much like a prion when placed inside yeast cells. It appears to exist in two forms and, in a prion-like fashion, when CPEB turns into the “infectious” form it induces other CPEB molecules to do the same. Once in this form, Aplysia’s CPEB becomes active, awakening dormant messenger RNAs.
Kandel’s team has yet to prove that this is how the protein behaves inside the sea slug’s neurons. But if it does, then CPEB would perfectly fit the profile of a candidate marker molecule. Kandel’s hypothesis is that CPEB shifts from an inactive shape to an active form, and that this is a stable, irreversible change that is highly localised and constitutes the synapse mark – in sea slugs, at least. “There has never been anything like this in the memory field,” says Jerry Yin of the University of Wisconsin-Madison.
Whether the results have any relevance for other organisms is another matter. “A pretty open question in my mind is whether these mechanisms are going to apply in mammals,” says neuroscientist J. David Sweatt from Baylor College of Medicine in Houston, Texas. He points out that Aplysia lives at 15° C, where controlled prion-like conversions might be chemically easier than in warmer conditions.
Jonathan Weissman, a prion scientist from the University of California, San Francisco, suggests that the next step for Kandel should be to show that there is a prion-like CPEB variant in other animals. “If they can demonstrate this in Drosophila or mammals it would be a profound result,” Weissman says. “I would certainly hope that it turns out to be true.”
